Compare commits

..

42 Commits

Author SHA1 Message Date
Joshua Lock
b29bd836ad runqemu: Fix path for i586 cross binaries
As spotted by Jessica Zhang

Signed-off-by: Joshua Lock <josh@linux.intel.com>
2010-04-28 14:12:03 +01:00
Richard Purdie
72eafe23fc Purple 3.2.1 Release 2010-03-19 14:37:29 +00:00
Graeme Gregory
5cc700daae zoom2.conf : switch to using the TI integration tree for kernel
Signed-off-by: Graeme Gregory <gg@slimlogic.co.uk>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-18 21:48:28 +00:00
Graeme Gregory
c25d4ba28c linux-omap-zoomsync_2.6.32.bb : add kernel from TI zoom2 integration tree
This is the latest code from dev.omapzoom.org and is also what OE is using
for zoom2 machine.

Signed-off-by: Graeme Gregory <gg@slimlogic.co.uk>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-18 21:48:21 +00:00
Joshua Lock
045b982378 handbook: Correction, we're building QEMU x86 not QEMU Arm image
Signed-off-by: Joshua Lock <josh@linux.intel.com>
2010-03-18 16:13:12 +00:00
Joshua Lock
f99f7e354f handbook: update references of stable version number
Signed-off-by: Joshua Lock <josh@linux.intel.com>
2010-03-18 16:13:03 +00:00
Richard Purdie
82925e8d7f handbook: Various tweaks for recent fop/pdf generation
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-18 15:44:50 +00:00
Richard Purdie
55e159c873 poky-doc-tools: Update to work with fop >= 0.95
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-18 15:40:28 +00:00
Joshua Lock
eddcc607c9 scripts/runqemu: Fix some typos
Fix typo in help message (we default to ext3 not ext2 now)
and a typo whereby moblin-image-sdk wouldn't be autofound.

Signed-off-by: Joshua Lock <josh@linux.intel.com>
2010-03-17 12:22:52 +00:00
Enric Balletbo i Serra
b82067c8e7 poky-eabi: Set missing PREFERRED_VERSION of binutils-cross-canadian
Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Signed-off-by: Joshua Lock <josh@linux.intel.com>
2010-03-17 12:09:10 +00:00
Enric Balletbo i Serra
b861de3d39 gcc: Fix fail to compile since they can't find limits.h include.
With gcc package on my build machine gcc can't find limits.h include.

$ cat x.c

$ gcc -M x.c
In file included from x.c:1:
/usr/include/limits.h:125:26: error: no include path in which to search for limits.h

This patch adds missing includes to gcc package.

Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-05 15:02:29 +00:00
Richard Purdie
10e9286aba gcc: Make sure target_* variables are set, fixing gcc search path issues
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-05 14:57:06 +00:00
Richard Purdie
e0964ab86c fakeroot: Stop setting LD_LIBRARY_PATH as it breaks things (see patch description)
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-05 14:56:55 +00:00
Richard Purdie
8f7bba3812 gcc-cross-canadian: Fix include paths and hence c++ compiler issues
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-05 14:56:37 +00:00
Richard Purdie
3a92d518b4 bitbake/data_smart.py: Fix error where update-rc.d would not get added to the dependency tree
If there was a variable such as:

X_${Y}_append = "Z"

The "Z" would be lost if X_${Y} was unset. This was due to a bug in the renameVar
function used by expandKeys().

Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2010-03-05 14:37:19 +00:00
Joshua Lock
9fad50948c insane.bbclass: Skip arch check for Cross Canadian
Signed-off-by: Joshua Lock <josh@linux.intel.com>
2009-12-14 12:01:15 +00:00
Joshua Lock
43a1d901f5 local.conf: Fix cross canadian variable, it's SDKMACHINE not SDK_ARCH 2009-12-11 17:25:52 +00:00
Joshua Lock
3ca17dfc31 local.conf: mention the Canadian Cross variable; SDK_ARCH 2009-12-11 15:03:45 +00:00
Joshua Lock
d7919b22fe local.conf: use qemux86 as default target 2009-12-11 15:03:45 +00:00
Joshua Lock
e8bb07c601 handbook: point people at purple in the introduction 2009-12-11 15:03:45 +00:00
Joshua Lock
8626d103f7 poky.conf: change version for release of purple-3.2 2009-12-11 15:03:45 +00:00
Joshua Lock
0b02870404 qemugl-nativesdk: Add a nativesdk package 2009-12-11 15:03:36 +00:00
Enric Balletbo i Serra
7ed5c2cebc udev: Add local.rules for the IGEP v2 to symlink the first framebuffer to /dev/fb
Signed-off-by: Enric Balletbo i Serra <eballetbo@iseebcn.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-10 16:50:20 +00:00
Joshua Lock
12e3b599ca qemugl: remove staging function, no need to stage anything at all 2009-12-09 16:23:45 +00:00
Richard Purdie
76f6ab09ca leadpad/pcmanfm/puzzles: Make libowl optional
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-08 21:09:40 +00:00
Richard Purdie
a68bc3912a tasks: Fix missing quote
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-08 21:09:22 +00:00
Richard Purdie
1411386a6b poky-fixed-revisions.inc: Update dates revision
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-08 21:08:41 +00:00
Richard Purdie
9cb8f533fe pimlico: Clean up libowl handling to be consistent 2009-12-08 21:08:17 +00:00
Richard Purdie
f2bc66b102 libowl: Remove stale comments and enable packaging
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-08 21:06:37 +00:00
Richard Purdie
5dfee8d314 bitbake: Fix dict changing size problems
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:02:41 +00:00
Richard Purdie
c1ee270d2c poky-init-build-env: Remove now unneeded return
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:02:34 +00:00
Richard Purdie
80ca44d44c distro/poky: Set EXTRAOPKGCONFIG
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:02:24 +00:00
Richard Purdie
1218b39f24 Add poky-feed-config-opkg (based off moblin version)
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:02:14 +00:00
Richard Purdie
c846e5848e scripts/poky-qemu-internal: Make sure qemuarm cursor is visible
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:02:02 +00:00
Richard Purdie
d2ce340b34 libxcalibrate: Fix with xcb
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:01:52 +00:00
Richard Purdie
5f72e5d0f2 pointercal: Update qemuarm calibration files after recent qemu changes to wacom code
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:01:35 +00:00
Richard Purdie
1a651898a8 udev-141: Backport fix from 145 which is also needed here
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:01:20 +00:00
Richard Purdie
0e1d3d91d6 qemu: Update to 0.10.6 and add fix wacom emulation issues
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 18:01:00 +00:00
Richard Purdie
935b598736 udev: Fix touchscreen link creation with latest version
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-12-03 17:59:39 +00:00
Richard Purdie
1a6a15e568 package.bbclass: Fix runtime_maping_rename so version information isn't removed
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-11-20 17:57:16 +00:00
Nicolas Dechesne
c7edcc0316 sato: move pam from meta-moblin to meta
pam recipe is required to build a sato image, and it was moved in the
meta-moblin collection by mistake, and sato build was broken.

pam is a dependency of policykit which is a dependency of networkmanager.

Signed-off-by: Nicolas Dechesne <n-dechesne@ti.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-11-20 17:49:17 +00:00
Nicolas Dechesne
6d80967bcf sato: move policykit from meta-moblin to met
policykit recipe is required to build a sato image, and it was moved in the
meta-moblin collection by mistake, and sato build was broken.

policykit is a dependency of networkmanager.

Signed-off-by: Nicolas Dechesne <n-dechesne@ti.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
2009-11-20 17:49:08 +00:00
8509 changed files with 5129789 additions and 714993 deletions

24
.gitignore vendored
View File

@@ -1,23 +1,13 @@
*.pyc
*.pyo
/*.patch
build*/
pyshtables.py
pstage/
scripts/oe-git-proxy-socks
build/conf/local.conf
build/tmp/
scripts/poky-git-proxy-socks
sources/
meta-*/
!meta-skeleton
!meta-hob
hob-image-*.bb
meta-darwin
meta-maemo
meta-prvt*
poky-autobuilder*
*.swp
*.orig
*.rej
*~
!meta-yocto
!meta-yocto-bsp
!meta-yocto-imported
documentation/user-manual/user-manual.html
documentation/user-manual/user-manual.pdf
documentation/user-manual/user-manual.tgz
pull-*/

View File

@@ -1,2 +0,0 @@
# Template settings
TEMPLATECONF=${TEMPLATECONF:-meta-yocto/conf}

13
LICENSE
View File

@@ -1,14 +1,11 @@
Different components of OpenEmbedded are under different licenses (a mix
of MIT and GPLv2). Please see:
Different components of Poky are under different licenses (a mix of
MIT and GPLv2). Please see:
meta/COPYING.GPLv2 (GPLv2)
bitbake/COPYING (GPLv2)
meta/COPYING.MIT (MIT)
meta-selftest/COPYING.MIT (MIT)
meta-skeleton/COPYING.MIT (MIT)
meta-extras/COPYING.MIT (MIT)
All metadata is MIT licensed unless otherwise stated. Source code
included in tree for individual recipes is under the LICENSE stated in
the associated recipe (.bb file) unless otherwise stated.
which cover the components in those subdirectories.
License information for any other files is either explicitly stated
or defaults to GPL version 2.

54
README
View File

@@ -1,49 +1,15 @@
Poky
====
Poky is an integration of various components to form a complete prepackaged
build system and development environment. It features support for building
customised embedded device style images. There are reference demo images
featuring a X11/Matchbox/GTK themed UI called Sato. The system supports
cross-architecture application development using QEMU emulation and a
standalone toolchain and SDK with IDE integration.
Poky platform builder is a combined cross build system and development
environment. It features support for building X11/Matchbox/GTK based
filesystem images for various embedded devices and boards. It also
supports cross-architecture application development using QEMU emulation
and a standalone toolchain and SDK with IDE integration.
Poky has an extensive handbook, the source of which is contained in
the handbook directory. For compiled HTML or pdf versions of this,
see the Poky website http://pokylinux.org.
Additional information on the specifics of hardware that Poky supports
is available in README.hardware. Further hardware support can easily be added
in the form of layers which extend the systems capabilities in a modular way.
As an integration layer Poky consists of several upstream projects such as
BitBake, OpenEmbedded-Core, Yocto documentation and various sources of information
e.g. for the hardware support. Poky is in turn a component of the Yocto Project.
The Yocto Project has extensive documentation about the system including a
reference manual which can be found at:
http://yoctoproject.org/documentation
OpenEmbedded-Core is a layer containing the core metadata for current versions
of OpenEmbedded. It is distro-less (can build a functional image with
DISTRO = "nodistro") and contains only emulated machine support.
For information about OpenEmbedded, see the OpenEmbedded website:
http://www.openembedded.org/
Where to Send Patches
=====================
As Poky is an integration repository, patches against the various components
should be sent to their respective upstreams.
bitbake:
bitbake-devel@lists.openembedded.org
meta-yocto:
poky@yoctoproject.org
Most everything else should be sent to the OpenEmbedded Core mailing list. If
in doubt, check the oe-core git repository for the content you intend to modify.
Before sending, be sure the patches apply cleanly to the current oe-core git
repository.
openembedded-core@lists.openembedded.org
Note: The scripts directory should be treated with extra care as it is a mix
of oe-core and poky-specific files.
is available in README.hardware.

View File

@@ -1,476 +1,436 @@
Poky Hardware README
====================
Poky Hardware Reference Guide
=============================
This file gives details about using Poky with the reference machines
supported out of the box. A full list of supported reference target machines
can be found by looking in the following directories:
This file gives details about using Poky with different hardware reference
boards and consumer devices. A full list of target machines can be found by
looking in the meta/conf/machine/ directory. If in doubt about using Poky with
your hardware, consult the documentation for your board/device. To discuss
support for further hardware reference boards/devices please contact OpenedHand.
meta/conf/machine/
meta-yocto-bsp/conf/machine/
If you are in doubt about using Poky/OpenEmbedded with your hardware, consult
the documentation for your board/device.
Support for additional devices is normally added by creating BSP layers - for
more information please see the Yocto Board Support Package (BSP) Developer's
Guide - documentation source is in documentation/bspguide or download the PDF
from:
http://yoctoproject.org/documentation
Support for physical reference hardware has now been split out into a
meta-yocto-bsp layer which can be removed separately from other layers if not
needed.
QEMU Emulation Targets
======================
To simplify development, the build system supports building images to
work with the QEMU emulator in system emulation mode. Several architectures
are currently supported:
* ARM (qemuarm)
* x86 (qemux86)
* x86-64 (qemux86-64)
* PowerPC (qemuppc)
* MIPS (qemumips)
Use of the QEMU images is covered in the Yocto Project Reference Manual.
The appropriate MACHINE variable value corresponding to the target is given
in brackets.
QEMU Emulation Images (qemuarm and qemux86)
===========================================
To simplify development Poky supports building images to work with the QEMU
emulator in system emulation mode. Two architectures are currently supported,
ARM (via qemuarm) and x86 (via qemux86). Use of the QEMU images is covered
in the Poky Handbook.
Hardware Reference Boards
=========================
The following boards are supported by the meta-yocto-bsp layer:
The following boards are supported by Poky:
* Texas Instruments Beaglebone (beaglebone)
* Freescale MPC8315E-RDB (mpc8315e-rdb)
For more information see the board's section below. The appropriate MACHINE
variable value corresponding to the board is given in brackets.
* Compulab CM-X270 (cm-x270)
* Compulab EM-X270 (em-x270)
* FreeScale iMX31ADS (mx31ads)
* Marvell PXA3xx Zylonite (zylonite)
* Logic iMX31 Lite Kit (mx31litekit)
* Phytec phyCORE-iMX31 (mx31phy)
For more information see board's section below. The Poky MACHINE setting
corresponding to the board is given in brackets.
Consumer Devices
================
The following consumer devices are supported by the meta-yocto-bsp layer:
The following consumer devices are supported by Poky:
* Intel x86 based PCs and devices (genericx86)
* Ubiquiti Networks EdgeRouter Lite (edgerouter)
* FIC Neo1973 GTA01 smartphone (fic-gta01)
* HTC Universal (htcuniversal)
* Nokia 770/N800/N810 Internet Tablets (nokia770 and nokia800)
* Sharp Zaurus SL-C7x0 series (c7x0)
* Sharp Zaurus SL-C1000 (akita)
* Sharp Zaurus SL-C3x00 series (spitz)
For more information see the device's section below. The appropriate MACHINE
variable value corresponding to the device is given in brackets.
For more information see board's section below. The Poky MACHINE setting
corresponding to the board is given in brackets.
Poky Boot CD (bootcdx86)
========================
The Poky boot CD iso images are designed as a demonstration of the Poky
environment and to show the versatile image formats Poky can generate. It will
run on Pentium2 or greater PC style computers. The iso image can be
burnt to CD and then booted from.
Specific Hardware Documentation
===============================
Hardware Reference Boards
=========================
Intel x86 based PCs and devices (genericx86)
==========================================
Compulab CM-X270 (cm-x270)
==========================
The genericx86 MACHINE is tested on the following platforms:
The bootloader on this board doesn't support writing jffs2 images directly to
NAND and normally uses a proprietary kernel flash driver. To allow the use of
jffs2 images, a two stage updating procedure is needed. Firstly, an initramfs
is booted which contains mtd utilities and this is then used to write the main
filesystem.
Intel Xeon/Core i-Series:
+ Intel Romley Server: Sandy Bridge Xeon processor, C600 PCH (Patsburg), (Canoe Pass CRB)
+ Intel Romley Server: Ivy Bridge Xeon processor, C600 PCH (Patsburg), (Intel SDP S2R3)
+ Intel Crystal Forest Server: Sandy Bridge Xeon processor, DH89xx PCH (Cave Creek), (Stargo CRB)
+ Intel Chief River Mobile: Ivy Bridge Mobile processor, QM77 PCH (Panther Point-M), (Emerald Lake II CRB, Sabino Canyon CRB)
+ Intel Huron River Mobile: Sandy Bridge processor, QM67 PCH (Cougar Point), (Emerald Lake CRB, EVOC EC7-1817LNAR board)
+ Intel Calpella Platform: Core i7 processor, QM57 PCH (Ibex Peak-M), (Red Fort CRB, Emerson MATXM CORE-411-B)
+ Intel Nehalem/Westmere-EP Server: Xeon 56xx/55xx processors, 5520 chipset, ICH10R IOH (82801), (Hanlan Creek CRB)
+ Intel Nehalem Workstation: Xeon 56xx/55xx processors, System SC5650SCWS (Greencity CRB)
+ Intel Picket Post Server: Xeon 56xx/55xx processors (Jasper Forest), 3420 chipset (Ibex Peak), (Osage CRB)
+ Intel Storage Platform: Sandy Bridge Xeon processor, C600 PCH (Patsburg), (Oak Creek Canyon CRB)
+ Intel Shark Bay Client Platform: Haswell processor, LynxPoint PCH, (Walnut Canyon CRB, Lava Canyon CRB, Basking Ridge CRB, Flathead Creek CRB)
+ Intel Shark Bay Ultrabook Platform: Haswell ULT processor, Lynx Point-LP PCH, (WhiteTip Mountain 1 CRB)
It is assumed the board is connected to a network where a TFTP server is
available and that a serial terminal is available to communicate with the
bootloader (38400, 8N1). If a DHCP server is available the device will use it
to obtain an IP address. If not, run:
Intel Atom platforms:
+ Intel embedded Menlow: Intel Atom Z510/530 CPU, System Controller Hub US15W (Portwell NANO-8044)
+ Intel Luna Pier: Intel Atom N4xx/D5xx series CPU (aka: Pineview-D & -M), 82801HM I/O Hub (ICH8M), (Advantech AIMB-212, Moon Creek CRB)
+ Intel Queens Bay platform: Intel Atom E6xx CPU (aka: Tunnel Creek), Topcliff EG20T I/O Hub (Emerson NITX-315, Crown Bay CRB, Minnow Board)
+ Intel Fish River Island platform: Intel Atom E6xx CPU (aka: Tunnel Creek), Topcliff EG20T I/O Hub (Kontron KM2M806)
+ Intel Cedar Trail platform: Intel Atom N2000 & D2000 series CPU (aka: Cedarview), NM10 Express Chipset (Norco kit BIS-6630, Cedar Rock CRB)
ARMmon > setip dhcp off
ARMmon > setip ip 192.168.1.203
ARMmon > setip mask 255.255.255.0
and is likely to work on many unlisted Atom/Core/Xeon based devices. The MACHINE
type supports ethernet, wifi, sound, and Intel/vesa graphics by default in
addition to common PC input devices, busses, and so on. Note that it does not
included the binary-only graphic drivers used on some Atom platforms, for
accelerated graphics on these machines please refer to meta-intel.
To reflash the kernel:
Depending on the device, it can boot from a traditional hard-disk, a USB device,
or over the network. Writing generated images to physical media is
straightforward with a caveat for USB devices. The following examples assume the
target boot device is /dev/sdb, be sure to verify this and use the correct
device as the following commands are run as root and are not reversable.
ARMmon > download kernel tftp zimage 192.168.1.202
ARMmon > flash kernel
USB Device:
1. Build a live image. This image type consists of a simple filesystem
without a partition table, which is suitable for USB keys, and with the
default setup for the genericx86 machine, this image type is built
automatically for any image you build. For example:
where zimage is the name of the kernel on the TFTP server and its IP address is
192.168.1.202. The names of the files must be all lowercase.
$ bitbake core-image-minimal
To reflash the initrd/initramfs:
2. Use the "dd" utility to write the image to the raw block device. For
example:
ARMmon > download ramdisk tftp diskimage 192.168.1.202
ARMmon > flash ramdisk
# dd if=core-image-minimal-genericx86.hddimg of=/dev/sdb
where diskimage is the name of the initramfs image (a cpio.gz file).
If the device fails to boot with "Boot error" displayed, or apparently
stops just after the SYSLINUX version banner, it is likely the BIOS cannot
understand the physical layout of the disk (or rather it expects a
particular layout and cannot handle anything else). There are two possible
solutions to this problem:
To boot the initramfs:
1. Change the BIOS USB Device setting to HDD mode. The label will vary by
device, but the idea is to force BIOS to read the Cylinder/Head/Sector
geometry from the device.
ARMmon > ramdisk on
ARMmon > bootos "console=ttyS0,38400 rdinit=/sbin/init"
2. Without such an option, the BIOS generally boots the device in USB-ZIP
mode. To write an image to a USB device that will be bootable in
USB-ZIP mode, carry out the following actions:
To reflash the main image login to the system as user "root", then run:
a. Determine the geometry of your USB device using fdisk:
# ifconfig eth0 192.168.1.203
# tftp -g -r mainimage 192.168.1.202
# flash_eraseall /dev/mtd1
# nandwrite /dev/mtd1 mainimage
# fdisk /dev/sdb
Command (m for help): p
which configures the network interface with the IP address 192.168.1.203,
downloads the "mainimage" file from the TFTP server at 192.168.1.202, erases
the flash and then writes the new image to the flash.
Disk /dev/sdb: 4011 MB, 4011491328 bytes
124 heads, 62 sectors/track, 1019 cylinders, total 7834944 sectors
...
The main image can then be booted with:
Command (m for help): q
ARMmon > bootos "console=ttyS0,38400 root=/dev/mtdblock1 rootfstype=jffs2"
b. Configure the USB device for USB-ZIP mode:
# mkdiskimage -4 /dev/sdb 1019 124 62
Note that the initramfs image is built by poky in a slightly different mode to
normal since it uses uclibc. To generate this use a command like:
Where 1019, 124 and 62 are the cylinder, head and sectors/track counts
as reported by fdisk (substitute the values reported for your device).
When the operation has finished and the access LED (if any) on the
device stops flashing, remove and reinsert the device to allow the
kernel to detect the new partition layout.
IMAGE_FSTYPES=cpio.gz MACHINE=cm-x270 POKYLIBC=uclibc bitbake poky-image-minimal-mtdutils
c. Copy the contents of the image to the USB-ZIP mode device:
# mkdir /tmp/image
# mkdir /tmp/usbkey
# mount -o loop core-image-minimal-genericx86.hddimg /tmp/image
# mount /dev/sdb4 /tmp/usbkey
# cp -rf /tmp/image/* /tmp/usbkey
Compulab EM-X270 (em-x270)
==========================
d. Install the syslinux boot loader:
Fetch the "Linux - kernel and run-time image (Angstrom)" ZIP file from the
Compulab website. Inside the images directory of this ZIP file is another ZIP
file called 'LiveDisk.zip'. Extract this over a cleanly formatted vfat USB flash
drive. Replace the 'em_x270.img' file with the 'updater-em-x270.ext2' file.
# syslinux /dev/sdb4
Insert this USB disk into the supplied adapter and connect this to the
board. Whilst holding down the the suspend button press the reset button. The
board will now boot off the USB key and into a version of Angstrom. On the
desktop is an icon labelled "Updater". Run this program to launch the updater
that will flash the Poky kernel and rootfs to the board.
e. Unmount everything:
# umount /tmp/image
# umount /tmp/usbkey
FreeScale iMX31ADS (mx31ads)
===========================
Install the boot device in the target board and configure the BIOS to boot
from it.
The correct serial port is the top-most female connector to the right of the
ethernet socket.
For more details on the USB-ZIP scenario, see the syslinux documentation:
http://git.kernel.org/?p=boot/syslinux/syslinux.git;a=blob_plain;f=doc/usbkey.txt;hb=HEAD
For uploading data to RedBoot we are going to use tftp. In this example we
assume that the tftpserver is on 192.168.9.1 and the board is on192.168.9.2.
To set the IP address, run:
Texas Instruments Beaglebone (beaglebone)
=========================================
ip_address -l 192.168.9.2/24 -h 192.168.9.1
The Beaglebone is an ARM Cortex-A8 development board with USB, Ethernet, 2D/3D
accelerated graphics, audio, serial, JTAG, and SD/MMC. The Black adds a faster
CPU, more RAM, eMMC flash and a micro HDMI port. The beaglebone MACHINE is
tested on the following platforms:
To download a kernel called "zimage" from the TFTP server, run:
o Beaglebone Black A6
o Beaglebone A6 (the original "White" model)
load -r -b 0x100000 zimage
The Beaglebone Black has eMMC, while the White does not. Pressing the USER/BOOT
button when powering on will temporarily change the boot order. But for the sake
of simplicity, these instructions assume you have erased the eMMC on the Black,
so its boot behavior matches that of the White and boots off of SD card. To do
this, issue the following commands from the u-boot prompt:
To write the kernel to flash run:
# mmc dev 1
# mmc erase 0 512
fis create kernel
To further tailor these instructions for your board, please refer to the
documentation at http://www.beagleboard.org/bone and http://www.beagleboard.org/black
To download a rootfs jffs2 image "rootfs" from the TFTP server, run:
From a Linux system with access to the image files perform the following steps
as root, replacing mmcblk0* with the SD card device on your machine (such as sdc
if used via a usb card reader):
load -r -b 0x100000 rootfs
1. Partition and format an SD card:
# fdisk -lu /dev/mmcblk0
To write the root filesystem to flash run:
Disk /dev/mmcblk0: 3951 MB, 3951034368 bytes
255 heads, 63 sectors/track, 480 cylinders, total 7716864 sectors
Units = sectors of 1 * 512 = 512 bytes
fis create root
Device Boot Start End Blocks Id System
/dev/mmcblk0p1 * 63 144584 72261 c Win95 FAT32 (LBA)
/dev/mmcblk0p2 144585 465884 160650 83 Linux
To load and boot a kernel and rootfs from flash:
# mkfs.vfat -F 16 -n "boot" /dev/mmcblk0p1
# mke2fs -j -L "root" /dev/mmcblk0p2
fis load kernel
exec -b 0x100000 -l 0x200000 -c "noinitrd console=ttymxc0,115200 root=/dev/mtdblock2 rootfstype=jffs2 init=linuxrc ip=none"
The following assumes the SD card partitions 1 and 2 are mounted at
/media/boot and /media/root respectively. Removing the card and reinserting
it will do just that on most modern Linux desktop environments.
To load and boot a kernel from a TFTP server with the rootfs over NFS:
The files referenced below are made available after the build in
build/tmp/deploy/images.
load -r -b 0x100000 zimage
exec -b 0x100000 -l 0x200000 -c "noinitrd console=ttymxc0,115200 root=/dev/nfs nfsroot=192.168.9.1:/mnt/nfsmx31 rw ip=192.168.9.2::192.168.9.1:255.255.255.0"
2. Install the boot loaders
# cp MLO-beaglebone /media/boot/MLO
# cp u-boot-beaglebone.img /media/boot/u-boot.img
The instructions above are for using the (default) NOR flash on the board,
there is also 128M of NAND flash. It is possible to install Poky to the NAND
flash which gives more space for the rootfs and instructions for using this are
given below. To switch to the NAND flash:
3. Install the root filesystem
# tar x -C /media/root -f core-image-$IMAGE_TYPE-beaglebone.tar.bz2
factive NAND
4. If using core-image-base or core-image-sato images, the SD card is ready
and rootfs already contains the kernel, modules and device tree (DTB)
files necessary to be booted with U-boot's default configuration, so
skip directly to step 8.
For core-image-minimal, proceed through next steps.
This will then restart RedBoot using the NAND rather than the NOR. If you
have not used the NAND before then it is unlikely that there will be a
partition table yet. You can get the list of partitions with 'fis list'.
5. If using core-image-minimal rootfs, install the modules
# tar x -C /media/root -f modules-beaglebone.tgz
If this shows no partitions then you can create them with:
6. If using core-image-minimal rootfs, install the kernel uImage into /boot
directory of rootfs
# cp uImage-beaglebone.bin /media/root/boot/uImage
fis init
7. If using core-image-minimal rootfs, also install device tree (DTB) files
into /boot directory of rootfs
# cp uImage-am335x-bone.dtb /media/root/boot/am335x-bone.dtb
# cp uImage-am335x-boneblack.dtb /media/root/boot/am335x-boneblack.dtb
The output of 'fis list' should now show:
8. Unmount the SD partitions, insert the SD card into the Beaglebone, and
boot the Beaglebone
Name FLASH addr Mem addr Length Entry point
RedBoot 0xE0000000 0xE0000000 0x00040000 0x00000000
FIS directory 0xE7FF4000 0xE7FF4000 0x00003000 0x00000000
RedBoot config 0xE7FF7000 0xE7FF7000 0x00001000 0x00000000
Partitions for the kernel and rootfs need to be created:
Freescale MPC8315E-RDB (mpc8315e-rdb)
=====================================
fis create -l 0x1A0000 -e 0x00100000 kernel
fis create -l 0x5000000 -e 0x00100000 root
The MPC8315 PowerPC reference platform (MPC8315E-RDB) is aimed at hardware and
software development of network attached storage (NAS) and digital media server
applications. The MPC8315E-RDB features the PowerQUICC II Pro processor, which
includes a built-in security accelerator.
You may now use the instructions above for flashing. However it is important
to note that the erase block size for the NAND is different to the NOR so the
JFFS erase size will need to be changed to 0x4000. Stardard images are built
for NOR and you will need to build custom images for NAND.
(Note: you may find it easier to order MPC8315E-RDBA; this appears to be the
same board in an enclosure with accessories. In any case it is fully
compatible with the instructions given here.)
You will also need to update the kernel command line to use the correct root
filesystem. This should be '/dev/mtdblock7' if you adhere to the partitioning
scheme shown above. If this fails then you can doublecheck against the output
from the kernel when it evaluates the available mtd partitions.
Setup instructions
------------------
You will need the following:
* NFS root setup on your workstation
* TFTP server installed on your workstation
* Straight-thru 9-conductor serial cable (DB9, M/F) connected from your
PC to UART1
* Ethernet connected to the first ethernet port on the board
Marvell PXA3xx Zylonite (zylonite)
==================================
--- Preparation ---
These instructions assume the Zylonite is connected to a machine running a TFTP
server at address 192.168.123.5 and that a serial link (38400 8N1) is available
to access the blob bootloader. The kernel is on the TFTP server as
"zylonite-kernel" and the root filesystem jffs2 file is "zylonite-rootfs" and
the images are to be saved in NAND flash.
Note: if you have altered your board's ethernet MAC address(es) from the
defaults, or you need to do so because you want multiple boards on the same
network, then you will need to change the values in the dts file (patch
linux/arch/powerpc/boot/dts/mpc8315erdb.dts within the kernel source). If
you have left them at the factory default then you shouldn't need to do
anything here.
The following commands setup blob:
--- Booting from NFS root ---
blob> setip client 192.168.123.4
blob> setip server 192.168.123.5
Load the kernel and dtb (device tree blob), and boot the system as follows:
To flash the kernel:
1. Get the kernel (uImage-mpc8315e-rdb.bin) and dtb (uImage-mpc8315e-rdb.dtb)
files from the tmp/deploy directory, and make them available on your TFTP
server.
blob> tftp zylonite-kernel
blob> nandwrite -j 0x80800000 0x60000 0x200000
2. Connect the board's first serial port to your workstation and then start up
your favourite serial terminal so that you will be able to interact with
the serial console. If you don't have a favourite, picocom is suggested:
To flash the rootfs:
$ picocom /dev/ttyUSB0 -b 115200
blob> tftp zylonite-rootfs
blob> nanderase -j 0x260000 0x5000000
blob> nandwrite -j 0x80800000 0x260000 <length>
3. Power up or reset the board and press a key on the terminal when prompted
to get to the U-Boot command line
(where <length> is the rootfs size which will be printed by the tftp step)
4. Set up the environment in U-Boot:
To boot the board:
=> setenv ipaddr <board ip>
=> setenv serverip <tftp server ip>
=> setenv bootargs root=/dev/nfs rw nfsroot=<nfsroot ip>:<rootfs path> ip=<board ip>:<server ip>:<gateway ip>:255.255.255.0:mpc8315e:eth0:off console=ttyS0,115200
blob> nkernel
blob> boot
5. Download the kernel and dtb, and boot:
=> tftp 1000000 uImage-mpc8315e-rdb.bin
=> tftp 2000000 uImage-mpc8315e-rdb.dtb
=> bootm 1000000 - 2000000
Logic iMX31 Lite Kit (mx31litekit)
===============================
The easiest method to boot this board is to take an MMC/SD card and format
the first partition as ext2, then extract the poky image onto this as root.
Assuming the board is network connected, a TFTP server is available at
192.168.1.33 and a serial terminal is available (115200 8N1), the following
commands will boot a kernel called "mx31kern" from the TFTP server:
Ubiquiti Networks EdgeRouter Lite (edgerouter)
==============================================
losh> ifconfig sm0 192.168.1.203 255.255.255.0 192.168.1.33
losh> load raw 0x80100000 0x200000 /tftp/192.168.1.33:mx31kern
losh> exec 0x80100000 -
The EdgeRouter Lite is part of the EdgeMax series. It is a MIPS64 router
(based on the Cavium Octeon processor) with 512MB of RAM, which uses an
internal USB pendrive for storage.
Setup instructions
------------------
Phytec phyCORE-iMX31 (mx31phy)
==============================
You will need the following:
* NFS root setup on your workstation
* TFTP server installed on your workstation
* RJ45 -> serial ("rollover") cable connected from your PC to the CONSOLE
port on the board
* Ethernet connected to the first ethernet port on the board
Support for this board is currently being developed. Experimental jffs2
images and a suitable kernel are available and are known to work with the
board.
--- Preparation ---
Build an image (e.g. core-image-minimal) using "edgerouter" as the MACHINE.
In the following instruction it is based on core-image-minimal. Another target
may be similiar with it.
Consumer Devices
================
--- Booting from NFS root ---
FIC Neo1973 GTA01 smartphone (fic-gta01)
========================================
Load the kernel, and boot the system as follows:
To install Poky on a GTA01 smartphone you will need "dfu-util" tool
which you can build with "bitbake dfu-util-native" command.
1. Get the kernel (vmlinux) file from the tmp/deploy/images/edgerouter
directory, and make them available on your TFTP server.
Flashing requires these steps:
2. Connect the board's first serial port to your workstation and then start up
your favourite serial terminal so that you will be able to interact with
the serial console. If you don't have a favourite, picocom is suggested:
1. Power down the device.
2. Connect the device to the host machine via USB.
3. Hold AUX key and press Power key. There should be a bootmenu
on screen.
4. Run "dfu-util -l" to check if the phone is visible on the USB bus.
The output should look like this:
$ picocom /dev/ttyS0 -b 115200
dfu-util - (C) 2007 by OpenMoko Inc.
This program is Free Software and has ABSOLUTELY NO WARRANTY
3. Power up or reset the board and press a key on the terminal when prompted
to get to the U-Boot command line
Found Runtime: [0x1457:0x5119] devnum=19, cfg=0, intf=2, alt=0, name="USB Device Firmware Upgrade"
4. Set up the environment in U-Boot:
5. Flash the kernel with "dfu-util -a kernel -D uImage-2.6.21.6-moko11-r2-fic-gta01.bin"
6. Flash rootfs with "dfu-util -a rootfs -D <image>", where <image> is the
jffs2 image file to use as the root filesystem
(e.g. ./tmp/deploy/images/poky-image-sato-fic-gta01.jffs2)
=> setenv ipaddr <board ip>
=> setenv serverip <tftp server ip>
5. Download the kernel and boot:
HTC Universal (htcuniversal)
============================
=> tftp tftp $loadaddr vmlinux
=> bootoctlinux $loadaddr coremask=0x3 root=/dev/nfs rw nfsroot=<nfsroot ip>:<rootfs path> ip=<board ip>:<server ip>:<gateway ip>:<netmask>:edgerouter:eth0:off mtdparts=phys_mapped_flash:512k(boot0),512k(boot1),64k@3072k(eeprom)
Note: HTC Universal support is highly experimental.
--- Booting from USB root ---
On the HTC Universal, entirely replacing the Windows installation is not
supported, instead Poky is booted from an MMC/SD card from Windows. Once Poky
has booted, Windows is no longer in memory or active but when power is removed,
the user will be returned to windows and will need to return to Linux from
there.
To boot from the USB disk, you either need to remove it from the edgerouter
box and populate it from another computer, or use a previously booted NFS
image and populate from the edgerouter itself.
Once an MMC/SD card is available it is suggested its split into two partitions,
one for a program called HaRET which lets you boot Linux from within Windows
and the second for the rootfs. The HaRET partition should be the first partition
on the card and be vfat formatted. It doesn't need to be large, just enough for
HaRET and a kernel (say 5MB max). The rootfs should be ext2 and is usually the
second partition. The first partition should be vfat so Windows recognises it
as if it doesn't, it has been known to reformat cards.
Type 1: Mounted USB disk
------------------------
On the first partition you need three files:
To boot from the USB disk there are two available partitions on the factory
USB storage. The rest of this guide assumes that these partitions are left
intact. If you change the partition scheme, you must update your boot method
appropriately.
* a HaRET binary (version 0.5.1 works well and a working version
should be part of the last Poky release)
* a kernel renamed to "zImage"
* a default.txt which contains:
The standard partitions are:
set kernel "zImage"
set mtype "855"
set cmdline "root=/dev/mmcblk0p2 rw console=ttyS0,115200n8 console=tty0 rootdelay=5 fbcon=rotate:1"
boot2
- 1: vfat partition containing factory kernels
- 2: ext3 partition for the root filesystem.
On the second parition the root file system is extracted as root. A different
partition layout or other kernel options can be changed in the default.txt file.
You can place the kernel on either partition 1, or partition 2, but the roofs
must go on partition 2 (due to its size).
When inserted into the device, Windows should see the card and let you browse
its contents using File Explorer. Running the HaRET binary will present a dialog
box (maybe after messages warning about running unsigned binaries) where you
select OK and you should then see Poky boot. Kernel messages can be seen by
adding psplash=false to the kernel commandline.
Note: If you place the kernel on the ext3 partition, you must re-create the
ext3 filesystem, since the factory u-boot can only handle 128 byte inodes and
cannot read the partition otherwise.
Steps:
Nokia 770/N800/N810 Internet Tablets (nokia770 and nokia800)
============================================================
1. Remove the USB disk from the edgerouter and insert it into a computer
that has access to your build artifacts.
Note: Nokia tablet support is highly experimental.
2. Copy the kernel image to the USB storage (assuming discovered as 'sdb' on
the development machine):
The Nokia internet tablet devices are OMAP based tablet formfactor devices
with large screens (800x480), wifi and touchscreen.
2a) if booting from vfat
# mount /dev/sdb1 /mnt
# cp tmp/deploy/images/edgerouter/vmlinux /mnt
# umount /mnt
To flash images to these devices you need the "flasher" utility which can be
downloaded from the http://tablets-dev.nokia.com/d3.php?f=flasher-3.0. This
utility needs to be run as root and the usb filesystem needs to be mounted
although most distributions will have done this for you. Once you have this
follow these steps:
2b) if booting from ext3
1. Power down the device.
2. Connect the device to the host machine via USB
(connecting power to the device doesn't hurt either).
3. Run "flasher -i"
4. Power on the device.
5. The program should give an indication it's found
a tablet device. If not, recheck the cables, make sure you're
root and usbfs/usbdevfs is mounted.
6. Run "flasher -r <image> -k <kernel> -f", where <image> is the
jffs2 image file to use as the root filesystem
(e.g. ./tmp/deploy/images/poky-image-sato-nokia800.jffs2)
and <kernel> is the kernel to use
(e.g. ./tmp/deploy/images/zImage-nokia800.bin).
7. Run "flasher -R" to reboot the device.
8. The device should boot into Poky.
# mkfs.ext3 -I 128 /dev/sdb2
# mount /dev/sdb2 /mnt
# mkdir /mnt/boot
# cp tmp/deploy/images/edgerouter/vmlinux /mnt/boot
# umount /mnt
The nokia800 images and kernel will run on both the N800 and N810.
3. Extract the rootfs to the USB storage ext3 partition
# mount /dev/sdb2 /mnt
# tar -xvjpf core-image-minimal-XXX.tar.bz2 -C /mnt
# umount /mnt
Sharp Zaurus SL-C7x0 series (c7x0)
==================================
4. Reboot the board and press a key on the terminal when prompted to get to the U-Boot
command line:
The Sharp Zaurus c7x0 series (SL-C700, SL-C750, SL-C760, SL-C860, SL-7500)
are PXA25x based handheld PDAs with VGA screens. To install Poky images on
these devices follow these steps:
5. Load the kernel and boot:
1. Obtain an SD/MMC or CF card with a vfat or ext2 filesystem.
2. Copy a jffs2 image file (e.g. poky-image-sato-c7x0.jffs2) onto the
card as "initrd.bin":
5a) vfat boot
$ cp ./tmp/deploy/images/poky-image-sato-c7x0.jffs2 /path/to/my-cf-card/initrd.bin
=> fatload usb 0:1 $loadaddr vmlinux
3. Copy an Linux kernel file (zImage-c7x0.bin) onto the card as
"zImage.bin":
5b) ext3 boot
$ cp ./tmp/deploy/images/zImage-c7x0.bin /path/to/my-cf-card/zImage.bin
=> ext2load usb 0:2 $loadaddr boot/vmlinux
=> bootoctlinux $loadaddr coremask=0x3 root=/dev/sda2 rw rootwait mtdparts=phys_mapped_flash:512k(boot0),512k(boot1),64k@3072k(eeprom)
4. Copy an updater script (updater.sh.c7x0) onto the card
as "updater.sh":
Type 2: NFS
-----------
$ cp ./tmp/deploy/images/updater.sh.c7x0 /path/to/my-cf-card/updater.sh
Note: If you place the kernel on the ext3 partition, you must re-create the
ext3 filesystem, since the factory u-boot can only handle 128 byte inodes and
cannot read the partition otherwise.
5. Power down the Zaurus.
6. Hold "OK" key and power on the device. An update menu should appear
(in Japanese).
7. Choose "Update" (item 4).
8. The next screen will ask for the source, choose the appropriate
card (CF or SD).
9. Make sure AC power is connected.
10. The next screen asks for confirmation, choose "Yes" (the left button).
11. The update process will start, flash the files on the card onto
the device and the device will then reboot into Poky.
These boot instructions assume that you have recreated the ext3 filesystem with
128 byte inodes, you have an updated uboot or you are running and image capable
of making the filesystem on the board itself.
Sharp Zaurus SL-C1000 (akita)
=============================
1. Boot from NFS root
The Sharp Zaurus SL-C1000 is a PXA270 based device otherwise similar to the
c7x0. To install Poky images on this device follow the instructions for
the c7x0 but replace "c7x0" with "akita" where appropriate.
2. Mount the USB disk partition 2 and then extract the contents of
tmp/deploy/core-image-XXXX.tar.bz2 into it.
Before starting, copy core-image-minimal-xxx.tar.bz2 and vmlinux into
rootfs path on your workstation.
Sharp Zaurus SL-C3x00 series (spitz)
====================================
and then,
# mount /dev/sda2 /media/sda2
# tar -xvjpf core-image-minimal-XXX.tar.bz2 -C /media/sda2
# cp vmlinux /media/sda2/boot/vmlinux
# umount /media/sda2
# reboot
The Sharp Zaurus SL-C3x00 devices are PXA270 based devices similar
to akita but with an internal microdrive. The installation procedure
assumes a standard microdrive based device where the root (first)
partition has been enlarged to fit the image (at least 100MB,
400MB for the SDK).
3. Reboot the board and press a key on the terminal when prompted to get to the U-Boot
command line:
The procedure is the same as for the c7x0 and akita models with the
following differences:
1. Instead of a jffs2 image you need to copy a compressed tarball of the
root fileystem (e.g. poky-image-sato-spitz.tar.gz) onto the
card as "hdimage1.tgz":
$ cp ./tmp/deploy/images/poky-image-sato-spitz.tar.gz /path/to/my-cf-card/hdimage1.tgz
2. You additionally need to copy a special tar utility (gnu-tar) onto
the card as "gnu-tar":
$ cp ./tmp/deploy/images/gnu-tar /path/to/my-cf-card/gnu-tar
# reboot
4. Load the kernel and boot:
=> ext2load usb 0:2 $loadaddr boot/vmlinux
=> bootoctlinux $loadaddr coremask=0x3 root=/dev/sda2 rw rootwait mtdparts=phys_mapped_flash:512k(boot0),512k(boot1),64k@3072k(eeprom)

10
bitbake-dev/AUTHORS Normal file
View File

@@ -0,0 +1,10 @@
Tim Ansell <mithro@mithis.net>
Phil Blundell <pb@handhelds.org>
Seb Frankengul <seb@frankengul.org>
Holger Freyther <zecke@handhelds.org>
Marcin Juszkiewicz <marcin@haerwu.biz>
Chris Larson <kergoth@handhelds.org>
Ulrich Luckas <luckas@musoft.de>
Mickey Lauer <mickey@Vanille.de>
Richard Purdie <rpurdie@rpsys.net>
Holger Schurig <holgerschurig@gmx.de>

315
bitbake-dev/ChangeLog Normal file
View File

@@ -0,0 +1,315 @@
Changes in Bitbake 1.9.x:
- Add PE (Package Epoch) support from Philipp Zabel (pH5)
- Treat python functions the same as shell functions for logging
- Use TMPDIR/anonfunc as a __anonfunc temp directory (T)
- Catch truncated cache file errors
- Allow operations other than assignment on flag variables
- Add code to handle inter-task dependencies
- Fix cache errors when generation dotGraphs
- Make sure __inherit_cache is updated before calling include() (from Michael Krelin)
- Fix bug when target was in ASSUME_PROVIDED (#2236)
- Raise ParseError for filenames with multiple underscores instead of infinitely looping (#2062)
- Fix invalid regexp in BBMASK error handling (missing import) (#1124)
- Promote certain warnings from debug to note 2 level
- Update manual
- Correctly redirect stdin when forking
- If parsing errors are found, exit, too many users miss the errors
- Remove supriours PREFERRED_PROVIDER warnings
- svn fetcher: Add _buildsvncommand function
- Improve certain error messages
- Rewrite svn fetcher to make adding extra operations easier
as part of future SRCDATE="now" fixes
(requires new FETCHCMD_svn definition in bitbake.conf)
- Change SVNDIR layout to be more unique (fixes #2644 and #2624)
- Add ConfigParsed Event after configuration parsing is complete
- Add SRCREV support for svn fetcher
- data.emit_var() - only call getVar if we need the variable
- Stop generating the A variable (seems to be legacy code)
- Make sure intertask depends get processed correcting in recursive depends
- Add pn-PN to overrides when evaluating PREFERRED_VERSION
- Improve the progress indicator by skipping tasks that have
already run before starting the build rather than during it
- Add profiling option (-P)
- Add BB_SRCREV_POLICY variable (clear or cache) to control SRCREV cache
- Add SRCREV_FORMAT support
- Fix local fetcher's localpath return values
- Apply OVERRIDES before performing immediate expansions
- Allow the -b -e option combination to take regular expressions
- Fix handling of variables with expansion in the name using _append/_prepend
e.g. RRECOMMENDS_${PN}_append_xyz = "abc"
- Add plain message function to bb.msg
- Sort the list of providers before processing so dependency problems are
reproducible rather than effectively random
- Fix/improve bitbake -s output
- Add locking for fetchers so only one tries to fetch a given file at a given time
- Fix int(0)/None confusion in runqueue.py which causes random gaps in dependency chains
- Expand data in addtasks
- Print the list of missing DEPENDS,RDEPENDS for the "No buildable providers available for required...."
error message.
- Rework add_task to be more efficient (6% speedup, 7% number of function calls reduction)
- Sort digraph output to make builds more reproducible
- Split expandKeys into two for loops to benefit from the expand_cache (12% speedup)
- runqueue.py: Fix idepends handling to avoid dependency errors
- Clear the terminal TOSTOP flag if set (and warn the user)
- Fix regression from r653 and make SRCDATE/CVSDATE work for packages again
- Fix a bug in bb.decodeurl where http://some.where.com/somefile.tgz decoded to host="" (#1530)
- Warn about malformed PREFERRED_PROVIDERS (#1072)
- Add support for BB_NICE_LEVEL option (#1627)
- Psyco is used only on x86 as there is no support for other architectures.
- Sort initial providers list by default preference (#1145, #2024)
- Improve provider sorting so prefered versions have preference over latest versions (#768)
- Detect builds of tasks with overlapping providers and warn (will become a fatal error) (#1359)
- Add MULTI_PROVIDER_WHITELIST variable to allow known safe multiple providers to be listed
- Handle paths in svn fetcher module parameter
- Support the syntax "export VARIABLE"
- Add bzr fetcher
- Add support for cleaning directories before a task in the form:
do_taskname[cleandirs] = "dir"
- bzr fetcher tweaks from Robert Schuster (#2913)
- Add mercurial (hg) fetcher from Robert Schuster (#2913)
- Don't add duplicates to BBPATH
- Fix preferred_version return values (providers.py)
- Fix 'depends' flag splitting
- Fix unexport handling (#3135)
- Add bb.copyfile function similar to bb.movefile (and improve movefile error reporting)
- Allow multiple options for deptask flag
- Use git-fetch instead of git-pull removing any need for merges when
fetching (we don't care about the index). Fixes fetch errors.
- Add BB_GENERATE_MIRROR_TARBALLS option, set to 0 to make git fetches
faster at the expense of not creating mirror tarballs.
- SRCREV handling updates, improvements and fixes from Poky
- Add bb.utils.lockfile() and bb.utils.unlockfile() from Poky
- Add support for task selfstamp and lockfiles flags
- Disable task number acceleration since it can allow the tasks to run
out of sequence
- Improve runqueue code comments
- Add task scheduler abstraction and some example schedulers
- Improve circular dependency chain debugging code and user feedback
- Don't give a stacktrace for invalid tasks, have a user friendly message (#3431)
- Add support for "-e target" (#3432)
- Fix shell showdata command (#3259)
- Fix shell data updating problems (#1880)
- Properly raise errors for invalid source URI protocols
- Change the wget fetcher failure handling to avoid lockfile problems
- Add support for branches in git fetcher (Otavio Salvador, Michael Lauer)
- Make taskdata and runqueue errors more user friendly
- Add norecurse and fullpath options to cvs fetcher
- Fix exit code for build failures in --continue mode
- Fix git branch tags fetching
- Change parseConfigurationFile so it works on real data, not a copy
- Handle 'base' inherit and all other INHERITs from parseConfigurationFile
instead of BBHandler
- Fix getVarFlags bug in data_smart
- Optmise cache handling by more quickly detecting an invalid cache, only
saving the cache when its changed, moving the cache validity check into
the parsing loop and factoring some getVar calls outside a for loop
- Cooker: Remove a debug message from the parsing loop to lower overhead
- Convert build.py exec_task to use getVarFlags
- Update shell to use cooker.buildFile
- Add StampUpdate event
- Convert -b option to use taskdata/runqueue
- Remove digraph and switch to new stamp checking code. exec_task no longer
honours dependencies
- Make fetcher timestamp updating non-fatal when permissions don't allow
updates
- Add BB_SCHEDULER variable/option ("completion" or "speed") controlling
the way bitbake schedules tasks
- Add BB_STAMP_POLICY variable/option ("perfile" or "full") controlling
how extensively stamps are looked at for validity
- When handling build target failures make sure idepends are checked and
failed where needed. Fixes --continue mode crashes.
- Fix -f (force) in conjunction with -b
- Fix problems with recrdeptask handling where some idepends weren't handled
correctly.
- Handle exit codes correctly (from pH5)
- Work around refs/HEAD issues with git over http (#3410)
- Add proxy support to the CVS fetcher (from Cyril Chemparathy)
- Improve runfetchcmd so errors are seen and various GIT variables are exported
- Add ability to fetchers to check URL validity without downloading
- Improve runtime PREFERRED_PROVIDERS warning message
- Add BB_STAMP_WHITELIST option which contains a list of stamps to ignore when
checking stamp dependencies and using a BB_STAMP_POLICY of "whitelist"
- No longer weight providers on the basis of a package being "already staged". This
leads to builds being non-deterministic.
- Flush stdout/stderr before forking to fix duplicate console output
- Make sure recrdeps tasks include all inter-task dependencies of a given fn
- Add bb.runqueue.check_stamp_fn() for use by packaged-staging
- Add PERSISTENT_DIR to store the PersistData in a persistent
directory != the cache dir.
- Add md5 and sha256 checksum generation functions to utils.py
- Correctly handle '-' characters in class names (#2958)
- Make sure expandKeys has been called on the data dictonary before running tasks
- Correctly add a task override in the form task-TASKNAME.
- Revert the '-' character fix in class names since it breaks things
- When a regexp fails to compile for PACKAGES_DYNAMIC, print a more useful error (#4444)
- Allow to checkout CVS by Date and Time. Just add HHmm to the SRCDATE.
- Move prunedir function to utils.py and add explode_dep_versions function
- Raise an exception if SRCREV == 'INVALID'
- Fix hg fetcher username/password handling and fix crash
- Fix PACKAGES_DYNAMIC handling of packages with '++' in the name
- Rename __depends to __base_depends after configuration parsing so we don't
recheck the validity of the config files time after time
- Add better environmental variable handling. By default it will now only pass certain
whitelisted variables into the data store. If BB_PRESERVE_ENV is set bitbake will use
all variable from the environment. If BB_ENV_WHITELIST is set, that whitelist will be
used instead of the internal bitbake one. Alternatively, BB_ENV_EXTRAWHITE can be used
to extend the internal whitelist.
- Perforce fetcher fix to use commandline options instead of being overriden by the environment
- bb.utils.prunedir can cope with symlinks to directoriees without exceptions
- use @rev when doing a svn checkout
- Add osc fetcher (from Joshua Lock in Poky)
- When SRCREV autorevisioning for a recipe is in use, don't cache the recipe
- Add tryaltconfigs option to control whether bitbake trys using alternative providers
to fulfil failed dependencies. It defaults to off, changing the default since this
behaviour confuses many users and isn't often useful.
- Improve lock file function error handling
- Add username handling to the git fetcher (Robert Bragg)
- Add support for HTTP_PROXY and HTTP_PROXY_IGNORE variables to the wget fetcher
- Export more variables to the fetcher commands to allow ssh checkouts and checkouts through
proxies to work better. (from Poky)
- Also allow user and pswd options in SRC_URIs globally (from Poky)
- Improve proxy handling when using mirrors (from Poky)
- Add bb.utils.prune_suffix function
- Fix hg checkouts of specific revisions (from Poky)
- Fix wget fetching of urls with parameters specified (from Poky)
- Add username handling to git fetcher (from Poky)
- Set HOME environmental variable when running fetcher commands (from Poky)
- Make sure allowed variables inherited from the environment are exported again (from Poky)
- When running a stage task in bbshell, run populate_staging, not the stage task (from Poky)
Changes in Bitbake 1.8.0:
- Release 1.7.x as a stable series
Changes in BitBake 1.7.x:
- Major updates of the dependency handling and execution
of tasks. Code from bin/bitbake replaced with runqueue.py
and taskdata.py
- New task execution code supports multithreading with a simplistic
threading algorithm controlled by BB_NUMBER_THREADS
- Change of the SVN Fetcher to keep the checkout around
courtsey of Paul Sokolovsky (#1367)
- PATH fix to bbimage (#1108)
- Allow debug domains to be specified on the commandline (-l)
- Allow 'interactive' tasks
- Logging message improvements
- Drop now uneeded BUILD_ALL_DEPS variable
- Add support for wildcards to -b option
- Major overhaul of the fetchers making a large amount of code common
including mirroring code
- Fetchers now touch md5 stamps upon access (to show activity)
- Fix -f force option when used without -b (long standing bug)
- Add expand_cache to data_cache.py, caching expanded data (speedup)
- Allow version field in DEPENDS (ignored for now)
- Add abort flag support to the shell
- Make inherit fail if the class doesn't exist (#1478)
- Fix data.emit_env() to expand keynames as well as values
- Add ssh fetcher
- Add perforce fetcher
- Make PREFERRED_PROVIDER_foobar defaults to foobar if available
- Share the parser's mtime_cache, reducing the number of stat syscalls
- Compile all anonfuncs at once!
*** Anonfuncs must now use common spacing format ***
- Memorise the list of handlers in __BBHANDLERS and tasks in __BBTASKS
This removes 2 million function calls resulting in a 5-10% speedup
- Add manpage
- Update generateDotGraph to use taskData/runQueue improving accuracy
and also adding a task dependency graph
- Fix/standardise on GPLv2 licence
- Move most functionality from bin/bitbake to cooker.py and split into
separate funcitons
- CVS fetcher: Added support for non-default port
- Add BBINCLUDELOGS_LINES, the number of lines to read from any logfile
- Drop shebangs from lib/bb scripts
Changes in Bitbake 1.6.0:
- Better msg handling
- COW dict implementation from Tim Ansell (mithro) leading
to better performance
- Speed up of -s
Changes in Bitbake 1.4.4:
- SRCDATE now handling courtsey Justin Patrin
- #1017 fix to work with rm_work
Changes in BitBake 1.4.2:
- Send logs to oe.pastebin.com instead of pastebin.com
fixes #856
- Copy the internal bitbake data before building the
dependency graph. This fixes nano not having a
virtual/libc dependency
- Allow multiple TARBALL_STASH entries
- Cache, check if the directory exists before changing
into it
- git speedup cloning by not doing a checkout
- allow to have spaces in filenames (.conf, .bb, .bbclass)
Changes in BitBake 1.4.0:
- Fix to check both RDEPENDS and RDEPENDS_${PN}
- Fix a RDEPENDS parsing bug in utils:explode_deps()
- Update git fetcher behaviour to match git changes
- ASSUME_PROVIDED allowed to include runtime packages
- git fetcher cleanup and efficency improvements
- Change the format of the cache
- Update usermanual to document the Fetchers
- Major changes to caching with a new strategy
giving a major performance increase when reparsing
with few data changes
Changes in BitBake 1.3.3:
- Create a new Fetcher module to ease the
development of new Fetchers.
Issue #438 fixed by rpurdie@openedhand.com
- Make the Subversion fetcher honor the SRC Date
(CVSDATE).
Issue #555 fixed by chris@openedhand.com
- Expand PREFERRED_PROVIDER properly
Issue #436 fixed by rprudie@openedhand.com
- Typo fix for Issue #531 by Philipp Zabel for the
BitBake Shell
- Introduce a new special variable SRCDATE as
a generic naming to replace CVSDATE.
- Introduce a new keyword 'required'. In contrast
to 'include' parsing will fail if a to be included
file can not be found.
- Remove hardcoding of the STAMP directory. Patch
courtsey pHilipp Zabel
- Track the RDEPENDS of each package (rpurdie@openedhand.com)
- Introduce BUILD_ALL_DEPS to build all RDEPENDS. E.g
this is used by the OpenEmbedded Meta Packages.
(rpurdie@openedhand.com).
Changes in BitBake 1.3.2:
- reintegration of make.py into BitBake
- bbread is gone, use bitbake -e
- lots of shell updates and bugfixes
- Introduction of the .= and =. operator
- Sort variables, keys and groups in bitdoc
- Fix regression in the handling of BBCOLLECTIONS
- Update the bitbake usermanual
Changes in BitBake 1.3.0:
- add bitbake interactive shell (bitbake -i)
- refactor bitbake utility in OO style
- kill default arguments in methods in the bb.data module
- kill default arguments in methods in the bb.fetch module
- the http/https/ftp fetcher will fail if the to be
downloaded file was not found in DL_DIR (this is needed
to avoid unpacking the sourceforge mirror page)
- Switch to a cow like data instance for persistent and non
persisting mode (called data_smart.py)
- Changed the callback of bb.make.collect_bbfiles to carry
additional parameters
- Drastically reduced the amount of needed RAM by not holding
each data instance in memory when using a cache/persistent
storage
Changes in BitBake 1.2.1:
The 1.2.1 release is meant as a intermediate release to lay the
ground for more radical changes. The most notable changes are:
- Do not hardcode {}, use bb.data.init() instead if you want to
get a instance of a data class
- bb.data.init() is a factory and the old bb.data methods are delegates
- Do not use deepcopy use bb.data.createCopy() instead.
- Removed default arguments in bb.fetch

207
bitbake-dev/bin/bitbake Executable file
View File

@@ -0,0 +1,207 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
# Copyright (C) 2003 - 2005 Michael 'Mickey' Lauer
# Copyright (C) 2005 Holger Hans Peter Freyther
# Copyright (C) 2005 ROAD GmbH
# Copyright (C) 2006 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import sys, os, getopt, re, time, optparse, xmlrpclib
sys.path.insert(0,os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb
from bb import cooker
from bb import daemonize
from bb import ui
from bb.ui import uievent
__version__ = "1.9.0"
if sys.hexversion < 0x020600F0:
print "Sorry, python 2.6 or later is required for this version of bitbake"
sys.exit(1)
#============================================================================#
# BBOptions
#============================================================================#
class BBConfiguration( object ):
"""
Manages build options and configurations for one run
"""
def __init__( self, options ):
for key, val in options.__dict__.items():
setattr( self, key, val )
def print_exception(exc, value, tb):
"""
Print the exception to stderr, only showing the traceback if bitbake
debugging is enabled.
"""
if not bb.msg.debug_level['default']:
tb = None
sys.__excepthook__(exc, value, tb)
#============================================================================#
# main
#============================================================================#
def main():
return_value = 0
pythonver = sys.version_info
if pythonver[0] < 2 or (pythonver[0] == 2 and pythonver[1] < 5):
print "Sorry, bitbake needs python 2.5 or later."
sys.exit(1)
parser = optparse.OptionParser( version = "BitBake Build Tool Core version %s, %%prog version %s" % ( bb.__version__, __version__ ),
usage = """%prog [options] [package ...]
Executes the specified task (default is 'build') for a given set of BitBake files.
It expects that BBFILES is defined, which is a space separated list of files to
be executed. BBFILES does support wildcards.
Default BBFILES are the .bb files in the current directory.""" )
parser.add_option( "-b", "--buildfile", help = "execute the task against this .bb file, rather than a package from BBFILES.",
action = "store", dest = "buildfile", default = None )
parser.add_option( "-k", "--continue", help = "continue as much as possible after an error. While the target that failed, and those that depend on it, cannot be remade, the other dependencies of these targets can be processed all the same.",
action = "store_false", dest = "abort", default = True )
parser.add_option( "-a", "--tryaltconfigs", help = "continue with builds by trying to use alternative providers where possible.",
action = "store_true", dest = "tryaltconfigs", default = False )
parser.add_option( "-f", "--force", help = "force run of specified cmd, regardless of stamp status",
action = "store_true", dest = "force", default = False )
parser.add_option( "-i", "--interactive", help = "drop into the interactive mode also called the BitBake shell.",
action = "store_true", dest = "interactive", default = False )
parser.add_option( "-c", "--cmd", help = "Specify task to execute. Note that this only executes the specified task for the providee and the packages it depends on, i.e. 'compile' does not implicitly call stage for the dependencies (IOW: use only if you know what you are doing). Depending on the base.bbclass a listtasks tasks is defined and will show available tasks",
action = "store", dest = "cmd" )
parser.add_option( "-r", "--read", help = "read the specified file before bitbake.conf",
action = "append", dest = "file", default = [] )
parser.add_option( "-v", "--verbose", help = "output more chit-chat to the terminal",
action = "store_true", dest = "verbose", default = False )
parser.add_option( "-D", "--debug", help = "Increase the debug level. You can specify this more than once.",
action = "count", dest="debug", default = 0)
parser.add_option( "-n", "--dry-run", help = "don't execute, just go through the motions",
action = "store_true", dest = "dry_run", default = False )
parser.add_option( "-p", "--parse-only", help = "quit after parsing the BB files (developers only)",
action = "store_true", dest = "parse_only", default = False )
parser.add_option( "-d", "--disable-psyco", help = "disable using the psyco just-in-time compiler (not recommended)",
action = "store_true", dest = "disable_psyco", default = False )
parser.add_option( "-s", "--show-versions", help = "show current and preferred versions of all packages",
action = "store_true", dest = "show_versions", default = False )
parser.add_option( "-e", "--environment", help = "show the global or per-package environment (this is what used to be bbread)",
action = "store_true", dest = "show_environment", default = False )
parser.add_option( "-g", "--graphviz", help = "emit the dependency trees of the specified packages in the dot syntax",
action = "store_true", dest = "dot_graph", default = False )
parser.add_option( "-I", "--ignore-deps", help = """Assume these dependencies don't exist and are already provided (equivalent to ASSUME_PROVIDED). Useful to make dependency graphs more appealing""",
action = "append", dest = "extra_assume_provided", default = [] )
parser.add_option( "-l", "--log-domains", help = """Show debug logging for the specified logging domains""",
action = "append", dest = "debug_domains", default = [] )
parser.add_option( "-P", "--profile", help = "profile the command and print a report",
action = "store_true", dest = "profile", default = False )
parser.add_option( "-u", "--ui", help = "userinterface to use",
action = "store", dest = "ui")
parser.add_option( "", "--revisions-changed", help = "Set the exit code depending on whether upstream floating revisions have changed or not",
action = "store_true", dest = "revisions_changed", default = False )
options, args = parser.parse_args(sys.argv)
configuration = BBConfiguration(options)
configuration.pkgs_to_build = []
configuration.pkgs_to_build.extend(args[1:])
cooker = bb.cooker.BBCooker(configuration)
# Clear away any spurious environment variables. But don't wipe the
# environment totally. This is necessary to ensure the correct operation
# of the UIs (e.g. for DISPLAY, etc.)
bb.utils.clean_environment()
cooker.parseCommandLine()
host = cooker.server.host
port = cooker.server.port
# Save a logfile for cooker into the current working directory. When the
# server is daemonized this logfile will be truncated.
cooker_logfile = os.path.join (os.getcwd(), "cooker.log")
daemonize.createDaemon(cooker.serve, cooker_logfile)
del cooker
# Setup a connection to the server (cooker)
server = xmlrpclib.Server("http://%s:%s" % (host, port), allow_none=True)
# Setup an event receiving queue
eventHandler = uievent.BBUIEventQueue(server)
# Launch the UI
if configuration.ui:
ui = configuration.ui
else:
ui = "knotty"
try:
# Dynamically load the UI based on the ui name. Although we
# suggest a fixed set this allows you to have flexibility in which
# ones are available.
exec "from bb.ui import " + ui
exec "return_value = " + ui + ".init(server, eventHandler)"
except ImportError:
print "FATAL: Invalid user interface '%s' specified. " % ui
print "Valid interfaces are 'ncurses', 'depexp' or the default, 'knotty'."
except Exception, e:
print "FATAL: Unable to start to '%s' UI: %s." % (configuration.ui, e.message)
finally:
# Don't wait for server indefinitely
import socket
socket.setdefaulttimeout(2)
try:
eventHandler.system_quit()
except:
pass
try:
server.terminateServer()
except:
pass
return return_value
if __name__ == "__main__":
print """WARNING, WARNING, WARNING
This is a Bitbake from the Unstable/Development 1.9 Branch. This software is a work in progress and should only be used by Bitbake developers/testers"""
ret = main()
sys.exit(ret)

534
bitbake-dev/bin/bitdoc Executable file
View File

@@ -0,0 +1,534 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2005 Holger Hans Peter Freyther
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import optparse, os, sys
# bitbake
sys.path.append(os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb
import bb.parse
from string import split, join
__version__ = "0.0.2"
class HTMLFormatter:
"""
Simple class to help to generate some sort of HTML files. It is
quite inferior solution compared to docbook, gtkdoc, doxygen but it
should work for now.
We've a global introduction site (index.html) and then one site for
the list of keys (alphabetical sorted) and one for the list of groups,
one site for each key with links to the relations and groups.
index.html
all_keys.html
all_groups.html
groupNAME.html
keyNAME.html
"""
def replace(self, text, *pairs):
"""
From pydoc... almost identical at least
"""
while pairs:
(a,b) = pairs[0]
text = join(split(text, a), b)
pairs = pairs[1:]
return text
def escape(self, text):
"""
Escape string to be conform HTML
"""
return self.replace(text,
('&', '&amp;'),
('<', '&lt;' ),
('>', '&gt;' ) )
def createNavigator(self):
"""
Create the navgiator
"""
return """<table class="navigation" width="100%" summary="Navigation header" cellpadding="2" cellspacing="2">
<tr valign="middle">
<td><a accesskey="g" href="index.html">Home</a></td>
<td><a accesskey="n" href="all_groups.html">Groups</a></td>
<td><a accesskey="u" href="all_keys.html">Keys</a></td>
</tr></table>
"""
def relatedKeys(self, item):
"""
Create HTML to link to foreign keys
"""
if len(item.related()) == 0:
return ""
txt = "<p><b>See also:</b><br>"
txts = []
for it in item.related():
txts.append("""<a href="key%(it)s.html">%(it)s</a>""" % vars() )
return txt + ",".join(txts)
def groups(self,item):
"""
Create HTML to link to related groups
"""
if len(item.groups()) == 0:
return ""
txt = "<p><b>See also:</b><br>"
txts = []
for group in item.groups():
txts.append( """<a href="group%s.html">%s</a> """ % (group,group) )
return txt + ",".join(txts)
def createKeySite(self,item):
"""
Create a site for a key. It contains the header/navigator, a heading,
the description, links to related keys and to the groups.
"""
return """<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head><title>Key %s</title></head>
<link rel="stylesheet" href="style.css" type="text/css">
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
%s
<h2><span class="refentrytitle">%s</span></h2>
<div class="refsynopsisdiv">
<h2>Synopsis</h2>
<p>
%s
</p>
</div>
<div class="refsynopsisdiv">
<h2>Related Keys</h2>
<p>
%s
</p>
</div>
<div class="refsynopsisdiv">
<h2>Groups</h2>
<p>
%s
</p>
</div>
</body>
""" % (item.name(), self.createNavigator(), item.name(),
self.escape(item.description()), self.relatedKeys(item), self.groups(item))
def createGroupsSite(self, doc):
"""
Create the Group Overview site
"""
groups = ""
sorted_groups = doc.groups()
sorted_groups.sort()
for group in sorted_groups:
groups += """<a href="group%s.html">%s</a><br>""" % (group, group)
return """<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head><title>Group overview</title></head>
<link rel="stylesheet" href="style.css" type="text/css">
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
%s
<h2>Available Groups</h2>
%s
</body>
""" % (self.createNavigator(), groups)
def createIndex(self):
"""
Create the index file
"""
return """<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head><title>Bitbake Documentation</title></head>
<link rel="stylesheet" href="style.css" type="text/css">
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
%s
<h2>Documentation Entrance</h2>
<a href="all_groups.html">All available groups</a><br>
<a href="all_keys.html">All available keys</a><br>
</body>
""" % self.createNavigator()
def createKeysSite(self, doc):
"""
Create Overview of all avilable keys
"""
keys = ""
sorted_keys = doc.doc_keys()
sorted_keys.sort()
for key in sorted_keys:
keys += """<a href="key%s.html">%s</a><br>""" % (key, key)
return """<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head><title>Key overview</title></head>
<link rel="stylesheet" href="style.css" type="text/css">
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
%s
<h2>Available Keys</h2>
%s
</body>
""" % (self.createNavigator(), keys)
def createGroupSite(self, gr, items, _description = None):
"""
Create a site for a group:
Group the name of the group, items contain the name of the keys
inside this group
"""
groups = ""
description = ""
# create a section with the group descriptions
if _description:
description += "<h2 Description of Grozp %s</h2>" % gr
description += _description
items.sort(lambda x,y:cmp(x.name(),y.name()))
for group in items:
groups += """<a href="key%s.html">%s</a><br>""" % (group.name(), group.name())
return """<!doctype html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html><head><title>Group %s</title></head>
<link rel="stylesheet" href="style.css" type="text/css">
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF">
%s
%s
<div class="refsynopsisdiv">
<h2>Keys in Group %s</h2>
<pre class="synopsis">
%s
</pre>
</div>
</body>
""" % (gr, self.createNavigator(), description, gr, groups)
def createCSS(self):
"""
Create the CSS file
"""
return """.synopsis, .classsynopsis
{
background: #eeeeee;
border: solid 1px #aaaaaa;
padding: 0.5em;
}
.programlisting
{
background: #eeeeff;
border: solid 1px #aaaaff;
padding: 0.5em;
}
.variablelist
{
padding: 4px;
margin-left: 3em;
}
.variablelist td:first-child
{
vertical-align: top;
}
table.navigation
{
background: #ffeeee;
border: solid 1px #ffaaaa;
margin-top: 0.5em;
margin-bottom: 0.5em;
}
.navigation a
{
color: #770000;
}
.navigation a:visited
{
color: #550000;
}
.navigation .title
{
font-size: 200%;
}
div.refnamediv
{
margin-top: 2em;
}
div.gallery-float
{
float: left;
padding: 10px;
}
div.gallery-float img
{
border-style: none;
}
div.gallery-spacer
{
clear: both;
}
a
{
text-decoration: none;
}
a:hover
{
text-decoration: underline;
color: #FF0000;
}
"""
class DocumentationItem:
"""
A class to hold information about a configuration
item. It contains the key name, description, a list of related names,
and the group this item is contained in.
"""
def __init__(self):
self._groups = []
self._related = []
self._name = ""
self._desc = ""
def groups(self):
return self._groups
def name(self):
return self._name
def description(self):
return self._desc
def related(self):
return self._related
def setName(self, name):
self._name = name
def setDescription(self, desc):
self._desc = desc
def addGroup(self, group):
self._groups.append(group)
def addRelation(self,relation):
self._related.append(relation)
def sort(self):
self._related.sort()
self._groups.sort()
class Documentation:
"""
Holds the documentation... with mappings from key to items...
"""
def __init__(self):
self.__keys = {}
self.__groups = {}
def insert_doc_item(self, item):
"""
Insert the Doc Item into the internal list
of representation
"""
item.sort()
self.__keys[item.name()] = item
for group in item.groups():
if not group in self.__groups:
self.__groups[group] = []
self.__groups[group].append(item)
self.__groups[group].sort()
def doc_item(self, key):
"""
Return the DocumentationInstance describing the key
"""
try:
return self.__keys[key]
except KeyError:
return None
def doc_keys(self):
"""
Return the documented KEYS (names)
"""
return self.__keys.keys()
def groups(self):
"""
Return the names of available groups
"""
return self.__groups.keys()
def group_content(self,group_name):
"""
Return a list of keys/names that are in a specefic
group or the empty list
"""
try:
return self.__groups[group_name]
except KeyError:
return []
def parse_cmdline(args):
"""
Parse the CMD line and return the result as a n-tuple
"""
parser = optparse.OptionParser( version = "Bitbake Documentation Tool Core version %s, %%prog version %s" % (bb.__version__,__version__))
usage = """%prog [options]
Create a set of html pages (documentation) for a bitbake.conf....
"""
# Add the needed options
parser.add_option( "-c", "--config", help = "Use the specified configuration file as source",
action = "store", dest = "config", default = os.path.join("conf", "documentation.conf") )
parser.add_option( "-o", "--output", help = "Output directory for html files",
action = "store", dest = "output", default = "html/" )
parser.add_option( "-D", "--debug", help = "Increase the debug level",
action = "count", dest = "debug", default = 0 )
parser.add_option( "-v","--verbose", help = "output more chit-char to the terminal",
action = "store_true", dest = "verbose", default = False )
options, args = parser.parse_args( sys.argv )
if options.debug:
bb.msg.set_debug_level(options.debug)
return options.config, options.output
def main():
"""
The main Method
"""
(config_file,output_dir) = parse_cmdline( sys.argv )
# right to let us load the file now
try:
documentation = bb.parse.handle( config_file, bb.data.init() )
except IOError:
bb.fatal( "Unable to open %s" % config_file )
except bb.parse.ParseError:
bb.fatal( "Unable to parse %s" % config_file )
if isinstance(documentation, dict):
documentation = documentation[""]
# Assuming we've the file loaded now, we will initialize the 'tree'
doc = Documentation()
# defined states
state_begin = 0
state_see = 1
state_group = 2
for key in bb.data.keys(documentation):
data = bb.data.getVarFlag(key, "doc", documentation)
if not data:
continue
# The Documentation now starts
doc_ins = DocumentationItem()
doc_ins.setName(key)
tokens = data.split(' ')
state = state_begin
string= ""
for token in tokens:
token = token.strip(',')
if not state == state_see and token == "@see":
state = state_see
continue
elif not state == state_group and token == "@group":
state = state_group
continue
if state == state_begin:
string += " %s" % token
elif state == state_see:
doc_ins.addRelation(token)
elif state == state_group:
doc_ins.addGroup(token)
# set the description
doc_ins.setDescription(string)
doc.insert_doc_item(doc_ins)
# let us create the HTML now
bb.mkdirhier(output_dir)
os.chdir(output_dir)
# Let us create the sites now. We do it in the following order
# Start with the index.html. It will point to sites explaining all
# keys and groups
html_slave = HTMLFormatter()
f = file('style.css', 'w')
print >> f, html_slave.createCSS()
f = file('index.html', 'w')
print >> f, html_slave.createIndex()
f = file('all_groups.html', 'w')
print >> f, html_slave.createGroupsSite(doc)
f = file('all_keys.html', 'w')
print >> f, html_slave.createKeysSite(doc)
# now for each group create the site
for group in doc.groups():
f = file('group%s.html' % group, 'w')
print >> f, html_slave.createGroupSite(group, doc.group_content(group))
# now for the keys
for key in doc.doc_keys():
f = file('key%s.html' % doc.doc_item(key).name(), 'w')
print >> f, html_slave.createKeySite(doc.doc_item(key))
if __name__ == "__main__":
main()

318
bitbake-dev/lib/bb/COW.py Normal file
View File

@@ -0,0 +1,318 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# This is a copy on write dictionary and set which abuses classes to try and be nice and fast.
#
# Copyright (C) 2006 Tim Amsell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#Please Note:
# Be careful when using mutable types (ie Dict and Lists) - operations involving these are SLOW.
# Assign a file to __warn__ to get warnings about slow operations.
#
import copy
import types
types.ImmutableTypes = tuple([ \
types.BooleanType, \
types.ComplexType, \
types.FloatType, \
types.IntType, \
types.LongType, \
types.NoneType, \
types.TupleType, \
frozenset] + \
list(types.StringTypes))
MUTABLE = "__mutable__"
class COWMeta(type):
pass
class COWDictMeta(COWMeta):
__warn__ = False
__hasmutable__ = False
__marker__ = tuple()
def __str__(cls):
# FIXME: I have magic numbers!
return "<COWDict Level: %i Current Keys: %i>" % (cls.__count__, len(cls.__dict__) - 3)
__repr__ = __str__
def cow(cls):
class C(cls):
__count__ = cls.__count__ + 1
return C
copy = cow
__call__ = cow
def __setitem__(cls, key, value):
if not isinstance(value, types.ImmutableTypes):
if not isinstance(value, COWMeta):
cls.__hasmutable__ = True
key += MUTABLE
setattr(cls, key, value)
def __getmutable__(cls, key, readonly=False):
nkey = key + MUTABLE
try:
return cls.__dict__[nkey]
except KeyError:
pass
value = getattr(cls, nkey)
if readonly:
return value
if not cls.__warn__ is False and not isinstance(value, COWMeta):
print >> cls.__warn__, "Warning: Doing a copy because %s is a mutable type." % key
try:
value = value.copy()
except AttributeError, e:
value = copy.copy(value)
setattr(cls, nkey, value)
return value
__getmarker__ = []
def __getreadonly__(cls, key, default=__getmarker__):
"""\
Get a value (even if mutable) which you promise not to change.
"""
return cls.__getitem__(key, default, True)
def __getitem__(cls, key, default=__getmarker__, readonly=False):
try:
try:
value = getattr(cls, key)
except AttributeError:
value = cls.__getmutable__(key, readonly)
# This is for values which have been deleted
if value is cls.__marker__:
raise AttributeError("key %s does not exist." % key)
return value
except AttributeError, e:
if not default is cls.__getmarker__:
return default
raise KeyError(str(e))
def __delitem__(cls, key):
cls.__setitem__(key, cls.__marker__)
def __revertitem__(cls, key):
if not cls.__dict__.has_key(key):
key += MUTABLE
delattr(cls, key)
def has_key(cls, key):
value = cls.__getreadonly__(key, cls.__marker__)
if value is cls.__marker__:
return False
return True
def iter(cls, type, readonly=False):
for key in dir(cls):
if key.startswith("__"):
continue
if key.endswith(MUTABLE):
key = key[:-len(MUTABLE)]
if type == "keys":
yield key
try:
if readonly:
value = cls.__getreadonly__(key)
else:
value = cls[key]
except KeyError:
continue
if type == "values":
yield value
if type == "items":
yield (key, value)
raise StopIteration()
def iterkeys(cls):
return cls.iter("keys")
def itervalues(cls, readonly=False):
if not cls.__warn__ is False and cls.__hasmutable__ and readonly is False:
print >> cls.__warn__, "Warning: If you arn't going to change any of the values call with True."
return cls.iter("values", readonly)
def iteritems(cls, readonly=False):
if not cls.__warn__ is False and cls.__hasmutable__ and readonly is False:
print >> cls.__warn__, "Warning: If you arn't going to change any of the values call with True."
return cls.iter("items", readonly)
class COWSetMeta(COWDictMeta):
def __str__(cls):
# FIXME: I have magic numbers!
return "<COWSet Level: %i Current Keys: %i>" % (cls.__count__, len(cls.__dict__) -3)
__repr__ = __str__
def cow(cls):
class C(cls):
__count__ = cls.__count__ + 1
return C
def add(cls, value):
COWDictMeta.__setitem__(cls, repr(hash(value)), value)
def remove(cls, value):
COWDictMeta.__delitem__(cls, repr(hash(value)))
def __in__(cls, value):
return COWDictMeta.has_key(repr(hash(value)))
def iterkeys(cls):
raise TypeError("sets don't have keys")
def iteritems(cls):
raise TypeError("sets don't have 'items'")
# These are the actual classes you use!
class COWDictBase(object):
__metaclass__ = COWDictMeta
__count__ = 0
class COWSetBase(object):
__metaclass__ = COWSetMeta
__count__ = 0
if __name__ == "__main__":
import sys
COWDictBase.__warn__ = sys.stderr
a = COWDictBase()
print "a", a
a['a'] = 'a'
a['b'] = 'b'
a['dict'] = {}
b = a.copy()
print "b", b
b['c'] = 'b'
print
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems():
print x
print
b['dict']['a'] = 'b'
b['a'] = 'c'
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems():
print x
print
try:
b['dict2']
except KeyError, e:
print "Okay!"
a['set'] = COWSetBase()
a['set'].add("o1")
a['set'].add("o1")
a['set'].add("o2")
print "a", a
for x in a['set'].itervalues():
print x
print "--"
print "b", b
for x in b['set'].itervalues():
print x
print
b['set'].add('o3')
print "a", a
for x in a['set'].itervalues():
print x
print "--"
print "b", b
for x in b['set'].itervalues():
print x
print
a['set2'] = set()
a['set2'].add("o1")
a['set2'].add("o1")
a['set2'].add("o2")
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print x
print
del b['b']
try:
print b['b']
except KeyError:
print "Yay! deleted key raises error"
if b.has_key('b'):
print "Boo!"
else:
print "Yay - has_key with delete works!"
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print x
print
b.__revertitem__('b')
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print x
print
b.__revertitem__('dict')
print "a", a
for x in a.iteritems():
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print x
print

File diff suppressed because it is too large Load Diff

394
bitbake-dev/lib/bb/build.py Normal file
View File

@@ -0,0 +1,394 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# BitBake 'Build' implementation
#
# Core code for function execution and task handling in the
# BitBake build tools.
#
# Copyright (C) 2003, 2004 Chris Larson
#
# Based on Gentoo's portage.py.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#Based on functions from the base bb module, Copyright 2003 Holger Schurig
from bb import data, event, mkdirhier, utils
import bb, os, sys
# When we execute a python function we'd like certain things
# in all namespaces, hence we add them to __builtins__
# If we do not do this and use the exec globals, they will
# not be available to subfunctions.
__builtins__['bb'] = bb
__builtins__['os'] = os
# events
class FuncFailed(Exception):
"""
Executed function failed
First parameter a message
Second paramter is a logfile (optional)
"""
class EventException(Exception):
"""Exception which is associated with an Event."""
def __init__(self, msg, event):
self.args = msg, event
class TaskBase(event.Event):
"""Base class for task events"""
def __init__(self, t, d ):
self._task = t
self._package = bb.data.getVar("PF", d, 1)
event.Event.__init__(self, d)
self._message = "package %s: task %s: %s" % (bb.data.getVar("PF", d, 1), t, bb.event.getName(self)[4:])
def getTask(self):
return self._task
def setTask(self, task):
self._task = task
task = property(getTask, setTask, None, "task property")
class TaskStarted(TaskBase):
"""Task execution started"""
class TaskSucceeded(TaskBase):
"""Task execution completed"""
class TaskFailed(TaskBase):
"""Task execution failed"""
def __init__(self, msg, logfile, t, d ):
self.logfile = logfile
self.msg = msg
TaskBase.__init__(self, t, d)
class InvalidTask(TaskBase):
"""Invalid Task"""
# functions
def exec_func(func, d, dirs = None):
"""Execute an BB 'function'"""
body = data.getVar(func, d)
if not body:
return
flags = data.getVarFlags(func, d)
for item in ['deps', 'check', 'interactive', 'python', 'cleandirs', 'dirs', 'lockfiles', 'fakeroot']:
if not item in flags:
flags[item] = None
ispython = flags['python']
cleandirs = (data.expand(flags['cleandirs'], d) or "").split()
for cdir in cleandirs:
os.system("rm -rf %s" % cdir)
if dirs:
dirs = data.expand(dirs, d)
else:
dirs = (data.expand(flags['dirs'], d) or "").split()
for adir in dirs:
mkdirhier(adir)
if len(dirs) > 0:
adir = dirs[-1]
else:
adir = data.getVar('B', d, 1)
# Save current directory
try:
prevdir = os.getcwd()
except OSError:
prevdir = data.getVar('TOPDIR', d, True)
# Setup logfiles
t = data.getVar('T', d, 1)
if not t:
bb.msg.fatal(bb.msg.domain.Build, "T not set")
mkdirhier(t)
# Gross hack, FIXME
import random
logfile = "%s/log.%s.%s.%s" % (t, func, str(os.getpid()),random.random())
runfile = "%s/run.%s.%s" % (t, func, str(os.getpid()))
# Change to correct directory (if specified)
if adir and os.access(adir, os.F_OK):
os.chdir(adir)
# Handle logfiles
si = file('/dev/null', 'r')
try:
if bb.msg.debug_level['default'] > 0 or ispython:
so = os.popen("tee \"%s\"" % logfile, "w")
else:
so = file(logfile, 'w')
except OSError, e:
bb.msg.error(bb.msg.domain.Build, "opening log file: %s" % e)
pass
se = so
# Dup the existing fds so we dont lose them
osi = [os.dup(sys.stdin.fileno()), sys.stdin.fileno()]
oso = [os.dup(sys.stdout.fileno()), sys.stdout.fileno()]
ose = [os.dup(sys.stderr.fileno()), sys.stderr.fileno()]
# Replace those fds with our own
os.dup2(si.fileno(), osi[1])
os.dup2(so.fileno(), oso[1])
os.dup2(se.fileno(), ose[1])
locks = []
lockfiles = (data.expand(flags['lockfiles'], d) or "").split()
for lock in lockfiles:
locks.append(bb.utils.lockfile(lock))
try:
# Run the function
if ispython:
exec_func_python(func, d, runfile, logfile)
else:
exec_func_shell(func, d, runfile, logfile, flags)
# Restore original directory
try:
os.chdir(prevdir)
except:
pass
finally:
# Unlock any lockfiles
for lock in locks:
bb.utils.unlockfile(lock)
# Restore the backup fds
os.dup2(osi[0], osi[1])
os.dup2(oso[0], oso[1])
os.dup2(ose[0], ose[1])
# Close our logs
si.close()
so.close()
se.close()
if os.path.exists(logfile) and os.path.getsize(logfile) == 0:
bb.msg.debug(2, bb.msg.domain.Build, "Zero size logfile %s, removing" % logfile)
os.remove(logfile)
# Close the backup fds
os.close(osi[0])
os.close(oso[0])
os.close(ose[0])
def exec_func_python(func, d, runfile, logfile):
"""Execute a python BB 'function'"""
import re, os
bbfile = bb.data.getVar('FILE', d, 1)
tmp = "def " + func + "():\n%s" % data.getVar(func, d)
tmp += '\n' + func + '()'
f = open(runfile, "w")
f.write(tmp)
comp = utils.better_compile(tmp, func, bbfile)
g = {} # globals
g['d'] = d
try:
utils.better_exec(comp, g, tmp, bbfile)
except:
(t,value,tb) = sys.exc_info()
if t in [bb.parse.SkipPackage, bb.build.FuncFailed]:
raise
bb.msg.error(bb.msg.domain.Build, "Function %s failed" % func)
raise FuncFailed("function %s failed" % func, logfile)
def exec_func_shell(func, d, runfile, logfile, flags):
"""Execute a shell BB 'function' Returns true if execution was successful.
For this, it creates a bash shell script in the tmp dectory, writes the local
data into it and finally executes. The output of the shell will end in a log file and stdout.
Note on directory behavior. The 'dirs' varflag should contain a list
of the directories you need created prior to execution. The last
item in the list is where we will chdir/cd to.
"""
deps = flags['deps']
check = flags['check']
if check in globals():
if globals()[check](func, deps):
return
f = open(runfile, "w")
f.write("#!/bin/sh -e\n")
if bb.msg.debug_level['default'] > 0: f.write("set -x\n")
data.emit_env(f, d)
f.write("cd %s\n" % os.getcwd())
if func: f.write("%s\n" % func)
f.close()
os.chmod(runfile, 0775)
if not func:
bb.msg.error(bb.msg.domain.Build, "Function not specified")
raise FuncFailed("Function not specified for exec_func_shell")
# execute function
if flags['fakeroot']:
maybe_fakeroot = "PATH=\"%s\" fakeroot " % bb.data.getVar("PATH", d, 1)
else:
maybe_fakeroot = ''
lang_environment = "LC_ALL=C "
ret = os.system('%s%ssh -e %s' % (lang_environment, maybe_fakeroot, runfile))
if ret == 0:
return
bb.msg.error(bb.msg.domain.Build, "Function %s failed" % func)
raise FuncFailed("function %s failed" % func, logfile)
def exec_task(task, d):
"""Execute an BB 'task'
The primary difference between executing a task versus executing
a function is that a task exists in the task digraph, and therefore
has dependencies amongst other tasks."""
# Check whther this is a valid task
if not data.getVarFlag(task, 'task', d):
raise EventException("No such task", InvalidTask(task, d))
try:
bb.msg.debug(1, bb.msg.domain.Build, "Executing task %s" % task)
old_overrides = data.getVar('OVERRIDES', d, 0)
localdata = data.createCopy(d)
data.setVar('OVERRIDES', 'task-%s:%s' % (task[3:], old_overrides), localdata)
data.update_data(localdata)
data.expandKeys(localdata)
event.fire(TaskStarted(task, localdata))
exec_func(task, localdata)
event.fire(TaskSucceeded(task, localdata))
except FuncFailed, message:
# Try to extract the optional logfile
try:
(msg, logfile) = message
except:
logfile = None
msg = message
bb.msg.note(1, bb.msg.domain.Build, "Task failed: %s" % message )
failedevent = TaskFailed(msg, logfile, task, d)
event.fire(failedevent)
raise EventException("Function failed in task: %s" % message, failedevent)
# make stamp, or cause event and raise exception
if not data.getVarFlag(task, 'nostamp', d) and not data.getVarFlag(task, 'selfstamp', d):
make_stamp(task, d)
def extract_stamp(d, fn):
"""
Extracts stamp format which is either a data dictonary (fn unset)
or a dataCache entry (fn set).
"""
if fn:
return d.stamp[fn]
return data.getVar('STAMP', d, 1)
def stamp_internal(task, d, file_name):
"""
Internal stamp helper function
Removes any stamp for the given task
Makes sure the stamp directory exists
Returns the stamp path+filename
"""
stamp = extract_stamp(d, file_name)
if not stamp:
return
stamp = "%s.%s" % (stamp, task)
mkdirhier(os.path.dirname(stamp))
# Remove the file and recreate to force timestamp
# change on broken NFS filesystems
if os.access(stamp, os.F_OK):
os.remove(stamp)
return stamp
def make_stamp(task, d, file_name = None):
"""
Creates/updates a stamp for a given task
(d can be a data dict or dataCache)
"""
stamp = stamp_internal(task, d, file_name)
if stamp:
f = open(stamp, "w")
f.close()
def del_stamp(task, d, file_name = None):
"""
Removes a stamp for a given task
(d can be a data dict or dataCache)
"""
stamp_internal(task, d, file_name)
def add_tasks(tasklist, d):
task_deps = data.getVar('_task_deps', d)
if not task_deps:
task_deps = {}
if not 'tasks' in task_deps:
task_deps['tasks'] = []
if not 'parents' in task_deps:
task_deps['parents'] = {}
for task in tasklist:
task = data.expand(task, d)
data.setVarFlag(task, 'task', 1, d)
if not task in task_deps['tasks']:
task_deps['tasks'].append(task)
flags = data.getVarFlags(task, d)
def getTask(name):
if not name in task_deps:
task_deps[name] = {}
if name in flags:
deptask = data.expand(flags[name], d)
task_deps[name][task] = deptask
getTask('depends')
getTask('deptask')
getTask('rdeptask')
getTask('recrdeptask')
getTask('nostamp')
task_deps['parents'][task] = []
for dep in flags['deps']:
dep = data.expand(dep, d)
task_deps['parents'][task].append(dep)
# don't assume holding a reference
data.setVar('_task_deps', task_deps, d)
def remove_task(task, kill, d):
"""Remove an BB 'task'.
If kill is 1, also remove tasks that depend on this task."""
data.delVarFlag(task, 'task', d)

534
bitbake-dev/lib/bb/cache.py Normal file
View File

@@ -0,0 +1,534 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# BitBake 'Event' implementation
#
# Caching of bitbake variables before task execution
# Copyright (C) 2006 Richard Purdie
# but small sections based on code from bin/bitbake:
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
# Copyright (C) 2003 - 2005 Michael 'Mickey' Lauer
# Copyright (C) 2005 Holger Hans Peter Freyther
# Copyright (C) 2005 ROAD GmbH
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os, re
import bb.data
import bb.utils
try:
import cPickle as pickle
except ImportError:
import pickle
bb.msg.note(1, bb.msg.domain.Cache, "Importing cPickle failed. Falling back to a very slow implementation.")
__cache_version__ = "130"
class Cache:
"""
BitBake Cache implementation
"""
def __init__(self, cooker):
self.cachedir = bb.data.getVar("CACHE", cooker.configuration.data, True)
self.clean = {}
self.checked = {}
self.depends_cache = {}
self.data = None
self.data_fn = None
self.cacheclean = True
if self.cachedir in [None, '']:
self.has_cache = False
bb.msg.note(1, bb.msg.domain.Cache, "Not using a cache. Set CACHE = <directory> to enable.")
return
self.has_cache = True
self.cachefile = os.path.join(self.cachedir,"bb_cache.dat")
bb.msg.debug(1, bb.msg.domain.Cache, "Using cache in '%s'" % self.cachedir)
try:
os.stat( self.cachedir )
except OSError:
bb.mkdirhier( self.cachedir )
# If any of configuration.data's dependencies are newer than the
# cache there isn't even any point in loading it...
newest_mtime = 0
deps = bb.data.getVar("__depends", cooker.configuration.data, True)
for f,old_mtime in deps:
if old_mtime > newest_mtime:
newest_mtime = old_mtime
if bb.parse.cached_mtime_noerror(self.cachefile) >= newest_mtime:
try:
p = pickle.Unpickler(file(self.cachefile, "rb"))
self.depends_cache, version_data = p.load()
if version_data['CACHE_VER'] != __cache_version__:
raise ValueError, 'Cache Version Mismatch'
if version_data['BITBAKE_VER'] != bb.__version__:
raise ValueError, 'Bitbake Version Mismatch'
except EOFError:
bb.msg.note(1, bb.msg.domain.Cache, "Truncated cache found, rebuilding...")
self.depends_cache = {}
except:
bb.msg.note(1, bb.msg.domain.Cache, "Invalid cache found, rebuilding...")
self.depends_cache = {}
else:
try:
os.stat( self.cachefile )
bb.msg.note(1, bb.msg.domain.Cache, "Out of date cache found, rebuilding...")
except OSError:
pass
def getVar(self, var, fn, exp = 0):
"""
Gets the value of a variable
(similar to getVar in the data class)
There are two scenarios:
1. We have cached data - serve from depends_cache[fn]
2. We're learning what data to cache - serve from data
backend but add a copy of the data to the cache.
"""
if fn in self.clean:
return self.depends_cache[fn][var]
if not fn in self.depends_cache:
self.depends_cache[fn] = {}
if fn != self.data_fn:
# We're trying to access data in the cache which doesn't exist
# yet setData hasn't been called to setup the right access. Very bad.
bb.msg.error(bb.msg.domain.Cache, "Parsing error data_fn %s and fn %s don't match" % (self.data_fn, fn))
self.cacheclean = False
result = bb.data.getVar(var, self.data, exp)
self.depends_cache[fn][var] = result
return result
def setData(self, virtualfn, fn, data):
"""
Called to prime bb_cache ready to learn which variables to cache.
Will be followed by calls to self.getVar which aren't cached
but can be fulfilled from self.data.
"""
self.data_fn = virtualfn
self.data = data
# Make sure __depends makes the depends_cache
self.getVar("__depends", virtualfn, True)
self.depends_cache[virtualfn]["CACHETIMESTAMP"] = bb.parse.cached_mtime(fn)
def virtualfn2realfn(self, virtualfn):
"""
Convert a virtual file name to a real one + the associated subclass keyword
"""
fn = virtualfn
cls = ""
if virtualfn.startswith('virtual:'):
cls = virtualfn.split(':', 2)[1]
fn = virtualfn.replace('virtual:' + cls + ':', '')
#bb.msg.debug(2, bb.msg.domain.Cache, "virtualfn2realfn %s to %s %s" % (virtualfn, fn, cls))
return (fn, cls)
def realfn2virtual(self, realfn, cls):
"""
Convert a real filename + the associated subclass keyword to a virtual filename
"""
if cls == "":
#bb.msg.debug(2, bb.msg.domain.Cache, "realfn2virtual %s and '%s' to %s" % (realfn, cls, realfn))
return realfn
#bb.msg.debug(2, bb.msg.domain.Cache, "realfn2virtual %s and %s to %s" % (realfn, cls, "virtual:" + cls + ":" + realfn))
return "virtual:" + cls + ":" + realfn
def loadDataFull(self, virtualfn, cfgData):
"""
Return a complete set of data for fn.
To do this, we need to parse the file.
"""
(fn, cls) = self.virtualfn2realfn(virtualfn)
bb.msg.debug(1, bb.msg.domain.Cache, "Parsing %s (full)" % fn)
bb_data, skipped = self.load_bbfile(fn, cfgData)
if isinstance(bb_data, dict):
return bb_data[cls]
return bb_data
def loadData(self, fn, cfgData, cacheData):
"""
Load a subset of data for fn.
If the cached data is valid we do nothing,
To do this, we need to parse the file and set the system
to record the variables accessed.
Return the cache status and whether the file was skipped when parsed
"""
if fn not in self.checked:
self.cacheValidUpdate(fn)
if self.cacheValid(fn):
if "SKIPPED" in self.depends_cache[fn]:
return True, True
self.handle_data(fn, cacheData)
multi = self.getVar('BBCLASSEXTEND', fn, True)
if multi:
for cls in multi.split():
virtualfn = self.realfn2virtual(fn, cls)
# Pretend we're clean so getVar works
self.clean[virtualfn] = ""
self.handle_data(virtualfn, cacheData)
return True, False
bb.msg.debug(1, bb.msg.domain.Cache, "Parsing %s" % fn)
bb_data, skipped = self.load_bbfile(fn, cfgData)
if skipped:
if isinstance(bb_data, dict):
self.setData(fn, fn, bb_data[""])
else:
self.setData(fn, fn, bb_data)
return False, skipped
if isinstance(bb_data, dict):
for data in bb_data:
virtualfn = self.realfn2virtual(fn, data)
self.setData(virtualfn, fn, bb_data[data])
self.handle_data(virtualfn, cacheData)
return False, skipped
self.setData(fn, fn, bb_data)
self.handle_data(fn, cacheData)
return False, skipped
def cacheValid(self, fn):
"""
Is the cache valid for fn?
Fast version, no timestamps checked.
"""
# Is cache enabled?
if not self.has_cache:
return False
if fn in self.clean:
return True
return False
def cacheValidUpdate(self, fn):
"""
Is the cache valid for fn?
Make thorough (slower) checks including timestamps.
"""
# Is cache enabled?
if not self.has_cache:
return False
self.checked[fn] = ""
# Pretend we're clean so getVar works
self.clean[fn] = ""
# File isn't in depends_cache
if not fn in self.depends_cache:
bb.msg.debug(2, bb.msg.domain.Cache, "Cache: %s is not cached" % fn)
self.remove(fn)
return False
mtime = bb.parse.cached_mtime_noerror(fn)
# Check file still exists
if mtime == 0:
bb.msg.debug(2, bb.msg.domain.Cache, "Cache: %s not longer exists" % fn)
self.remove(fn)
return False
# Check the file's timestamp
if mtime != self.getVar("CACHETIMESTAMP", fn, True):
bb.msg.debug(2, bb.msg.domain.Cache, "Cache: %s changed" % fn)
self.remove(fn)
return False
# Check dependencies are still valid
depends = self.getVar("__depends", fn, True)
if depends:
for f,old_mtime in depends:
fmtime = bb.parse.cached_mtime_noerror(f)
# Check if file still exists
if old_mtime != 0 and fmtime == 0:
self.remove(fn)
return False
if (fmtime != old_mtime):
bb.msg.debug(2, bb.msg.domain.Cache, "Cache: %s's dependency %s changed" % (fn, f))
self.remove(fn)
return False
#bb.msg.debug(2, bb.msg.domain.Cache, "Depends Cache: %s is clean" % fn)
if not fn in self.clean:
self.clean[fn] = ""
return True
def skip(self, fn):
"""
Mark a fn as skipped
Called from the parser
"""
if not fn in self.depends_cache:
self.depends_cache[fn] = {}
self.depends_cache[fn]["SKIPPED"] = "1"
def remove(self, fn):
"""
Remove a fn from the cache
Called from the parser in error cases
"""
bb.msg.debug(1, bb.msg.domain.Cache, "Removing %s from cache" % fn)
if fn in self.depends_cache:
del self.depends_cache[fn]
if fn in self.clean:
del self.clean[fn]
def sync(self):
"""
Save the cache
Called from the parser when complete (or exiting)
"""
import copy
if not self.has_cache:
return
if self.cacheclean:
bb.msg.note(1, bb.msg.domain.Cache, "Cache is clean, not saving.")
return
version_data = {}
version_data['CACHE_VER'] = __cache_version__
version_data['BITBAKE_VER'] = bb.__version__
cache_data = copy.deepcopy(self.depends_cache)
for fn in self.depends_cache:
if '__BB_DONT_CACHE' in self.depends_cache[fn] and self.depends_cache[fn]['__BB_DONT_CACHE']:
bb.msg.debug(2, bb.msg.domain.Cache, "Not caching %s, marked as not cacheable" % fn)
del cache_data[fn]
elif 'PV' in self.depends_cache[fn] and 'SRCREVINACTION' in self.depends_cache[fn]['PV']:
bb.msg.error(bb.msg.domain.Cache, "Not caching %s as it had SRCREVINACTION in PV. Please report this bug" % fn)
del cache_data[fn]
p = pickle.Pickler(file(self.cachefile, "wb" ), -1 )
p.dump([cache_data, version_data])
def mtime(self, cachefile):
return bb.parse.cached_mtime_noerror(cachefile)
def handle_data(self, file_name, cacheData):
"""
Save data we need into the cache
"""
pn = self.getVar('PN', file_name, True)
pe = self.getVar('PE', file_name, True) or "0"
pv = self.getVar('PV', file_name, True)
if 'SRCREVINACTION' in pv:
bb.note("Found SRCREVINACTION in PV (%s) or %s. Please report this bug." % (pv, file_name))
pr = self.getVar('PR', file_name, True)
dp = int(self.getVar('DEFAULT_PREFERENCE', file_name, True) or "0")
depends = bb.utils.explode_deps(self.getVar("DEPENDS", file_name, True) or "")
packages = (self.getVar('PACKAGES', file_name, True) or "").split()
packages_dynamic = (self.getVar('PACKAGES_DYNAMIC', file_name, True) or "").split()
rprovides = (self.getVar("RPROVIDES", file_name, True) or "").split()
cacheData.task_deps[file_name] = self.getVar("_task_deps", file_name, True)
# build PackageName to FileName lookup table
if pn not in cacheData.pkg_pn:
cacheData.pkg_pn[pn] = []
cacheData.pkg_pn[pn].append(file_name)
cacheData.stamp[file_name] = self.getVar('STAMP', file_name, True)
# build FileName to PackageName lookup table
cacheData.pkg_fn[file_name] = pn
cacheData.pkg_pepvpr[file_name] = (pe,pv,pr)
cacheData.pkg_dp[file_name] = dp
provides = [pn]
for provide in (self.getVar("PROVIDES", file_name, True) or "").split():
if provide not in provides:
provides.append(provide)
# Build forward and reverse provider hashes
# Forward: virtual -> [filenames]
# Reverse: PN -> [virtuals]
if pn not in cacheData.pn_provides:
cacheData.pn_provides[pn] = []
cacheData.fn_provides[file_name] = provides
for provide in provides:
if provide not in cacheData.providers:
cacheData.providers[provide] = []
cacheData.providers[provide].append(file_name)
if not provide in cacheData.pn_provides[pn]:
cacheData.pn_provides[pn].append(provide)
cacheData.deps[file_name] = []
for dep in depends:
if not dep in cacheData.deps[file_name]:
cacheData.deps[file_name].append(dep)
if not dep in cacheData.all_depends:
cacheData.all_depends.append(dep)
# Build reverse hash for PACKAGES, so runtime dependencies
# can be be resolved (RDEPENDS, RRECOMMENDS etc.)
for package in packages:
if not package in cacheData.packages:
cacheData.packages[package] = []
cacheData.packages[package].append(file_name)
rprovides += (self.getVar("RPROVIDES_%s" % package, file_name, 1) or "").split()
for package in packages_dynamic:
if not package in cacheData.packages_dynamic:
cacheData.packages_dynamic[package] = []
cacheData.packages_dynamic[package].append(file_name)
for rprovide in rprovides:
if not rprovide in cacheData.rproviders:
cacheData.rproviders[rprovide] = []
cacheData.rproviders[rprovide].append(file_name)
# Build hash of runtime depends and rececommends
if not file_name in cacheData.rundeps:
cacheData.rundeps[file_name] = {}
if not file_name in cacheData.runrecs:
cacheData.runrecs[file_name] = {}
rdepends = self.getVar('RDEPENDS', file_name, True) or ""
rrecommends = self.getVar('RRECOMMENDS', file_name, True) or ""
for package in packages + [pn]:
if not package in cacheData.rundeps[file_name]:
cacheData.rundeps[file_name][package] = []
if not package in cacheData.runrecs[file_name]:
cacheData.runrecs[file_name][package] = []
cacheData.rundeps[file_name][package] = rdepends + " " + (self.getVar("RDEPENDS_%s" % package, file_name, True) or "")
cacheData.runrecs[file_name][package] = rrecommends + " " + (self.getVar("RRECOMMENDS_%s" % package, file_name, True) or "")
# Collect files we may need for possible world-dep
# calculations
if not self.getVar('BROKEN', file_name, True) and not self.getVar('EXCLUDE_FROM_WORLD', file_name, True):
cacheData.possible_world.append(file_name)
# Touch this to make sure its in the cache
self.getVar('__BB_DONT_CACHE', file_name, True)
self.getVar('BBCLASSEXTEND', file_name, True)
def load_bbfile( self, bbfile , config):
"""
Load and parse one .bb build file
Return the data and whether parsing resulted in the file being skipped
"""
import bb
from bb import utils, data, parse, debug, event, fatal
# expand tmpdir to include this topdir
data.setVar('TMPDIR', data.getVar('TMPDIR', config, 1) or "", config)
bbfile_loc = os.path.abspath(os.path.dirname(bbfile))
oldpath = os.path.abspath(os.getcwd())
if bb.parse.cached_mtime_noerror(bbfile_loc):
os.chdir(bbfile_loc)
bb_data = data.init_db(config)
try:
bb_data = parse.handle(bbfile, bb_data) # read .bb data
os.chdir(oldpath)
return bb_data, False
except bb.parse.SkipPackage:
os.chdir(oldpath)
return bb_data, True
except:
os.chdir(oldpath)
raise
def init(cooker):
"""
The Objective: Cache the minimum amount of data possible yet get to the
stage of building packages (i.e. tryBuild) without reparsing any .bb files.
To do this, we intercept getVar calls and only cache the variables we see
being accessed. We rely on the cache getVar calls being made for all
variables bitbake might need to use to reach this stage. For each cached
file we need to track:
* Its mtime
* The mtimes of all its dependencies
* Whether it caused a parse.SkipPackage exception
Files causing parsing errors are evicted from the cache.
"""
return Cache(cooker)
#============================================================================#
# CacheData
#============================================================================#
class CacheData:
"""
The data structures we compile from the cached data
"""
def __init__(self):
"""
Direct cache variables
(from Cache.handle_data)
"""
self.providers = {}
self.rproviders = {}
self.packages = {}
self.packages_dynamic = {}
self.possible_world = []
self.pkg_pn = {}
self.pkg_fn = {}
self.pkg_pepvpr = {}
self.pkg_dp = {}
self.pn_provides = {}
self.fn_provides = {}
self.all_depends = []
self.deps = {}
self.rundeps = {}
self.runrecs = {}
self.task_queues = {}
self.task_deps = {}
self.stamp = {}
self.preferred = {}
"""
Indirect Cache variables
(set elsewhere)
"""
self.ignored_dependencies = []
self.world_target = set()
self.bbfile_priority = {}
self.bbfile_config_priorities = []

View File

@@ -0,0 +1,272 @@
"""
BitBake 'Command' module
Provide an interface to interact with the bitbake server through 'commands'
"""
# Copyright (C) 2006-2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
"""
The bitbake server takes 'commands' from its UI/commandline.
Commands are either synchronous or asynchronous.
Async commands return data to the client in the form of events.
Sync commands must only return data through the function return value
and must not trigger events, directly or indirectly.
Commands are queued in a CommandQueue
"""
import bb
async_cmds = {}
sync_cmds = {}
class Command:
"""
A queue of asynchronous commands for bitbake
"""
def __init__(self, cooker):
self.cooker = cooker
self.cmds_sync = CommandsSync()
self.cmds_async = CommandsAsync()
# FIXME Add lock for this
self.currentAsyncCommand = None
for attr in CommandsSync.__dict__:
command = attr[:].lower()
method = getattr(CommandsSync, attr)
sync_cmds[command] = (method)
for attr in CommandsAsync.__dict__:
command = attr[:].lower()
method = getattr(CommandsAsync, attr)
async_cmds[command] = (method)
def runCommand(self, commandline):
bb.debug("Running command %s" % commandline)
try:
command = commandline.pop(0)
if command in CommandsSync.__dict__:
# Can run synchronous commands straight away
return getattr(CommandsSync, command)(self.cmds_sync, self, commandline)
if self.currentAsyncCommand is not None:
return "Busy (%s in progress)" % self.currentAsyncCommand[0]
if command not in CommandsAsync.__dict__:
return "No such command"
self.currentAsyncCommand = (command, commandline)
self.cooker.server.register_idle_function(self.cooker.runCommands, self.cooker)
return True
except:
import traceback
return traceback.format_exc()
def runAsyncCommand(self):
try:
if self.currentAsyncCommand is not None:
(command, options) = self.currentAsyncCommand
commandmethod = getattr(CommandsAsync, command)
needcache = getattr( commandmethod, "needcache" )
if needcache and self.cooker.cookerState != bb.cooker.cookerParsed:
self.cooker.updateCache()
return True
else:
commandmethod(self.cmds_async, self, options)
return False
else:
return False
except:
import traceback
self.finishAsyncCommand(traceback.format_exc())
return False
def finishAsyncCommand(self, error = None):
if error:
bb.event.fire(bb.command.CookerCommandFailed(self.cooker.configuration.event_data, error))
else:
bb.event.fire(bb.command.CookerCommandCompleted(self.cooker.configuration.event_data))
self.currentAsyncCommand = None
class CommandsSync:
"""
A class of synchronous commands
These should run quickly so as not to hurt interactive performance.
These must not influence any running synchronous command.
"""
def stateShutdown(self, command, params):
"""
Trigger cooker 'shutdown' mode
"""
command.cooker.cookerAction = bb.cooker.cookerShutdown
def stateStop(self, command, params):
"""
Stop the cooker
"""
command.cooker.cookerAction = bb.cooker.cookerStop
def getCmdLineAction(self, command, params):
"""
Get any command parsed from the commandline
"""
return command.cooker.commandlineAction
def getVariable(self, command, params):
"""
Read the value of a variable from configuration.data
"""
varname = params[0]
expand = True
if len(params) > 1:
expand = params[1]
return bb.data.getVar(varname, command.cooker.configuration.data, expand)
def setVariable(self, command, params):
"""
Set the value of variable in configuration.data
"""
varname = params[0]
value = params[1]
bb.data.setVar(varname, value, command.cooker.configuration.data)
class CommandsAsync:
"""
A class of asynchronous commands
These functions communicate via generated events.
Any function that requires metadata parsing should be here.
"""
def buildFile(self, command, params):
"""
Build a single specified .bb file
"""
bfile = params[0]
task = params[1]
command.cooker.buildFile(bfile, task)
buildFile.needcache = False
def buildTargets(self, command, params):
"""
Build a set of targets
"""
pkgs_to_build = params[0]
task = params[1]
command.cooker.buildTargets(pkgs_to_build, task)
buildTargets.needcache = True
def generateDepTreeEvent(self, command, params):
"""
Generate an event containing the dependency information
"""
pkgs_to_build = params[0]
task = params[1]
command.cooker.generateDepTreeEvent(pkgs_to_build, task)
command.finishAsyncCommand()
generateDepTreeEvent.needcache = True
def generateDotGraph(self, command, params):
"""
Dump dependency information to disk as .dot files
"""
pkgs_to_build = params[0]
task = params[1]
command.cooker.generateDotGraphFiles(pkgs_to_build, task)
command.finishAsyncCommand()
generateDotGraph.needcache = True
def showVersions(self, command, params):
"""
Show the currently selected versions
"""
command.cooker.showVersions()
command.finishAsyncCommand()
showVersions.needcache = True
def showEnvironmentTarget(self, command, params):
"""
Print the environment of a target recipe
(needs the cache to work out which recipe to use)
"""
pkg = params[0]
command.cooker.showEnvironment(None, pkg)
command.finishAsyncCommand()
showEnvironmentTarget.needcache = True
def showEnvironment(self, command, params):
"""
Print the standard environment
or if specified the environment for a specified recipe
"""
bfile = params[0]
command.cooker.showEnvironment(bfile)
command.finishAsyncCommand()
showEnvironment.needcache = False
def parseFiles(self, command, params):
"""
Parse the .bb files
"""
command.cooker.updateCache()
command.finishAsyncCommand()
parseFiles.needcache = True
def compareRevisions(self, command, params):
"""
Parse the .bb files
"""
command.cooker.compareRevisions()
command.finishAsyncCommand()
compareRevisions.needcache = True
#
# Events
#
class CookerCommandCompleted(bb.event.Event):
"""
Cooker command completed
"""
def __init__(self, data):
bb.event.Event.__init__(self, data)
class CookerCommandFailed(bb.event.Event):
"""
Cooker command completed
"""
def __init__(self, data, error):
bb.event.Event.__init__(self, data)
self.error = error
class CookerCommandSetExitCode(bb.event.Event):
"""
Set the exit code for a cooker command
"""
def __init__(self, data, exitcode):
bb.event.Event.__init__(self, data)
self.exitcode = int(exitcode)

View File

@@ -0,0 +1,977 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
# Copyright (C) 2003 - 2005 Michael 'Mickey' Lauer
# Copyright (C) 2005 Holger Hans Peter Freyther
# Copyright (C) 2005 ROAD GmbH
# Copyright (C) 2006 - 2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import sys, os, getopt, glob, copy, os.path, re, time
import bb
from bb import utils, data, parse, event, cache, providers, taskdata, runqueue
from bb import xmlrpcserver, command
import itertools, sre_constants
class MultipleMatches(Exception):
"""
Exception raised when multiple file matches are found
"""
class ParsingErrorsFound(Exception):
"""
Exception raised when parsing errors are found
"""
class NothingToBuild(Exception):
"""
Exception raised when there is nothing to build
"""
# Different states cooker can be in
cookerClean = 1
cookerParsing = 2
cookerParsed = 3
# Different action states the cooker can be in
cookerRun = 1 # Cooker is running normally
cookerShutdown = 2 # Active tasks should be brought to a controlled stop
cookerStop = 3 # Stop, now!
#============================================================================#
# BBCooker
#============================================================================#
class BBCooker:
"""
Manages one bitbake build run
"""
def __init__(self, configuration):
self.status = None
self.cache = None
self.bb_cache = None
self.server = bb.xmlrpcserver.BitBakeXMLRPCServer(self)
#self.server.register_function(self.showEnvironment)
self.configuration = configuration
if self.configuration.verbose:
bb.msg.set_verbose(True)
if self.configuration.debug:
bb.msg.set_debug_level(self.configuration.debug)
else:
bb.msg.set_debug_level(0)
if self.configuration.debug_domains:
bb.msg.set_debug_domains(self.configuration.debug_domains)
self.configuration.data = bb.data.init()
bb.data.inheritFromOS(self.configuration.data)
for f in self.configuration.file:
self.parseConfigurationFile( f )
self.parseConfigurationFile( os.path.join( "conf", "bitbake.conf" ) )
if not self.configuration.cmd:
self.configuration.cmd = bb.data.getVar("BB_DEFAULT_TASK", self.configuration.data, True) or "build"
bbpkgs = bb.data.getVar('BBPKGS', self.configuration.data, True)
if bbpkgs and len(self.configuration.pkgs_to_build) == 0:
self.configuration.pkgs_to_build.extend(bbpkgs.split())
#
# Special updated configuration we use for firing events
#
self.configuration.event_data = bb.data.createCopy(self.configuration.data)
bb.data.update_data(self.configuration.event_data)
# TOSTOP must not be set or our children will hang when they output
fd = sys.stdout.fileno()
if os.isatty(fd):
import termios
tcattr = termios.tcgetattr(fd)
if tcattr[3] & termios.TOSTOP:
bb.msg.note(1, bb.msg.domain.Build, "The terminal had the TOSTOP bit set, clearing...")
tcattr[3] = tcattr[3] & ~termios.TOSTOP
termios.tcsetattr(fd, termios.TCSANOW, tcattr)
self.command = bb.command.Command(self)
self.cookerState = cookerClean
self.cookerAction = cookerRun
def parseConfiguration(self):
# Change nice level if we're asked to
nice = bb.data.getVar("BB_NICE_LEVEL", self.configuration.data, True)
if nice:
curnice = os.nice(0)
nice = int(nice) - curnice
bb.msg.note(2, bb.msg.domain.Build, "Renice to %s " % os.nice(nice))
def parseCommandLine(self):
# Parse any commandline into actions
if self.configuration.show_environment:
self.commandlineAction = None
if 'world' in self.configuration.pkgs_to_build:
bb.error("'world' is not a valid target for --environment.")
elif len(self.configuration.pkgs_to_build) > 1:
bb.error("Only one target can be used with the --environment option.")
elif self.configuration.buildfile and len(self.configuration.pkgs_to_build) > 0:
bb.error("No target should be used with the --environment and --buildfile options.")
elif len(self.configuration.pkgs_to_build) > 0:
self.commandlineAction = ["showEnvironmentTarget", self.configuration.pkgs_to_build]
else:
self.commandlineAction = ["showEnvironment", self.configuration.buildfile]
elif self.configuration.buildfile is not None:
self.commandlineAction = ["buildFile", self.configuration.buildfile, self.configuration.cmd]
elif self.configuration.revisions_changed:
self.commandlineAction = ["compareRevisions"]
elif self.configuration.show_versions:
self.commandlineAction = ["showVersions"]
elif self.configuration.parse_only:
self.commandlineAction = ["parseFiles"]
# FIXME - implement
#elif self.configuration.interactive:
# self.interactiveMode()
elif self.configuration.dot_graph:
if self.configuration.pkgs_to_build:
self.commandlineAction = ["generateDotGraph", self.configuration.pkgs_to_build, self.configuration.cmd]
else:
self.commandlineAction = None
bb.error("Please specify a package name for dependency graph generation.")
else:
if self.configuration.pkgs_to_build:
self.commandlineAction = ["buildTargets", self.configuration.pkgs_to_build, self.configuration.cmd]
else:
self.commandlineAction = None
bb.error("Nothing to do. Use 'bitbake world' to build everything, or run 'bitbake --help' for usage information.")
def runCommands(self, server, data, abort):
"""
Run any queued asynchronous command
This is done by the idle handler so it runs in true context rather than
tied to any UI.
"""
return self.command.runAsyncCommand()
def tryBuildPackage(self, fn, item, task, the_data):
"""
Build one task of a package, optionally build following task depends
"""
try:
if not self.configuration.dry_run:
bb.build.exec_task('do_%s' % task, the_data)
return True
except bb.build.FuncFailed:
bb.msg.error(bb.msg.domain.Build, "task stack execution failed")
raise
except bb.build.EventException, e:
event = e.args[1]
bb.msg.error(bb.msg.domain.Build, "%s event exception, aborting" % bb.event.getName(event))
raise
def tryBuild(self, fn, task):
"""
Build a provider and its dependencies.
build_depends is a list of previous build dependencies (not runtime)
If build_depends is empty, we're dealing with a runtime depends
"""
the_data = self.bb_cache.loadDataFull(fn, self.configuration.data)
item = self.status.pkg_fn[fn]
#if bb.build.stamp_is_current('do_%s' % self.configuration.cmd, the_data):
# return True
return self.tryBuildPackage(fn, item, task, the_data)
def showVersions(self):
# Need files parsed
self.updateCache()
pkg_pn = self.status.pkg_pn
preferred_versions = {}
latest_versions = {}
# Sort by priority
for pn in pkg_pn.keys():
(last_ver,last_file,pref_ver,pref_file) = bb.providers.findBestProvider(pn, self.configuration.data, self.status)
preferred_versions[pn] = (pref_ver, pref_file)
latest_versions[pn] = (last_ver, last_file)
pkg_list = pkg_pn.keys()
pkg_list.sort()
bb.msg.plain("%-35s %25s %25s" % ("Package Name", "Latest Version", "Preferred Version"))
bb.msg.plain("%-35s %25s %25s\n" % ("============", "==============", "================="))
for p in pkg_list:
pref = preferred_versions[p]
latest = latest_versions[p]
prefstr = pref[0][0] + ":" + pref[0][1] + '-' + pref[0][2]
lateststr = latest[0][0] + ":" + latest[0][1] + "-" + latest[0][2]
if pref == latest:
prefstr = ""
bb.msg.plain("%-35s %25s %25s" % (p, lateststr, prefstr))
def compareRevisions(self):
ret = bb.fetch.fetcher_compare_revisons(self.configuration.data)
bb.event.fire(bb.command.CookerCommandSetExitCode(self.configuration.event_data, ret))
def showEnvironment(self, buildfile = None, pkgs_to_build = []):
"""
Show the outer or per-package environment
"""
fn = None
envdata = None
if buildfile:
self.cb = None
self.bb_cache = bb.cache.init(self)
fn = self.matchFile(buildfile)
elif len(pkgs_to_build) == 1:
self.updateCache()
localdata = data.createCopy(self.configuration.data)
bb.data.update_data(localdata)
bb.data.expandKeys(localdata)
taskdata = bb.taskdata.TaskData(self.configuration.abort)
taskdata.add_provider(localdata, self.status, pkgs_to_build[0])
taskdata.add_unresolved(localdata, self.status)
targetid = taskdata.getbuild_id(pkgs_to_build[0])
fnid = taskdata.build_targets[targetid][0]
fn = taskdata.fn_index[fnid]
else:
envdata = self.configuration.data
if fn:
try:
envdata = self.bb_cache.loadDataFull(fn, self.configuration.data)
except IOError, e:
bb.msg.error(bb.msg.domain.Parsing, "Unable to read %s: %s" % (fn, e))
raise
except Exception, e:
bb.msg.error(bb.msg.domain.Parsing, "%s" % e)
raise
class dummywrite:
def __init__(self):
self.writebuf = ""
def write(self, output):
self.writebuf = self.writebuf + output
# emit variables and shell functions
try:
data.update_data(envdata)
wb = dummywrite()
data.emit_env(wb, envdata, True)
bb.msg.plain(wb.writebuf)
except Exception, e:
bb.msg.fatal(bb.msg.domain.Parsing, "%s" % e)
# emit the metadata which isnt valid shell
data.expandKeys(envdata)
for e in envdata.keys():
if data.getVarFlag( e, 'python', envdata ):
bb.msg.plain("\npython %s () {\n%s}\n" % (e, data.getVar(e, envdata, 1)))
def generateDepTreeData(self, pkgs_to_build, task):
"""
Create a dependency tree of pkgs_to_build, returning the data.
"""
# Need files parsed
self.updateCache()
# If we are told to do the None task then query the default task
if (task == None):
task = self.configuration.cmd
pkgs_to_build = self.checkPackages(pkgs_to_build)
localdata = data.createCopy(self.configuration.data)
bb.data.update_data(localdata)
bb.data.expandKeys(localdata)
taskdata = bb.taskdata.TaskData(self.configuration.abort)
runlist = []
for k in pkgs_to_build:
taskdata.add_provider(localdata, self.status, k)
runlist.append([k, "do_%s" % task])
taskdata.add_unresolved(localdata, self.status)
rq = bb.runqueue.RunQueue(self, self.configuration.data, self.status, taskdata, runlist)
rq.prepare_runqueue()
seen_fnids = []
depend_tree = {}
depend_tree["depends"] = {}
depend_tree["tdepends"] = {}
depend_tree["pn"] = {}
depend_tree["rdepends-pn"] = {}
depend_tree["packages"] = {}
depend_tree["rdepends-pkg"] = {}
depend_tree["rrecs-pkg"] = {}
for task in range(len(rq.runq_fnid)):
taskname = rq.runq_task[task]
fnid = rq.runq_fnid[task]
fn = taskdata.fn_index[fnid]
pn = self.status.pkg_fn[fn]
version = "%s:%s-%s" % self.status.pkg_pepvpr[fn]
if pn not in depend_tree["pn"]:
depend_tree["pn"][pn] = {}
depend_tree["pn"][pn]["filename"] = fn
depend_tree["pn"][pn]["version"] = version
for dep in rq.runq_depends[task]:
depfn = taskdata.fn_index[rq.runq_fnid[dep]]
deppn = self.status.pkg_fn[depfn]
dotname = "%s.%s" % (pn, rq.runq_task[task])
if not dotname in depend_tree["tdepends"]:
depend_tree["tdepends"][dotname] = []
depend_tree["tdepends"][dotname].append("%s.%s" % (deppn, rq.runq_task[dep]))
if fnid not in seen_fnids:
seen_fnids.append(fnid)
packages = []
depend_tree["depends"][pn] = []
for dep in taskdata.depids[fnid]:
depend_tree["depends"][pn].append(taskdata.build_names_index[dep])
depend_tree["rdepends-pn"][pn] = []
for rdep in taskdata.rdepids[fnid]:
depend_tree["rdepends-pn"][pn].append(taskdata.run_names_index[rdep])
rdepends = self.status.rundeps[fn]
for package in rdepends:
depend_tree["rdepends-pkg"][package] = []
for rdepend in rdepends[package]:
depend_tree["rdepends-pkg"][package].append(rdepend)
packages.append(package)
rrecs = self.status.runrecs[fn]
for package in rrecs:
depend_tree["rrecs-pkg"][package] = []
for rdepend in rrecs[package]:
depend_tree["rrecs-pkg"][package].append(rdepend)
if not package in packages:
packages.append(package)
for package in packages:
if package not in depend_tree["packages"]:
depend_tree["packages"][package] = {}
depend_tree["packages"][package]["pn"] = pn
depend_tree["packages"][package]["filename"] = fn
depend_tree["packages"][package]["version"] = version
return depend_tree
def generateDepTreeEvent(self, pkgs_to_build, task):
"""
Create a task dependency graph of pkgs_to_build.
Generate an event with the result
"""
depgraph = self.generateDepTreeData(pkgs_to_build, task)
bb.event.fire(bb.event.DepTreeGenerated(self.configuration.data, depgraph))
def generateDotGraphFiles(self, pkgs_to_build, task):
"""
Create a task dependency graph of pkgs_to_build.
Save the result to a set of .dot files.
"""
depgraph = self.generateDepTreeData(pkgs_to_build, task)
# Prints a flattened form of package-depends below where subpackages of a package are merged into the main pn
depends_file = file('pn-depends.dot', 'w' )
print >> depends_file, "digraph depends {"
for pn in depgraph["pn"]:
fn = depgraph["pn"][pn]["filename"]
version = depgraph["pn"][pn]["version"]
print >> depends_file, '"%s" [label="%s %s\\n%s"]' % (pn, pn, version, fn)
for pn in depgraph["depends"]:
for depend in depgraph["depends"][pn]:
print >> depends_file, '"%s" -> "%s"' % (pn, depend)
for pn in depgraph["rdepends-pn"]:
for rdepend in depgraph["rdepends-pn"][pn]:
print >> depends_file, '"%s" -> "%s" [style=dashed]' % (pn, rdepend)
print >> depends_file, "}"
bb.msg.plain("PN dependencies saved to 'pn-depends.dot'")
depends_file = file('package-depends.dot', 'w' )
print >> depends_file, "digraph depends {"
for package in depgraph["packages"]:
pn = depgraph["packages"][package]["pn"]
fn = depgraph["packages"][package]["filename"]
version = depgraph["packages"][package]["version"]
if package == pn:
print >> depends_file, '"%s" [label="%s %s\\n%s"]' % (pn, pn, version, fn)
else:
print >> depends_file, '"%s" [label="%s(%s) %s\\n%s"]' % (package, package, pn, version, fn)
for depend in depgraph["depends"][pn]:
print >> depends_file, '"%s" -> "%s"' % (package, depend)
for package in depgraph["rdepends-pkg"]:
for rdepend in depgraph["rdepends-pkg"][package]:
print >> depends_file, '"%s" -> "%s" [style=dashed]' % (package, rdepend)
for package in depgraph["rrecs-pkg"]:
for rdepend in depgraph["rrecs-pkg"][package]:
print >> depends_file, '"%s" -> "%s" [style=dashed]' % (package, rdepend)
print >> depends_file, "}"
bb.msg.plain("Package dependencies saved to 'package-depends.dot'")
tdepends_file = file('task-depends.dot', 'w' )
print >> tdepends_file, "digraph depends {"
for task in depgraph["tdepends"]:
(pn, taskname) = task.rsplit(".", 1)
fn = depgraph["pn"][pn]["filename"]
version = depgraph["pn"][pn]["version"]
print >> tdepends_file, '"%s.%s" [label="%s %s\\n%s\\n%s"]' % (pn, taskname, pn, taskname, version, fn)
for dep in depgraph["tdepends"][task]:
print >> tdepends_file, '"%s" -> "%s"' % (task, dep)
print >> tdepends_file, "}"
bb.msg.plain("Task dependencies saved to 'task-depends.dot'")
def buildDepgraph( self ):
all_depends = self.status.all_depends
pn_provides = self.status.pn_provides
localdata = data.createCopy(self.configuration.data)
bb.data.update_data(localdata)
bb.data.expandKeys(localdata)
def calc_bbfile_priority(filename):
for (regex, pri) in self.status.bbfile_config_priorities:
if regex.match(filename):
return pri
return 0
# Handle PREFERRED_PROVIDERS
for p in (bb.data.getVar('PREFERRED_PROVIDERS', localdata, 1) or "").split():
try:
(providee, provider) = p.split(':')
except:
bb.msg.fatal(bb.msg.domain.Provider, "Malformed option in PREFERRED_PROVIDERS variable: %s" % p)
continue
if providee in self.status.preferred and self.status.preferred[providee] != provider:
bb.msg.error(bb.msg.domain.Provider, "conflicting preferences for %s: both %s and %s specified" % (providee, provider, self.status.preferred[providee]))
self.status.preferred[providee] = provider
# Calculate priorities for each file
for p in self.status.pkg_fn.keys():
self.status.bbfile_priority[p] = calc_bbfile_priority(p)
def buildWorldTargetList(self):
"""
Build package list for "bitbake world"
"""
all_depends = self.status.all_depends
pn_provides = self.status.pn_provides
bb.msg.debug(1, bb.msg.domain.Parsing, "collating packages for \"world\"")
for f in self.status.possible_world:
terminal = True
pn = self.status.pkg_fn[f]
for p in pn_provides[pn]:
if p.startswith('virtual/'):
bb.msg.debug(2, bb.msg.domain.Parsing, "World build skipping %s due to %s provider starting with virtual/" % (f, p))
terminal = False
break
for pf in self.status.providers[p]:
if self.status.pkg_fn[pf] != pn:
bb.msg.debug(2, bb.msg.domain.Parsing, "World build skipping %s due to both us and %s providing %s" % (f, pf, p))
terminal = False
break
if terminal:
self.status.world_target.add(pn)
# drop reference count now
self.status.possible_world = None
self.status.all_depends = None
def interactiveMode( self ):
"""Drop off into a shell"""
try:
from bb import shell
except ImportError, details:
bb.msg.fatal(bb.msg.domain.Parsing, "Sorry, shell not available (%s)" % details )
else:
shell.start( self )
def parseConfigurationFile( self, afile ):
try:
self.configuration.data = bb.parse.handle( afile, self.configuration.data )
# Handle any INHERITs and inherit the base class
inherits = ["base"] + (bb.data.getVar('INHERIT', self.configuration.data, True ) or "").split()
for inherit in inherits:
self.configuration.data = bb.parse.handle(os.path.join('classes', '%s.bbclass' % inherit), self.configuration.data, True )
# Nomally we only register event handlers at the end of parsing .bb files
# We register any handlers we've found so far here...
for var in data.getVar('__BBHANDLERS', self.configuration.data) or []:
bb.event.register(var,bb.data.getVar(var, self.configuration.data))
bb.fetch.fetcher_init(self.configuration.data)
bb.event.fire(bb.event.ConfigParsed(self.configuration.data))
except IOError, e:
bb.msg.fatal(bb.msg.domain.Parsing, "Error when parsing %s: %s" % (afile, str(e)))
except bb.parse.ParseError, details:
bb.msg.fatal(bb.msg.domain.Parsing, "Unable to parse %s (%s)" % (afile, details) )
def handleCollections( self, collections ):
"""Handle collections"""
if collections:
collection_list = collections.split()
for c in collection_list:
regex = bb.data.getVar("BBFILE_PATTERN_%s" % c, self.configuration.data, 1)
if regex == None:
bb.msg.error(bb.msg.domain.Parsing, "BBFILE_PATTERN_%s not defined" % c)
continue
priority = bb.data.getVar("BBFILE_PRIORITY_%s" % c, self.configuration.data, 1)
if priority == None:
bb.msg.error(bb.msg.domain.Parsing, "BBFILE_PRIORITY_%s not defined" % c)
continue
try:
cre = re.compile(regex)
except re.error:
bb.msg.error(bb.msg.domain.Parsing, "BBFILE_PATTERN_%s \"%s\" is not a valid regular expression" % (c, regex))
continue
try:
pri = int(priority)
self.status.bbfile_config_priorities.append((cre, pri))
except ValueError:
bb.msg.error(bb.msg.domain.Parsing, "invalid value for BBFILE_PRIORITY_%s: \"%s\"" % (c, priority))
def buildSetVars(self):
"""
Setup any variables needed before starting a build
"""
if not bb.data.getVar("BUILDNAME", self.configuration.data):
bb.data.setVar("BUILDNAME", os.popen('date +%Y%m%d%H%M').readline().strip(), self.configuration.data)
bb.data.setVar("BUILDSTART", time.strftime('%m/%d/%Y %H:%M:%S',time.gmtime()), self.configuration.data)
def matchFiles(self, buildfile):
"""
Find the .bb files which match the expression in 'buildfile'.
"""
bf = os.path.abspath(buildfile)
try:
os.stat(bf)
return [bf]
except OSError:
(filelist, masked) = self.collect_bbfiles()
regexp = re.compile(buildfile)
matches = []
for f in filelist:
if regexp.search(f) and os.path.isfile(f):
bf = f
matches.append(f)
return matches
def matchFile(self, buildfile):
"""
Find the .bb file which matches the expression in 'buildfile'.
Raise an error if multiple files
"""
matches = self.matchFiles(buildfile)
if len(matches) != 1:
bb.msg.error(bb.msg.domain.Parsing, "Unable to match %s (%s matches found):" % (buildfile, len(matches)))
for f in matches:
bb.msg.error(bb.msg.domain.Parsing, " %s" % f)
raise MultipleMatches
return matches[0]
def buildFile(self, buildfile, task):
"""
Build the file matching regexp buildfile
"""
# Parse the configuration here. We need to do it explicitly here since
# buildFile() doesn't use the cache
self.parseConfiguration()
# If we are told to do the None task then query the default task
if (task == None):
task = self.configuration.cmd
fn = self.matchFile(buildfile)
self.buildSetVars()
# Load data into the cache for fn and parse the loaded cache data
self.bb_cache = bb.cache.init(self)
self.status = bb.cache.CacheData()
self.bb_cache.loadData(fn, self.configuration.data, self.status)
# Tweak some variables
item = self.bb_cache.getVar('PN', fn, True)
self.status.ignored_dependencies = set()
self.status.bbfile_priority[fn] = 1
# Remove external dependencies
self.status.task_deps[fn]['depends'] = {}
self.status.deps[fn] = []
self.status.rundeps[fn] = []
self.status.runrecs[fn] = []
# Remove stamp for target if force mode active
if self.configuration.force:
bb.msg.note(2, bb.msg.domain.RunQueue, "Remove stamp %s, %s" % (task, fn))
bb.build.del_stamp('do_%s' % task, self.status, fn)
# Setup taskdata structure
taskdata = bb.taskdata.TaskData(self.configuration.abort)
taskdata.add_provider(self.configuration.data, self.status, item)
buildname = bb.data.getVar("BUILDNAME", self.configuration.data)
bb.event.fire(bb.event.BuildStarted(buildname, [item], self.configuration.event_data))
# Execute the runqueue
runlist = [[item, "do_%s" % task]]
rq = bb.runqueue.RunQueue(self, self.configuration.data, self.status, taskdata, runlist)
def buildFileIdle(server, rq, abort):
if abort or self.cookerAction == cookerStop:
rq.finish_runqueue(True)
elif self.cookerAction == cookerShutdown:
rq.finish_runqueue(False)
failures = 0
try:
retval = rq.execute_runqueue()
except runqueue.TaskFailure, fnids:
for fnid in fnids:
bb.msg.error(bb.msg.domain.Build, "'%s' failed" % taskdata.fn_index[fnid])
failures = failures + 1
retval = False
if not retval:
self.command.finishAsyncCommand()
bb.event.fire(bb.event.BuildCompleted(buildname, targets, self.configuration.event_data, failures))
return False
return 0.5
self.server.register_idle_function(buildFileIdle, rq)
def buildTargets(self, targets, task):
"""
Attempt to build the targets specified
"""
# Need files parsed
self.updateCache()
# If we are told to do the NULL task then query the default task
if (task == None):
task = self.configuration.cmd
targets = self.checkPackages(targets)
def buildTargetsIdle(server, rq, abort):
if abort or self.cookerAction == cookerStop:
rq.finish_runqueue(True)
elif self.cookerAction == cookerShutdown:
rq.finish_runqueue(False)
failures = 0
try:
retval = rq.execute_runqueue()
except runqueue.TaskFailure, fnids:
for fnid in fnids:
bb.msg.error(bb.msg.domain.Build, "'%s' failed" % taskdata.fn_index[fnid])
failures = failures + 1
retval = False
if not retval:
self.command.finishAsyncCommand()
bb.event.fire(bb.event.BuildCompleted(buildname, targets, self.configuration.event_data, failures))
return None
return 0.5
self.buildSetVars()
buildname = bb.data.getVar("BUILDNAME", self.configuration.data)
bb.event.fire(bb.event.BuildStarted(buildname, targets, self.configuration.event_data))
localdata = data.createCopy(self.configuration.data)
bb.data.update_data(localdata)
bb.data.expandKeys(localdata)
taskdata = bb.taskdata.TaskData(self.configuration.abort)
runlist = []
for k in targets:
taskdata.add_provider(localdata, self.status, k)
runlist.append([k, "do_%s" % task])
taskdata.add_unresolved(localdata, self.status)
rq = bb.runqueue.RunQueue(self, self.configuration.data, self.status, taskdata, runlist)
self.server.register_idle_function(buildTargetsIdle, rq)
def updateCache(self):
if self.cookerState == cookerParsed:
return
if self.cookerState != cookerParsing:
self.parseConfiguration ()
# Import Psyco if available and not disabled
import platform
if platform.machine() in ['i386', 'i486', 'i586', 'i686']:
if not self.configuration.disable_psyco:
try:
import psyco
except ImportError:
bb.msg.note(1, bb.msg.domain.Collection, "Psyco JIT Compiler (http://psyco.sf.net) not available. Install it to increase performance.")
else:
psyco.bind( CookerParser.parse_next )
else:
bb.msg.note(1, bb.msg.domain.Collection, "You have disabled Psyco. This decreases performance.")
self.status = bb.cache.CacheData()
ignore = bb.data.getVar("ASSUME_PROVIDED", self.configuration.data, 1) or ""
self.status.ignored_dependencies = set(ignore.split())
for dep in self.configuration.extra_assume_provided:
self.status.ignored_dependencies.add(dep)
self.handleCollections( bb.data.getVar("BBFILE_COLLECTIONS", self.configuration.data, 1) )
bb.msg.debug(1, bb.msg.domain.Collection, "collecting .bb files")
(filelist, masked) = self.collect_bbfiles()
bb.data.renameVar("__depends", "__base_depends", self.configuration.data)
self.parser = CookerParser(self, filelist, masked)
self.cookerState = cookerParsing
if not self.parser.parse_next():
bb.msg.debug(1, bb.msg.domain.Collection, "parsing complete")
self.buildDepgraph()
self.cookerState = cookerParsed
return None
return 0.00001
def checkPackages(self, pkgs_to_build):
if len(pkgs_to_build) == 0:
raise NothingToBuild
if 'world' in pkgs_to_build:
self.buildWorldTargetList()
pkgs_to_build.remove('world')
for t in self.status.world_target:
pkgs_to_build.append(t)
return pkgs_to_build
def get_bbfiles( self, path = os.getcwd() ):
"""Get list of default .bb files by reading out the current directory"""
contents = os.listdir(path)
bbfiles = []
for f in contents:
(root, ext) = os.path.splitext(f)
if ext == ".bb":
bbfiles.append(os.path.abspath(os.path.join(os.getcwd(),f)))
return bbfiles
def find_bbfiles( self, path ):
"""Find all the .bb files in a directory"""
from os.path import join
found = []
for dir, dirs, files in os.walk(path):
for ignored in ('SCCS', 'CVS', '.svn'):
if ignored in dirs:
dirs.remove(ignored)
found += [join(dir,f) for f in files if f.endswith('.bb')]
return found
def collect_bbfiles( self ):
"""Collect all available .bb build files"""
parsed, cached, skipped, masked = 0, 0, 0, 0
self.bb_cache = bb.cache.init(self)
files = (data.getVar( "BBFILES", self.configuration.data, 1 ) or "").split()
data.setVar("BBFILES", " ".join(files), self.configuration.data)
if not len(files):
files = self.get_bbfiles()
if not len(files):
bb.msg.error(bb.msg.domain.Collection, "no files to build.")
newfiles = []
for f in files:
if os.path.isdir(f):
dirfiles = self.find_bbfiles(f)
if dirfiles:
newfiles += dirfiles
continue
else:
globbed = glob.glob(f)
if not globbed and os.path.exists(f):
globbed = [f]
newfiles += globbed
bbmask = bb.data.getVar('BBMASK', self.configuration.data, 1)
if not bbmask:
return (newfiles, 0)
try:
bbmask_compiled = re.compile(bbmask)
except sre_constants.error:
bb.msg.fatal(bb.msg.domain.Collection, "BBMASK is not a valid regular expression.")
finalfiles = []
for f in newfiles:
if bbmask_compiled.search(f):
bb.msg.debug(1, bb.msg.domain.Collection, "skipping masked file %s" % f)
masked += 1
continue
finalfiles.append(f)
return (finalfiles, masked)
def serve(self):
# Empty the environment. The environment will be populated as
# necessary from the data store.
bb.utils.empty_environment()
if self.configuration.profile:
try:
import cProfile as profile
except:
import profile
profile.runctx("self.server.serve_forever()", globals(), locals(), "profile.log")
# Redirect stdout to capture profile information
pout = open('profile.log.processed', 'w')
so = sys.stdout.fileno()
os.dup2(pout.fileno(), so)
import pstats
p = pstats.Stats('profile.log')
p.sort_stats('time')
p.print_stats()
p.print_callers()
p.sort_stats('cumulative')
p.print_stats()
os.dup2(so, pout.fileno())
pout.flush()
pout.close()
else:
self.server.serve_forever()
bb.event.fire(CookerExit(self.configuration.event_data))
class CookerExit(bb.event.Event):
"""
Notify clients of the Cooker shutdown
"""
def __init__(self, d):
bb.event.Event.__init__(self, d)
class CookerParser:
def __init__(self, cooker, filelist, masked):
# Internal data
self.filelist = filelist
self.cooker = cooker
# Accounting statistics
self.parsed = 0
self.cached = 0
self.skipped = 0
self.error = 0
self.masked = masked
self.total = len(filelist)
# Pointer to the next file to parse
self.pointer = 0
def parse_next(self):
print "Pointer %d" % self.pointer
if self.pointer < len(self.filelist):
f = self.filelist[self.pointer]
cooker = self.cooker
try:
fromCache, skip = cooker.bb_cache.loadData(f, cooker.configuration.data, cooker.status)
if skip:
self.skipped += 1
bb.msg.debug(2, bb.msg.domain.Collection, "skipping %s" % f)
cooker.bb_cache.skip(f)
elif fromCache: self.cached += 1
else: self.parsed += 1
except IOError, e:
self.error += 1
cooker.bb_cache.remove(f)
bb.msg.error(bb.msg.domain.Collection, "opening %s: %s" % (f, e))
pass
except KeyboardInterrupt:
cooker.bb_cache.remove(f)
cooker.bb_cache.sync()
raise
except Exception, e:
self.error += 1
cooker.bb_cache.remove(f)
bb.msg.error(bb.msg.domain.Collection, "%s while parsing %s" % (e, f))
except:
cooker.bb_cache.remove(f)
raise
finally:
bb.event.fire(bb.event.ParseProgress(cooker.configuration.event_data, self.cached, self.parsed, self.skipped, self.masked, self.error, self.total))
self.pointer += 1
if self.pointer >= self.total:
cooker.bb_cache.sync()
if self.error > 0:
raise ParsingErrorsFound
return False
return True

View File

@@ -0,0 +1,191 @@
"""
Python Deamonizing helper
Configurable daemon behaviors:
1.) The current working directory set to the "/" directory.
2.) The current file creation mode mask set to 0.
3.) Close all open files (1024).
4.) Redirect standard I/O streams to "/dev/null".
A failed call to fork() now raises an exception.
References:
1) Advanced Programming in the Unix Environment: W. Richard Stevens
2) Unix Programming Frequently Asked Questions:
http://www.erlenstar.demon.co.uk/unix/faq_toc.html
Modified to allow a function to be daemonized and return for
bitbake use by Richard Purdie
"""
__author__ = "Chad J. Schroeder"
__copyright__ = "Copyright (C) 2005 Chad J. Schroeder"
__version__ = "0.2"
# Standard Python modules.
import os # Miscellaneous OS interfaces.
import sys # System-specific parameters and functions.
# Default daemon parameters.
# File mode creation mask of the daemon.
# For BitBake's children, we do want to inherit the parent umask.
UMASK = None
# Default maximum for the number of available file descriptors.
MAXFD = 1024
# The standard I/O file descriptors are redirected to /dev/null by default.
if (hasattr(os, "devnull")):
REDIRECT_TO = os.devnull
else:
REDIRECT_TO = "/dev/null"
def createDaemon(function, logfile):
"""
Detach a process from the controlling terminal and run it in the
background as a daemon, returning control to the caller.
"""
try:
# Fork a child process so the parent can exit. This returns control to
# the command-line or shell. It also guarantees that the child will not
# be a process group leader, since the child receives a new process ID
# and inherits the parent's process group ID. This step is required
# to insure that the next call to os.setsid is successful.
pid = os.fork()
except OSError, e:
raise Exception, "%s [%d]" % (e.strerror, e.errno)
if (pid == 0): # The first child.
# To become the session leader of this new session and the process group
# leader of the new process group, we call os.setsid(). The process is
# also guaranteed not to have a controlling terminal.
os.setsid()
# Is ignoring SIGHUP necessary?
#
# It's often suggested that the SIGHUP signal should be ignored before
# the second fork to avoid premature termination of the process. The
# reason is that when the first child terminates, all processes, e.g.
# the second child, in the orphaned group will be sent a SIGHUP.
#
# "However, as part of the session management system, there are exactly
# two cases where SIGHUP is sent on the death of a process:
#
# 1) When the process that dies is the session leader of a session that
# is attached to a terminal device, SIGHUP is sent to all processes
# in the foreground process group of that terminal device.
# 2) When the death of a process causes a process group to become
# orphaned, and one or more processes in the orphaned group are
# stopped, then SIGHUP and SIGCONT are sent to all members of the
# orphaned group." [2]
#
# The first case can be ignored since the child is guaranteed not to have
# a controlling terminal. The second case isn't so easy to dismiss.
# The process group is orphaned when the first child terminates and
# POSIX.1 requires that every STOPPED process in an orphaned process
# group be sent a SIGHUP signal followed by a SIGCONT signal. Since the
# second child is not STOPPED though, we can safely forego ignoring the
# SIGHUP signal. In any case, there are no ill-effects if it is ignored.
#
# import signal # Set handlers for asynchronous events.
# signal.signal(signal.SIGHUP, signal.SIG_IGN)
try:
# Fork a second child and exit immediately to prevent zombies. This
# causes the second child process to be orphaned, making the init
# process responsible for its cleanup. And, since the first child is
# a session leader without a controlling terminal, it's possible for
# it to acquire one by opening a terminal in the future (System V-
# based systems). This second fork guarantees that the child is no
# longer a session leader, preventing the daemon from ever acquiring
# a controlling terminal.
pid = os.fork() # Fork a second child.
except OSError, e:
raise Exception, "%s [%d]" % (e.strerror, e.errno)
if (pid == 0): # The second child.
# We probably don't want the file mode creation mask inherited from
# the parent, so we give the child complete control over permissions.
if UMASK is not None:
os.umask(UMASK)
else:
# Parent (the first child) of the second child.
os._exit(0)
else:
# exit() or _exit()?
# _exit is like exit(), but it doesn't call any functions registered
# with atexit (and on_exit) or any registered signal handlers. It also
# closes any open file descriptors. Using exit() may cause all stdio
# streams to be flushed twice and any temporary files may be unexpectedly
# removed. It's therefore recommended that child branches of a fork()
# and the parent branch(es) of a daemon use _exit().
return
# Close all open file descriptors. This prevents the child from keeping
# open any file descriptors inherited from the parent. There is a variety
# of methods to accomplish this task. Three are listed below.
#
# Try the system configuration variable, SC_OPEN_MAX, to obtain the maximum
# number of open file descriptors to close. If it doesn't exists, use
# the default value (configurable).
#
# try:
# maxfd = os.sysconf("SC_OPEN_MAX")
# except (AttributeError, ValueError):
# maxfd = MAXFD
#
# OR
#
# if (os.sysconf_names.has_key("SC_OPEN_MAX")):
# maxfd = os.sysconf("SC_OPEN_MAX")
# else:
# maxfd = MAXFD
#
# OR
#
# Use the getrlimit method to retrieve the maximum file descriptor number
# that can be opened by this process. If there is not limit on the
# resource, use the default value.
#
import resource # Resource usage information.
maxfd = resource.getrlimit(resource.RLIMIT_NOFILE)[1]
if (maxfd == resource.RLIM_INFINITY):
maxfd = MAXFD
# Iterate through and close all file descriptors.
# for fd in range(0, maxfd):
# try:
# os.close(fd)
# except OSError: # ERROR, fd wasn't open to begin with (ignored)
# pass
# Redirect the standard I/O file descriptors to the specified file. Since
# the daemon has no controlling terminal, most daemons redirect stdin,
# stdout, and stderr to /dev/null. This is done to prevent side-effects
# from reads and writes to the standard I/O file descriptors.
# This call to open is guaranteed to return the lowest file descriptor,
# which will be 0 (stdin), since it was closed above.
# os.open(REDIRECT_TO, os.O_RDWR) # standard input (0)
# Duplicate standard input to standard output and standard error.
# os.dup2(0, 1) # standard output (1)
# os.dup2(0, 2) # standard error (2)
si = file('/dev/null', 'r')
so = file(logfile, 'w')
se = so
# Replace those fds with our own
os.dup2(si.fileno(), sys.stdin.fileno())
os.dup2(so.fileno(), sys.stdout.fileno())
os.dup2(se.fileno(), sys.stderr.fileno())
function()
os._exit(0)

562
bitbake-dev/lib/bb/data.py Normal file
View File

@@ -0,0 +1,562 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Data' implementations
Functions for interacting with the data structure used by the
BitBake build tools.
The expandData and update_data are the most expensive
operations. At night the cookie monster came by and
suggested 'give me cookies on setting the variables and
things will work out'. Taking this suggestion into account
applying the skills from the not yet passed 'Entwurf und
Analyse von Algorithmen' lecture and the cookie
monster seems to be right. We will track setVar more carefully
to have faster update_data and expandKeys operations.
This is a treade-off between speed and memory again but
the speed is more critical here.
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2005 Holger Hans Peter Freyther
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#Based on functions from the base bb module, Copyright 2003 Holger Schurig
import sys, os, re, types
if sys.argv[0][-5:] == "pydoc":
path = os.path.dirname(os.path.dirname(sys.argv[1]))
else:
path = os.path.dirname(os.path.dirname(sys.argv[0]))
sys.path.insert(0,path)
from bb import data_smart
import bb
_dict_type = data_smart.DataSmart
def init():
return _dict_type()
def init_db(parent = None):
if parent:
return parent.createCopy()
else:
return _dict_type()
def createCopy(source):
"""Link the source set to the destination
If one does not find the value in the destination set,
search will go on to the source set to get the value.
Value from source are copy-on-write. i.e. any try to
modify one of them will end up putting the modified value
in the destination set.
"""
return source.createCopy()
def initVar(var, d):
"""Non-destructive var init for data structure"""
d.initVar(var)
def setVar(var, value, d):
"""Set a variable to a given value
Example:
>>> d = init()
>>> setVar('TEST', 'testcontents', d)
>>> print getVar('TEST', d)
testcontents
"""
d.setVar(var,value)
def getVar(var, d, exp = 0):
"""Gets the value of a variable
Example:
>>> d = init()
>>> setVar('TEST', 'testcontents', d)
>>> print getVar('TEST', d)
testcontents
"""
return d.getVar(var,exp)
def renameVar(key, newkey, d):
"""Renames a variable from key to newkey
Example:
>>> d = init()
>>> setVar('TEST', 'testcontents', d)
>>> renameVar('TEST', 'TEST2', d)
>>> print getVar('TEST2', d)
testcontents
"""
d.renameVar(key, newkey)
def delVar(var, d):
"""Removes a variable from the data set
Example:
>>> d = init()
>>> setVar('TEST', 'testcontents', d)
>>> print getVar('TEST', d)
testcontents
>>> delVar('TEST', d)
>>> print getVar('TEST', d)
None
"""
d.delVar(var)
def setVarFlag(var, flag, flagvalue, d):
"""Set a flag for a given variable to a given value
Example:
>>> d = init()
>>> setVarFlag('TEST', 'python', 1, d)
>>> print getVarFlag('TEST', 'python', d)
1
"""
d.setVarFlag(var,flag,flagvalue)
def getVarFlag(var, flag, d):
"""Gets given flag from given var
Example:
>>> d = init()
>>> setVarFlag('TEST', 'python', 1, d)
>>> print getVarFlag('TEST', 'python', d)
1
"""
return d.getVarFlag(var,flag)
def delVarFlag(var, flag, d):
"""Removes a given flag from the variable's flags
Example:
>>> d = init()
>>> setVarFlag('TEST', 'testflag', 1, d)
>>> print getVarFlag('TEST', 'testflag', d)
1
>>> delVarFlag('TEST', 'testflag', d)
>>> print getVarFlag('TEST', 'testflag', d)
None
"""
d.delVarFlag(var,flag)
def setVarFlags(var, flags, d):
"""Set the flags for a given variable
Note:
setVarFlags will not clear previous
flags. Think of this method as
addVarFlags
Example:
>>> d = init()
>>> myflags = {}
>>> myflags['test'] = 'blah'
>>> setVarFlags('TEST', myflags, d)
>>> print getVarFlag('TEST', 'test', d)
blah
"""
d.setVarFlags(var,flags)
def getVarFlags(var, d):
"""Gets a variable's flags
Example:
>>> d = init()
>>> setVarFlag('TEST', 'test', 'blah', d)
>>> print getVarFlags('TEST', d)['test']
blah
"""
return d.getVarFlags(var)
def delVarFlags(var, d):
"""Removes a variable's flags
Example:
>>> data = init()
>>> setVarFlag('TEST', 'testflag', 1, data)
>>> print getVarFlag('TEST', 'testflag', data)
1
>>> delVarFlags('TEST', data)
>>> print getVarFlags('TEST', data)
None
"""
d.delVarFlags(var)
def keys(d):
"""Return a list of keys in d
Example:
>>> d = init()
>>> setVar('TEST', 1, d)
>>> setVar('MOO' , 2, d)
>>> setVarFlag('TEST', 'test', 1, d)
>>> keys(d)
['TEST', 'MOO']
"""
return d.keys()
def getData(d):
"""Returns the data object used"""
return d
def setData(newData, d):
"""Sets the data object to the supplied value"""
d = newData
##
## Cookie Monsters' query functions
##
def _get_override_vars(d, override):
"""
Internal!!!
Get the Names of Variables that have a specific
override. This function returns a iterable
Set or an empty list
"""
return []
def _get_var_flags_triple(d):
"""
Internal!!!
"""
return []
__expand_var_regexp__ = re.compile(r"\${[^{}]+}")
__expand_python_regexp__ = re.compile(r"\${@.+?}")
def expand(s, d, varname = None):
"""Variable expansion using the data store.
Example:
Standard expansion:
>>> d = init()
>>> setVar('A', 'sshd', d)
>>> print expand('/usr/bin/${A}', d)
/usr/bin/sshd
Python expansion:
>>> d = init()
>>> print expand('result: ${@37 * 72}', d)
result: 2664
Shell expansion:
>>> d = init()
>>> print expand('${TARGET_MOO}', d)
${TARGET_MOO}
>>> setVar('TARGET_MOO', 'yupp', d)
>>> print expand('${TARGET_MOO}',d)
yupp
>>> setVar('SRC_URI', 'http://somebug.${TARGET_MOO}', d)
>>> delVar('TARGET_MOO', d)
>>> print expand('${SRC_URI}', d)
http://somebug.${TARGET_MOO}
"""
return d.expand(s, varname)
def expandKeys(alterdata, readdata = None):
if readdata == None:
readdata = alterdata
todolist = {}
for key in keys(alterdata):
if not '${' in key:
continue
ekey = expand(key, readdata)
if key == ekey:
continue
todolist[key] = ekey
# These two for loops are split for performance to maximise the
# usefulness of the expand cache
for key in todolist:
ekey = todolist[key]
renameVar(key, ekey, alterdata)
def expandData(alterdata, readdata = None):
"""For each variable in alterdata, expand it, and update the var contents.
Replacements use data from readdata.
Example:
>>> a=init()
>>> b=init()
>>> setVar("dlmsg", "dl_dir is ${DL_DIR}", a)
>>> setVar("DL_DIR", "/path/to/whatever", b)
>>> expandData(a, b)
>>> print getVar("dlmsg", a)
dl_dir is /path/to/whatever
"""
if readdata == None:
readdata = alterdata
for key in keys(alterdata):
val = getVar(key, alterdata)
if type(val) is not types.StringType:
continue
expanded = expand(val, readdata)
# print "key is %s, val is %s, expanded is %s" % (key, val, expanded)
if val != expanded:
setVar(key, expanded, alterdata)
def inheritFromOS(d):
"""Inherit variables from the environment."""
for s in os.environ.keys():
try:
setVar(s, os.environ[s], d)
setVarFlag(s, "export", True, d)
except TypeError:
pass
def emit_var(var, o=sys.__stdout__, d = init(), all=False):
"""Emit a variable to be sourced by a shell."""
if getVarFlag(var, "python", d):
return 0
export = getVarFlag(var, "export", d)
unexport = getVarFlag(var, "unexport", d)
func = getVarFlag(var, "func", d)
if not all and not export and not unexport and not func:
return 0
try:
if all:
oval = getVar(var, d, 0)
val = getVar(var, d, 1)
except KeyboardInterrupt:
raise
except:
excname = str(sys.exc_info()[0])
if excname == "bb.build.FuncFailed":
raise
o.write('# expansion of %s threw %s\n' % (var, excname))
return 0
if all:
o.write('# %s=%s\n' % (var, oval))
if type(val) is not types.StringType:
return 0
if (var.find("-") != -1 or var.find(".") != -1 or var.find('{') != -1 or var.find('}') != -1 or var.find('+') != -1) and not all:
return 0
varExpanded = expand(var, d)
if unexport:
o.write('unset %s\n' % varExpanded)
return 1
val.rstrip()
if not val:
return 0
if func:
# NOTE: should probably check for unbalanced {} within the var
o.write("%s() {\n%s\n}\n" % (varExpanded, val))
return 1
if export:
o.write('export ')
# if we're going to output this within doublequotes,
# to a shell, we need to escape the quotes in the var
alter = re.sub('"', '\\"', val.strip())
o.write('%s="%s"\n' % (varExpanded, alter))
return 1
def emit_env(o=sys.__stdout__, d = init(), all=False):
"""Emits all items in the data store in a format such that it can be sourced by a shell."""
env = keys(d)
for e in env:
if getVarFlag(e, "func", d):
continue
emit_var(e, o, d, all) and o.write('\n')
for e in env:
if not getVarFlag(e, "func", d):
continue
emit_var(e, o, d) and o.write('\n')
def update_data(d):
"""Modifies the environment vars according to local overrides and commands.
Examples:
Appending to a variable:
>>> d = init()
>>> setVar('TEST', 'this is a', d)
>>> setVar('TEST_append', ' test', d)
>>> setVar('TEST_append', ' of the emergency broadcast system.', d)
>>> update_data(d)
>>> print getVar('TEST', d)
this is a test of the emergency broadcast system.
Prepending to a variable:
>>> setVar('TEST', 'virtual/libc', d)
>>> setVar('TEST_prepend', 'virtual/tmake ', d)
>>> setVar('TEST_prepend', 'virtual/patcher ', d)
>>> update_data(d)
>>> print getVar('TEST', d)
virtual/patcher virtual/tmake virtual/libc
Overrides:
>>> setVar('TEST_arm', 'target', d)
>>> setVar('TEST_ramses', 'machine', d)
>>> setVar('TEST_local', 'local', d)
>>> setVar('OVERRIDES', 'arm', d)
>>> setVar('TEST', 'original', d)
>>> update_data(d)
>>> print getVar('TEST', d)
target
>>> setVar('OVERRIDES', 'arm:ramses:local', d)
>>> setVar('TEST', 'original', d)
>>> update_data(d)
>>> print getVar('TEST', d)
local
CopyMonster:
>>> e = d.createCopy()
>>> setVar('TEST_foo', 'foo', e)
>>> update_data(e)
>>> print getVar('TEST', e)
local
>>> setVar('OVERRIDES', 'arm:ramses:local:foo', e)
>>> update_data(e)
>>> print getVar('TEST', e)
foo
>>> f = d.createCopy()
>>> setVar('TEST_moo', 'something', f)
>>> setVar('OVERRIDES', 'moo:arm:ramses:local:foo', e)
>>> update_data(e)
>>> print getVar('TEST', e)
foo
>>> h = init()
>>> setVar('SRC_URI', 'file://append.foo;patch=1 ', h)
>>> g = h.createCopy()
>>> setVar('SRC_URI_append_arm', 'file://other.foo;patch=1', g)
>>> setVar('OVERRIDES', 'arm:moo', g)
>>> update_data(g)
>>> print getVar('SRC_URI', g)
file://append.foo;patch=1 file://other.foo;patch=1
"""
bb.msg.debug(2, bb.msg.domain.Data, "update_data()")
# now ask the cookie monster for help
#print "Cookie Monster"
#print "Append/Prepend %s" % d._special_values
#print "Overrides %s" % d._seen_overrides
overrides = (getVar('OVERRIDES', d, 1) or "").split(':') or []
#
# Well let us see what breaks here. We used to iterate
# over each variable and apply the override and then
# do the line expanding.
# If we have bad luck - which we will have - the keys
# where in some order that is so important for this
# method which we don't have anymore.
# Anyway we will fix that and write test cases this
# time.
#
# First we apply all overrides
# Then we will handle _append and _prepend
#
for o in overrides:
# calculate '_'+override
l = len(o)+1
# see if one should even try
if not d._seen_overrides.has_key(o):
continue
vars = d._seen_overrides[o]
for var in vars:
name = var[:-l]
try:
d[name] = d[var]
except:
bb.msg.note(1, bb.msg.domain.Data, "Untracked delVar")
# now on to the appends and prepends
if d._special_values.has_key('_append'):
appends = d._special_values['_append'] or []
for append in appends:
for (a, o) in getVarFlag(append, '_append', d) or []:
# maybe the OVERRIDE was not yet added so keep the append
if (o and o in overrides) or not o:
delVarFlag(append, '_append', d)
if o and not o in overrides:
continue
sval = getVar(append,d) or ""
sval+=a
setVar(append, sval, d)
if d._special_values.has_key('_prepend'):
prepends = d._special_values['_prepend'] or []
for prepend in prepends:
for (a, o) in getVarFlag(prepend, '_prepend', d) or []:
# maybe the OVERRIDE was not yet added so keep the prepend
if (o and o in overrides) or not o:
delVarFlag(prepend, '_prepend', d)
if o and not o in overrides:
continue
sval = a + (getVar(prepend,d) or "")
setVar(prepend, sval, d)
def inherits_class(klass, d):
val = getVar('__inherit_cache', d) or []
if os.path.join('classes', '%s.bbclass' % klass) in val:
return True
return False
def _test():
"""Start a doctest run on this module"""
import doctest
import bb
from bb import data
bb.msg.set_debug_level(0)
doctest.testmod(data)
if __name__ == "__main__":
_test()

View File

@@ -0,0 +1,288 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake Smart Dictionary Implementation
Functions for interacting with the data structure used by the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2004, 2005 Seb Frankengul
# Copyright (C) 2005, 2006 Holger Hans Peter Freyther
# Copyright (C) 2005 Uli Luckas
# Copyright (C) 2005 ROAD GmbH
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import copy, os, re, sys, time, types
import bb
from bb import utils, methodpool
from COW import COWDictBase
from new import classobj
__setvar_keyword__ = ["_append","_prepend"]
__setvar_regexp__ = re.compile('(?P<base>.*?)(?P<keyword>_append|_prepend)(_(?P<add>.*))?')
__expand_var_regexp__ = re.compile(r"\${[^{}]+}")
__expand_python_regexp__ = re.compile(r"\${@.+?}")
class DataSmart:
def __init__(self, special = COWDictBase.copy(), seen = COWDictBase.copy() ):
self.dict = {}
# cookie monster tribute
self._special_values = special
self._seen_overrides = seen
self.expand_cache = {}
def expand(self,s, varname):
def var_sub(match):
key = match.group()[2:-1]
if varname and key:
if varname == key:
raise Exception("variable %s references itself!" % varname)
var = self.getVar(key, 1)
if var is not None:
return var
else:
return match.group()
def python_sub(match):
import bb
code = match.group()[3:-1]
locals()['d'] = self
s = eval(code)
if type(s) == types.IntType: s = str(s)
return s
if type(s) is not types.StringType: # sanity check
return s
if varname and varname in self.expand_cache:
return self.expand_cache[varname]
while s.find('${') != -1:
olds = s
try:
s = __expand_var_regexp__.sub(var_sub, s)
s = __expand_python_regexp__.sub(python_sub, s)
if s == olds: break
if type(s) is not types.StringType: # sanity check
bb.msg.error(bb.msg.domain.Data, 'expansion of %s returned non-string %s' % (olds, s))
except KeyboardInterrupt:
raise
except:
bb.msg.note(1, bb.msg.domain.Data, "%s:%s while evaluating:\n%s" % (sys.exc_info()[0], sys.exc_info()[1], s))
raise
if varname:
self.expand_cache[varname] = s
return s
def initVar(self, var):
self.expand_cache = {}
if not var in self.dict:
self.dict[var] = {}
def _findVar(self,var):
_dest = self.dict
while (_dest and var not in _dest):
if not "_data" in _dest:
_dest = None
break
_dest = _dest["_data"]
if _dest and var in _dest:
return _dest[var]
return None
def _makeShadowCopy(self, var):
if var in self.dict:
return
local_var = self._findVar(var)
if local_var:
self.dict[var] = copy.copy(local_var)
else:
self.initVar(var)
def setVar(self,var,value):
self.expand_cache = {}
match = __setvar_regexp__.match(var)
if match and match.group("keyword") in __setvar_keyword__:
base = match.group('base')
keyword = match.group("keyword")
override = match.group('add')
l = self.getVarFlag(base, keyword) or []
l.append([value, override])
self.setVarFlag(base, keyword, l)
# todo make sure keyword is not __doc__ or __module__
# pay the cookie monster
try:
self._special_values[keyword].add( base )
except:
self._special_values[keyword] = set()
self._special_values[keyword].add( base )
return
if not var in self.dict:
self._makeShadowCopy(var)
# more cookies for the cookie monster
if '_' in var:
override = var[var.rfind('_')+1:]
if not self._seen_overrides.has_key(override):
self._seen_overrides[override] = set()
self._seen_overrides[override].add( var )
# setting var
self.dict[var]["content"] = value
def getVar(self,var,exp):
value = self.getVarFlag(var,"content")
if exp and value:
return self.expand(value,var)
return value
def renameVar(self, key, newkey):
"""
Rename the variable key to newkey
"""
val = self.getVar(key, 0)
if val is None:
return
self.setVar(newkey, val)
for i in ('_append', '_prepend'):
dest = self.getVarFlag(newkey, i) or []
src = self.getVarFlag(key, i) or []
dest.extend(src)
self.setVarFlag(newkey, i, dest)
if self._special_values.has_key(i) and key in self._special_values[i]:
self._special_values[i].remove(key)
self._special_values[i].add(newkey)
self.delVar(key)
def delVar(self,var):
self.expand_cache = {}
self.dict[var] = {}
def setVarFlag(self,var,flag,flagvalue):
if not var in self.dict:
self._makeShadowCopy(var)
self.dict[var][flag] = flagvalue
def getVarFlag(self,var,flag):
local_var = self._findVar(var)
if local_var:
if flag in local_var:
return copy.copy(local_var[flag])
return None
def delVarFlag(self,var,flag):
local_var = self._findVar(var)
if not local_var:
return
if not var in self.dict:
self._makeShadowCopy(var)
if var in self.dict and flag in self.dict[var]:
del self.dict[var][flag]
def setVarFlags(self,var,flags):
if not var in self.dict:
self._makeShadowCopy(var)
for i in flags.keys():
if i == "content":
continue
self.dict[var][i] = flags[i]
def getVarFlags(self,var):
local_var = self._findVar(var)
flags = {}
if local_var:
for i in local_var.keys():
if i == "content":
continue
flags[i] = local_var[i]
if len(flags) == 0:
return None
return flags
def delVarFlags(self,var):
if not var in self.dict:
self._makeShadowCopy(var)
if var in self.dict:
content = None
# try to save the content
if "content" in self.dict[var]:
content = self.dict[var]["content"]
self.dict[var] = {}
self.dict[var]["content"] = content
else:
del self.dict[var]
def createCopy(self):
"""
Create a copy of self by setting _data to self
"""
# we really want this to be a DataSmart...
data = DataSmart(seen=self._seen_overrides.copy(), special=self._special_values.copy())
data.dict["_data"] = self.dict
return data
# Dictionary Methods
def keys(self):
def _keys(d, mykey):
if "_data" in d:
_keys(d["_data"],mykey)
for key in d.keys():
if key != "_data":
mykey[key] = None
keytab = {}
_keys(self.dict,keytab)
return keytab.keys()
def __getitem__(self,item):
#print "Warning deprecated"
return self.getVar(item, False)
def __setitem__(self,var,data):
#print "Warning deprecated"
self.setVar(var,data)

302
bitbake-dev/lib/bb/event.py Normal file
View File

@@ -0,0 +1,302 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Event' implementation
Classes and functions for manipulating 'events' in the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os, re
import bb.utils
# This is the pid for which we should generate the event. This is set when
# the runqueue forks off.
worker_pid = 0
class Event:
"""Base class for events"""
type = "Event"
def __init__(self, d):
self._data = d
self.pid = worker_pid
def getData(self):
return self._data
def setData(self, data):
self._data = data
data = property(getData, setData, None, "data property")
NotHandled = 0
Handled = 1
Registered = 10
AlreadyRegistered = 14
# Internal
_handlers = {}
_ui_handlers = {}
_ui_handler_seq = 0
def fire(event):
"""Fire off an Event"""
for handler in _handlers:
h = _handlers[handler]
if type(h).__name__ == "code":
exec(h)
tmpHandler(event)
else:
h(event)
# Remove the event data elements for UI handlers - too much data otherwise
# They can request data if they need it
event.data = None
event._data = None
errors = []
for h in _ui_handlers:
#print "Sending event %s" % event
classid = "%s.%s" % (event.__class__.__module__, event.__class__.__name__)
try:
_ui_handlers[h].event.send((classid, event))
except:
errors.append(h)
for h in errors:
del _ui_handlers[h]
def register(name, handler):
"""Register an Event handler"""
# already registered
if name in _handlers:
return AlreadyRegistered
if handler is not None:
# handle string containing python code
if type(handler).__name__ == "str":
tmp = "def tmpHandler(e):\n%s" % handler
comp = bb.utils.better_compile(tmp, "tmpHandler(e)", "bb.event._registerCode")
_handlers[name] = comp
else:
_handlers[name] = handler
return Registered
def remove(name, handler):
"""Remove an Event handler"""
_handlers.pop(name)
def register_UIHhandler(handler):
bb.event._ui_handler_seq = bb.event._ui_handler_seq + 1
_ui_handlers[_ui_handler_seq] = handler
return _ui_handler_seq
def unregister_UIHhandler(handlerNum):
if handlerNum in _ui_handlers:
del _ui_handlers[handlerNum]
return
def getName(e):
"""Returns the name of a class or class instance"""
if getattr(e, "__name__", None) == None:
return e.__class__.__name__
else:
return e.__name__
class ConfigParsed(Event):
"""Configuration Parsing Complete"""
class RecipeParsed(Event):
""" Recipe Parsing Complete """
def __init__(self, fn, d):
self.fn = fn
Event.__init__(self, d)
class StampUpdate(Event):
"""Trigger for any adjustment of the stamp files to happen"""
def __init__(self, targets, stampfns, d):
self._targets = targets
self._stampfns = stampfns
Event.__init__(self, d)
def getStampPrefix(self):
return self._stampfns
def getTargets(self):
return self._targets
stampPrefix = property(getStampPrefix)
targets = property(getTargets)
class PkgBase(Event):
"""Base class for package events"""
def __init__(self, t, d):
self._pkg = t
Event.__init__(self, d)
self._message = "package %s: %s" % (bb.data.getVar("P", d, 1), getName(self)[3:])
def getPkg(self):
return self._pkg
def setPkg(self, pkg):
self._pkg = pkg
pkg = property(getPkg, setPkg, None, "pkg property")
class BuildBase(Event):
"""Base class for bbmake run events"""
def __init__(self, n, p, c, failures = 0):
self._name = n
self._pkgs = p
Event.__init__(self, c)
self._failures = failures
def getPkgs(self):
return self._pkgs
def setPkgs(self, pkgs):
self._pkgs = pkgs
def getName(self):
return self._name
def setName(self, name):
self._name = name
def getCfg(self):
return self.data
def setCfg(self, cfg):
self.data = cfg
def getFailures(self):
"""
Return the number of failed packages
"""
return self._failures
pkgs = property(getPkgs, setPkgs, None, "pkgs property")
name = property(getName, setName, None, "name property")
cfg = property(getCfg, setCfg, None, "cfg property")
class DepBase(PkgBase):
"""Base class for dependency events"""
def __init__(self, t, data, d):
self._dep = d
PkgBase.__init__(self, t, data)
def getDep(self):
return self._dep
def setDep(self, dep):
self._dep = dep
dep = property(getDep, setDep, None, "dep property")
class BuildStarted(BuildBase):
"""bbmake build run started"""
class BuildCompleted(BuildBase):
"""bbmake build run completed"""
class UnsatisfiedDep(DepBase):
"""Unsatisfied Dependency"""
class RecursiveDep(DepBase):
"""Recursive Dependency"""
class NoProvider(Event):
"""No Provider for an Event"""
def __init__(self, item, data, runtime=False):
Event.__init__(self, data)
self._item = item
self._runtime = runtime
def getItem(self):
return self._item
def isRuntime(self):
return self._runtime
class MultipleProviders(Event):
"""Multiple Providers"""
def __init__(self, item, candidates, data, runtime = False):
Event.__init__(self, data)
self._item = item
self._candidates = candidates
self._is_runtime = runtime
def isRuntime(self):
"""
Is this a runtime issue?
"""
return self._is_runtime
def getItem(self):
"""
The name for the to be build item
"""
return self._item
def getCandidates(self):
"""
Get the possible Candidates for a PROVIDER.
"""
return self._candidates
class ParseProgress(Event):
"""
Parsing Progress Event
"""
def __init__(self, d, cached, parsed, skipped, masked, errors, total):
Event.__init__(self, d)
self.cached = cached
self.parsed = parsed
self.skipped = skipped
self.masked = masked
self.errors = errors
self.sofar = cached + parsed + skipped
self.total = total
class DepTreeGenerated(Event):
"""
Event when a dependency tree has been generated
"""
def __init__(self, d, depgraph):
Event.__init__(self, d)
self._depgraph = depgraph

View File

@@ -0,0 +1,606 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
Classes for obtaining upstream sources for the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os, re
import bb
from bb import data
from bb import persist_data
class FetchError(Exception):
"""Exception raised when a download fails"""
class NoMethodError(Exception):
"""Exception raised when there is no method to obtain a supplied url or set of urls"""
class MissingParameterError(Exception):
"""Exception raised when a fetch method is missing a critical parameter in the url"""
class ParameterError(Exception):
"""Exception raised when a url cannot be proccessed due to invalid parameters."""
class MD5SumError(Exception):
"""Exception raised when a MD5SUM of a file does not match the expected one"""
class InvalidSRCREV(Exception):
"""Exception raised when an invalid SRCREV is encountered"""
def uri_replace(uri, uri_find, uri_replace, d):
# bb.msg.note(1, bb.msg.domain.Fetcher, "uri_replace: operating on %s" % uri)
if not uri or not uri_find or not uri_replace:
bb.msg.debug(1, bb.msg.domain.Fetcher, "uri_replace: passed an undefined value, not replacing")
uri_decoded = list(bb.decodeurl(uri))
uri_find_decoded = list(bb.decodeurl(uri_find))
uri_replace_decoded = list(bb.decodeurl(uri_replace))
result_decoded = ['','','','','',{}]
for i in uri_find_decoded:
loc = uri_find_decoded.index(i)
result_decoded[loc] = uri_decoded[loc]
import types
if type(i) == types.StringType:
if (re.match(i, uri_decoded[loc])):
result_decoded[loc] = re.sub(i, uri_replace_decoded[loc], uri_decoded[loc])
if uri_find_decoded.index(i) == 2:
if d:
localfn = bb.fetch.localpath(uri, d)
if localfn:
result_decoded[loc] = os.path.dirname(result_decoded[loc]) + "/" + os.path.basename(bb.fetch.localpath(uri, d))
# bb.msg.note(1, bb.msg.domain.Fetcher, "uri_replace: matching %s against %s and replacing with %s" % (i, uri_decoded[loc], uri_replace_decoded[loc]))
else:
# bb.msg.note(1, bb.msg.domain.Fetcher, "uri_replace: no match")
return uri
# else:
# for j in i.keys():
# FIXME: apply replacements against options
return bb.encodeurl(result_decoded)
methods = []
urldata_cache = {}
saved_headrevs = {}
def fetcher_init(d):
"""
Called to initilize the fetchers once the configuration data is known
Calls before this must not hit the cache.
"""
pd = persist_data.PersistData(d)
# When to drop SCM head revisions controled by user policy
srcrev_policy = bb.data.getVar('BB_SRCREV_POLICY', d, 1) or "clear"
if srcrev_policy == "cache":
bb.msg.debug(1, bb.msg.domain.Fetcher, "Keeping SRCREV cache due to cache policy of: %s" % srcrev_policy)
elif srcrev_policy == "clear":
bb.msg.debug(1, bb.msg.domain.Fetcher, "Clearing SRCREV cache due to cache policy of: %s" % srcrev_policy)
try:
bb.fetch.saved_headrevs = pd.getKeyValues("BB_URI_HEADREVS")
except:
pass
pd.delDomain("BB_URI_HEADREVS")
else:
bb.msg.fatal(bb.msg.domain.Fetcher, "Invalid SRCREV cache policy of: %s" % srcrev_policy)
# Make sure our domains exist
pd.addDomain("BB_URI_HEADREVS")
pd.addDomain("BB_URI_LOCALCOUNT")
def fetcher_compare_revisons(d):
"""
Compare the revisions in the persistant cache with current values and
return true/false on whether they've changed.
"""
pd = persist_data.PersistData(d)
data = pd.getKeyValues("BB_URI_HEADREVS")
data2 = bb.fetch.saved_headrevs
changed = False
for key in data:
if key not in data2 or data2[key] != data[key]:
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s changed" % key)
changed = True
return True
else:
bb.msg.debug(2, bb.msg.domain.Fetcher, "%s did not change" % key)
return False
# Function call order is usually:
# 1. init
# 2. go
# 3. localpaths
# localpath can be called at any time
def init(urls, d, setup = True):
urldata = {}
fn = bb.data.getVar('FILE', d, 1)
if fn in urldata_cache:
urldata = urldata_cache[fn]
for url in urls:
if url not in urldata:
urldata[url] = FetchData(url, d)
if setup:
for url in urldata:
if not urldata[url].setup:
urldata[url].setup_localpath(d)
urldata_cache[fn] = urldata
return urldata
def go(d):
"""
Fetch all urls
init must have previously been called
"""
urldata = init([], d, True)
for u in urldata:
ud = urldata[u]
m = ud.method
if ud.localfile:
if not m.forcefetch(u, ud, d) and os.path.exists(ud.md5):
# File already present along with md5 stamp file
# Touch md5 file to show activity
try:
os.utime(ud.md5, None)
except:
# Errors aren't fatal here
pass
continue
lf = bb.utils.lockfile(ud.lockfile)
if not m.forcefetch(u, ud, d) and os.path.exists(ud.md5):
# If someone else fetched this before we got the lock,
# notice and don't try again
try:
os.utime(ud.md5, None)
except:
# Errors aren't fatal here
pass
bb.utils.unlockfile(lf)
continue
m.go(u, ud, d)
if ud.localfile:
if not m.forcefetch(u, ud, d):
Fetch.write_md5sum(u, ud, d)
bb.utils.unlockfile(lf)
def checkstatus(d):
"""
Check all urls exist upstream
init must have previously been called
"""
urldata = init([], d, True)
for u in urldata:
ud = urldata[u]
m = ud.method
bb.msg.note(1, bb.msg.domain.Fetcher, "Testing URL %s" % u)
ret = m.checkstatus(u, ud, d)
if not ret:
bb.msg.fatal(bb.msg.domain.Fetcher, "URL %s doesn't work" % u)
def localpaths(d):
"""
Return a list of the local filenames, assuming successful fetch
"""
local = []
urldata = init([], d, True)
for u in urldata:
ud = urldata[u]
local.append(ud.localpath)
return local
srcrev_internal_call = False
def get_srcrev(d):
"""
Return the version string for the current package
(usually to be used as PV)
Most packages usually only have one SCM so we just pass on the call.
In the multi SCM case, we build a value based on SRCREV_FORMAT which must
have been set.
"""
#
# Ugly code alert. localpath in the fetchers will try to evaluate SRCREV which
# could translate into a call to here. If it does, we need to catch this
# and provide some way so it knows get_srcrev is active instead of being
# some number etc. hence the srcrev_internal_call tracking and the magic
# "SRCREVINACTION" return value.
#
# Neater solutions welcome!
#
if bb.fetch.srcrev_internal_call:
return "SRCREVINACTION"
scms = []
# Only call setup_localpath on URIs which suppports_srcrev()
urldata = init(bb.data.getVar('SRC_URI', d, 1).split(), d, False)
for u in urldata:
ud = urldata[u]
if ud.method.suppports_srcrev():
if not ud.setup:
ud.setup_localpath(d)
scms.append(u)
if len(scms) == 0:
bb.msg.error(bb.msg.domain.Fetcher, "SRCREV was used yet no valid SCM was found in SRC_URI")
raise ParameterError
bb.data.setVar('__BB_DONT_CACHE','1', d)
if len(scms) == 1:
return urldata[scms[0]].method.sortable_revision(scms[0], urldata[scms[0]], d)
#
# Mutiple SCMs are in SRC_URI so we resort to SRCREV_FORMAT
#
format = bb.data.getVar('SRCREV_FORMAT', d, 1)
if not format:
bb.msg.error(bb.msg.domain.Fetcher, "The SRCREV_FORMAT variable must be set when multiple SCMs are used.")
raise ParameterError
for scm in scms:
if 'name' in urldata[scm].parm:
name = urldata[scm].parm["name"]
rev = urldata[scm].method.sortable_revision(scm, urldata[scm], d)
format = format.replace(name, rev)
return format
def localpath(url, d, cache = True):
"""
Called from the parser with cache=False since the cache isn't ready
at this point. Also called from classed in OE e.g. patch.bbclass
"""
ud = init([url], d)
if ud[url].method:
return ud[url].localpath
return url
def runfetchcmd(cmd, d, quiet = False):
"""
Run cmd returning the command output
Raise an error if interrupted or cmd fails
Optionally echo command output to stdout
"""
# Need to export PATH as binary could be in metadata paths
# rather than host provided
# Also include some other variables.
# FIXME: Should really include all export varaiables?
exportvars = ['PATH', 'GIT_PROXY_COMMAND', 'GIT_PROXY_HOST', 'GIT_PROXY_PORT', 'GIT_CONFIG', 'http_proxy', 'ftp_proxy', 'SSH_AUTH_SOCK', 'SSH_AGENT_PID', 'HOME']
for var in exportvars:
val = data.getVar(var, d, True)
if val:
cmd = 'export ' + var + '=%s; %s' % (val, cmd)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % cmd)
# redirect stderr to stdout
stdout_handle = os.popen(cmd + " 2>&1", "r")
output = ""
while 1:
line = stdout_handle.readline()
if not line:
break
if not quiet:
print line,
output += line
status = stdout_handle.close() or 0
signal = status >> 8
exitstatus = status & 0xff
if signal:
raise FetchError("Fetch command %s failed with signal %s, output:\n%s" % (cmd, signal, output))
elif status != 0:
raise FetchError("Fetch command %s failed with exit code %s, output:\n%s" % (cmd, status, output))
return output
class FetchData(object):
"""
A class which represents the fetcher state for a given URI.
"""
def __init__(self, url, d):
self.localfile = ""
(self.type, self.host, self.path, self.user, self.pswd, self.parm) = bb.decodeurl(data.expand(url, d))
self.date = Fetch.getSRCDate(self, d)
self.url = url
if not self.user and "user" in self.parm:
self.user = self.parm["user"]
if not self.pswd and "pswd" in self.parm:
self.pswd = self.parm["pswd"]
self.setup = False
for m in methods:
if m.supports(url, self, d):
self.method = m
return
raise NoMethodError("Missing implementation for url %s" % url)
def setup_localpath(self, d):
self.setup = True
if "localpath" in self.parm:
# if user sets localpath for file, use it instead.
self.localpath = self.parm["localpath"]
else:
try:
bb.fetch.srcrev_internal_call = True
self.localpath = self.method.localpath(self.url, self, d)
finally:
bb.fetch.srcrev_internal_call = False
# We have to clear data's internal caches since the cached value of SRCREV is now wrong.
# Horrible...
bb.data.delVar("ISHOULDNEVEREXIST", d)
self.md5 = self.localpath + '.md5'
self.lockfile = self.localpath + '.lock'
class Fetch(object):
"""Base class for 'fetch'ing data"""
def __init__(self, urls = []):
self.urls = []
def supports(self, url, urldata, d):
"""
Check to see if this fetch class supports a given url.
"""
return 0
def localpath(self, url, urldata, d):
"""
Return the local filename of a given url assuming a successful fetch.
Can also setup variables in urldata for use in go (saving code duplication
and duplicate code execution)
"""
return url
def setUrls(self, urls):
self.__urls = urls
def getUrls(self):
return self.__urls
urls = property(getUrls, setUrls, None, "Urls property")
def forcefetch(self, url, urldata, d):
"""
Force a fetch, even if localpath exists?
"""
return False
def suppports_srcrev(self):
"""
The fetcher supports auto source revisions (SRCREV)
"""
return False
def go(self, url, urldata, d):
"""
Fetch urls
Assumes localpath was called first
"""
raise NoMethodError("Missing implementation for url")
def checkstatus(self, url, urldata, d):
"""
Check the status of a URL
Assumes localpath was called first
"""
bb.msg.note(1, bb.msg.domain.Fetcher, "URL %s could not be checked for status since no method exists." % url)
return True
def getSRCDate(urldata, d):
"""
Return the SRC Date for the component
d the bb.data module
"""
if "srcdate" in urldata.parm:
return urldata.parm['srcdate']
pn = data.getVar("PN", d, 1)
if pn:
return data.getVar("SRCDATE_%s" % pn, d, 1) or data.getVar("CVSDATE_%s" % pn, d, 1) or data.getVar("SRCDATE", d, 1) or data.getVar("CVSDATE", d, 1) or data.getVar("DATE", d, 1)
return data.getVar("SRCDATE", d, 1) or data.getVar("CVSDATE", d, 1) or data.getVar("DATE", d, 1)
getSRCDate = staticmethod(getSRCDate)
def srcrev_internal_helper(ud, d):
"""
Return:
a) a source revision if specified
b) True if auto srcrev is in action
c) False otherwise
"""
if 'rev' in ud.parm:
return ud.parm['rev']
if 'tag' in ud.parm:
return ud.parm['tag']
rev = None
if 'name' in ud.parm:
pn = data.getVar("PN", d, 1)
rev = data.getVar("SRCREV_pn-" + pn + "_" + ud.parm['name'], d, 1)
if not rev:
rev = data.getVar("SRCREV", d, 1)
if rev == "INVALID":
raise InvalidSRCREV("Please set SRCREV to a valid value")
if not rev:
return False
if rev is "SRCREVINACTION":
return True
return rev
srcrev_internal_helper = staticmethod(srcrev_internal_helper)
def try_mirror(d, tarfn):
"""
Try to use a mirrored version of the sources. We do this
to avoid massive loads on foreign cvs and svn servers.
This method will be used by the different fetcher
implementations.
d Is a bb.data instance
tarfn is the name of the tarball
"""
tarpath = os.path.join(data.getVar("DL_DIR", d, 1), tarfn)
if os.access(tarpath, os.R_OK):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists, skipping checkout." % tarfn)
return True
pn = data.getVar('PN', d, True)
src_tarball_stash = None
if pn:
src_tarball_stash = (data.getVar('SRC_TARBALL_STASH_%s' % pn, d, True) or data.getVar('CVS_TARBALL_STASH_%s' % pn, d, True) or data.getVar('SRC_TARBALL_STASH', d, True) or data.getVar('CVS_TARBALL_STASH', d, True) or "").split()
ld = d.createCopy()
for stash in src_tarball_stash:
url = stash + tarfn
try:
ud = FetchData(url, ld)
except bb.fetch.NoMethodError:
bb.msg.debug(1, bb.msg.domain.Fetcher, "No method for %s" % url)
continue
ud.setup_localpath(ld)
try:
ud.method.go(url, ud, ld)
return True
except (bb.fetch.MissingParameterError,
bb.fetch.FetchError,
bb.fetch.MD5SumError):
import sys
(type, value, traceback) = sys.exc_info()
bb.msg.debug(2, bb.msg.domain.Fetcher, "Tarball stash fetch failure: %s" % value)
return False
try_mirror = staticmethod(try_mirror)
def verify_md5sum(ud, got_sum):
"""
Verify the md5sum we wanted with the one we got
"""
wanted_sum = None
if 'md5sum' in ud.parm:
wanted_sum = ud.parm['md5sum']
if not wanted_sum:
return True
return wanted_sum == got_sum
verify_md5sum = staticmethod(verify_md5sum)
def write_md5sum(url, ud, d):
md5data = bb.utils.md5_file(ud.localpath)
# verify the md5sum
if not Fetch.verify_md5sum(ud, md5data):
raise MD5SumError(url)
md5out = file(ud.md5, 'w')
md5out.write(md5data)
md5out.close()
write_md5sum = staticmethod(write_md5sum)
def latest_revision(self, url, ud, d):
"""
Look in the cache for the latest revision, if not present ask the SCM.
"""
if not hasattr(self, "_latest_revision"):
raise ParameterError
pd = persist_data.PersistData(d)
key = self.generate_revision_key(url, ud, d)
rev = pd.getValue("BB_URI_HEADREVS", key)
if rev != None:
return str(rev)
rev = self._latest_revision(url, ud, d)
pd.setValue("BB_URI_HEADREVS", key, rev)
return rev
def sortable_revision(self, url, ud, d):
"""
"""
has_sortable = hasattr(self, "_sortable_revision")
if has_sortable:
return self._sortable_revision(url, ud, d)
pd = persist_data.PersistData(d)
key = self.generate_revision_key(url, ud, d)
latest_rev = self._build_revision(url, ud, d)
last_rev = pd.getValue("BB_URI_LOCALCOUNT", key + "_rev")
count = pd.getValue("BB_URI_LOCALCOUNT", key + "_count")
if last_rev == latest_rev:
return str(count + "+" + latest_rev)
if count is None:
count = "0"
else:
count = str(int(count) + 1)
pd.setValue("BB_URI_LOCALCOUNT", key + "_rev", latest_rev)
pd.setValue("BB_URI_LOCALCOUNT", key + "_count", count)
return str(count + "+" + latest_rev)
def generate_revision_key(self, url, ud, d):
key = self._revision_key(url, ud, d)
return "%s-%s" % (key, bb.data.getVar("PN", d, True) or "")
import cvs
import git
import local
import svn
import wget
import svk
import ssh
import perforce
import bzr
import hg
import osc
methods.append(local.Local())
methods.append(wget.Wget())
methods.append(svn.Svn())
methods.append(git.Git())
methods.append(cvs.Cvs())
methods.append(svk.Svk())
methods.append(ssh.SSH())
methods.append(perforce.Perforce())
methods.append(bzr.Bzr())
methods.append(hg.Hg())
methods.append(osc.Osc())

View File

@@ -0,0 +1,153 @@
"""
BitBake 'Fetch' implementation for bzr.
"""
# Copyright (C) 2007 Ross Burton
# Copyright (C) 2007 Richard Purdie
#
# Classes for obtaining upstream sources for the
# BitBake build tools.
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import runfetchcmd
class Bzr(Fetch):
def supports(self, url, ud, d):
return ud.type in ['bzr']
def localpath (self, url, ud, d):
# Create paths to bzr checkouts
relpath = ud.path
if relpath.startswith('/'):
# Remove leading slash as os.path.join can't cope
relpath = relpath[1:]
ud.pkgdir = os.path.join(data.expand('${BZRDIR}', d), ud.host, relpath)
revision = Fetch.srcrev_internal_helper(ud, d)
if revision is True:
ud.revision = self.latest_revision(url, ud, d)
elif revision:
ud.revision = revision
if not ud.revision:
ud.revision = self.latest_revision(url, ud, d)
ud.localfile = data.expand('bzr_%s_%s_%s.tar.gz' % (ud.host, ud.path.replace('/', '.'), ud.revision), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def _buildbzrcommand(self, ud, d, command):
"""
Build up an bzr commandline based on ud
command is "fetch", "update", "revno"
"""
basecmd = data.expand('${FETCHCMD_bzr}', d)
proto = "http"
if "proto" in ud.parm:
proto = ud.parm["proto"]
bzrroot = ud.host + ud.path
options = []
if command is "revno":
bzrcmd = "%s revno %s %s://%s" % (basecmd, " ".join(options), proto, bzrroot)
else:
if ud.revision:
options.append("-r %s" % ud.revision)
if command is "fetch":
bzrcmd = "%s co %s %s://%s" % (basecmd, " ".join(options), proto, bzrroot)
elif command is "update":
bzrcmd = "%s pull %s --overwrite" % (basecmd, " ".join(options))
else:
raise FetchError("Invalid bzr command %s" % command)
return bzrcmd
def go(self, loc, ud, d):
"""Fetch url"""
# try to use the tarball stash
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping bzr checkout." % ud.localpath)
return
if os.access(os.path.join(ud.pkgdir, os.path.basename(ud.pkgdir), '.bzr'), os.R_OK):
bzrcmd = self._buildbzrcommand(ud, d, "update")
bb.msg.debug(1, bb.msg.domain.Fetcher, "BZR Update %s" % loc)
os.chdir(os.path.join (ud.pkgdir, os.path.basename(ud.path)))
runfetchcmd(bzrcmd, d)
else:
os.system("rm -rf %s" % os.path.join(ud.pkgdir, os.path.basename(ud.pkgdir)))
bzrcmd = self._buildbzrcommand(ud, d, "fetch")
bb.msg.debug(1, bb.msg.domain.Fetcher, "BZR Checkout %s" % loc)
bb.mkdirhier(ud.pkgdir)
os.chdir(ud.pkgdir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % bzrcmd)
runfetchcmd(bzrcmd, d)
os.chdir(ud.pkgdir)
# tar them up to a defined filename
try:
runfetchcmd("tar -czf %s %s" % (ud.localpath, os.path.basename(ud.pkgdir)), d)
except:
t, v, tb = sys.exc_info()
try:
os.unlink(ud.localpath)
except OSError:
pass
raise t, v, tb
def suppports_srcrev(self):
return True
def _revision_key(self, url, ud, d):
"""
Return a unique key for the url
"""
return "bzr:" + ud.pkgdir
def _latest_revision(self, url, ud, d):
"""
Return the latest upstream revision number
"""
bb.msg.debug(2, bb.msg.domain.Fetcher, "BZR fetcher hitting network for %s" % url)
output = runfetchcmd(self._buildbzrcommand(ud, d, "revno"), d, True)
return output.strip()
def _sortable_revision(self, url, ud, d):
"""
Return a sortable revision number which in our case is the revision number
"""
return self._build_revision(url, ud, d)
def _build_revision(self, url, ud, d):
return ud.revision

View File

@@ -0,0 +1,182 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
Classes for obtaining upstream sources for the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#Based on functions from the base bb module, Copyright 2003 Holger Schurig
#
import os
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import MissingParameterError
class Cvs(Fetch):
"""
Class to fetch a module or modules from cvs repositories
"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with cvs.
"""
return ud.type in ['cvs', 'pserver']
def localpath(self, url, ud, d):
if not "module" in ud.parm:
raise MissingParameterError("cvs method needs a 'module' parameter")
ud.module = ud.parm["module"]
ud.tag = ""
if 'tag' in ud.parm:
ud.tag = ud.parm['tag']
# Override the default date in certain cases
if 'date' in ud.parm:
ud.date = ud.parm['date']
elif ud.tag:
ud.date = ""
norecurse = ''
if 'norecurse' in ud.parm:
norecurse = '_norecurse'
fullpath = ''
if 'fullpath' in ud.parm:
fullpath = '_fullpath'
ud.localfile = data.expand('%s_%s_%s_%s%s%s.tar.gz' % (ud.module.replace('/', '.'), ud.host, ud.tag, ud.date, norecurse, fullpath), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def forcefetch(self, url, ud, d):
if (ud.date == "now"):
return True
return False
def go(self, loc, ud, d):
# try to use the tarball stash
if not self.forcefetch(loc, ud, d) and Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping cvs checkout." % ud.localpath)
return
method = "pserver"
if "method" in ud.parm:
method = ud.parm["method"]
localdir = ud.module
if "localdir" in ud.parm:
localdir = ud.parm["localdir"]
cvs_port = ""
if "port" in ud.parm:
cvs_port = ud.parm["port"]
cvs_rsh = None
if method == "ext":
if "rsh" in ud.parm:
cvs_rsh = ud.parm["rsh"]
if method == "dir":
cvsroot = ud.path
else:
cvsroot = ":" + method
cvsproxyhost = data.getVar('CVS_PROXY_HOST', d, True)
if cvsproxyhost:
cvsroot += ";proxy=" + cvsproxyhost
cvsproxyport = data.getVar('CVS_PROXY_PORT', d, True)
if cvsproxyport:
cvsroot += ";proxyport=" + cvsproxyport
cvsroot += ":" + ud.user
if ud.pswd:
cvsroot += ":" + ud.pswd
cvsroot += "@" + ud.host + ":" + cvs_port + ud.path
options = []
if 'norecurse' in ud.parm:
options.append("-l")
if ud.date:
# treat YYYYMMDDHHMM specially for CVS
if len(ud.date) == 12:
options.append("-D \"%s %s:%s UTC\"" % (ud.date[0:8], ud.date[8:10], ud.date[10:12]))
else:
options.append("-D \"%s UTC\"" % ud.date)
if ud.tag:
options.append("-r %s" % ud.tag)
localdata = data.createCopy(d)
data.setVar('OVERRIDES', "cvs:%s" % data.getVar('OVERRIDES', localdata), localdata)
data.update_data(localdata)
data.setVar('CVSROOT', cvsroot, localdata)
data.setVar('CVSCOOPTS', " ".join(options), localdata)
data.setVar('CVSMODULE', ud.module, localdata)
cvscmd = data.getVar('FETCHCOMMAND', localdata, 1)
cvsupdatecmd = data.getVar('UPDATECOMMAND', localdata, 1)
if cvs_rsh:
cvscmd = "CVS_RSH=\"%s\" %s" % (cvs_rsh, cvscmd)
cvsupdatecmd = "CVS_RSH=\"%s\" %s" % (cvs_rsh, cvsupdatecmd)
# create module directory
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: checking for module directory")
pkg = data.expand('${PN}', d)
pkgdir = os.path.join(data.expand('${CVSDIR}', localdata), pkg)
moddir = os.path.join(pkgdir,localdir)
if os.access(os.path.join(moddir,'CVS'), os.R_OK):
bb.msg.note(1, bb.msg.domain.Fetcher, "Update " + loc)
# update sources there
os.chdir(moddir)
myret = os.system(cvsupdatecmd)
else:
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
# check out sources there
bb.mkdirhier(pkgdir)
os.chdir(pkgdir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % cvscmd)
myret = os.system(cvscmd)
if myret != 0 or not os.access(moddir, os.R_OK):
try:
os.rmdir(moddir)
except OSError:
pass
raise FetchError(ud.module)
# tar them up to a defined filename
if 'fullpath' in ud.parm:
os.chdir(pkgdir)
myret = os.system("tar -czf %s %s" % (ud.localpath, localdir))
else:
os.chdir(moddir)
os.chdir('..')
myret = os.system("tar -czf %s %s" % (ud.localpath, os.path.basename(moddir)))
if myret != 0:
try:
os.unlink(ud.localpath)
except OSError:
pass
raise FetchError(ud.module)

View File

@@ -0,0 +1,188 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' git implementation
"""
#Copyright (C) 2005 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import runfetchcmd
class Git(Fetch):
"""Class to fetch a module or modules from git repositories"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with git.
"""
return ud.type in ['git']
def localpath(self, url, ud, d):
if 'protocol' in ud.parm:
ud.proto = ud.parm['protocol']
elif not ud.host:
ud.proto = 'file'
else:
ud.proto = "rsync"
ud.branch = ud.parm.get("branch", "master")
tag = Fetch.srcrev_internal_helper(ud, d)
if tag is True:
ud.tag = self.latest_revision(url, ud, d)
elif tag:
ud.tag = tag
if not ud.tag or ud.tag == "master":
ud.tag = self.latest_revision(url, ud, d)
ud.localfile = data.expand('git_%s%s_%s.tar.gz' % (ud.host, ud.path.replace('/', '.'), ud.tag), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def go(self, loc, ud, d):
"""Fetch url"""
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists (or was stashed). Skipping git checkout." % ud.localpath)
return
if ud.user:
username = ud.user + '@'
else:
username = ""
gitsrcname = '%s%s' % (ud.host, ud.path.replace('/', '.'))
repofilename = 'git_%s.tar.gz' % (gitsrcname)
repofile = os.path.join(data.getVar("DL_DIR", d, 1), repofilename)
repodir = os.path.join(data.expand('${GITDIR}', d), gitsrcname)
coname = '%s' % (ud.tag)
codir = os.path.join(repodir, coname)
if not os.path.exists(repodir):
if Fetch.try_mirror(d, repofilename):
bb.mkdirhier(repodir)
os.chdir(repodir)
runfetchcmd("tar -xzf %s" % (repofile), d)
else:
runfetchcmd("git clone -n %s://%s%s%s %s" % (ud.proto, username, ud.host, ud.path, repodir), d)
os.chdir(repodir)
# Remove all but the .git directory
if not self._contains_ref(ud.tag, d):
runfetchcmd("rm * -Rf", d)
runfetchcmd("git fetch %s://%s%s%s %s" % (ud.proto, username, ud.host, ud.path, ud.branch), d)
runfetchcmd("git fetch --tags %s://%s%s%s" % (ud.proto, username, ud.host, ud.path), d)
runfetchcmd("git prune-packed", d)
runfetchcmd("git pack-redundant --all | xargs -r rm", d)
os.chdir(repodir)
mirror_tarballs = data.getVar("BB_GENERATE_MIRROR_TARBALLS", d, True)
if mirror_tarballs != "0":
bb.msg.note(1, bb.msg.domain.Fetcher, "Creating tarball of git repository")
runfetchcmd("tar -czf %s %s" % (repofile, os.path.join(".", ".git", "*") ), d)
if os.path.exists(codir):
bb.utils.prunedir(codir)
bb.mkdirhier(codir)
os.chdir(repodir)
runfetchcmd("git read-tree %s" % (ud.tag), d)
runfetchcmd("git checkout-index -q -f --prefix=%s -a" % (os.path.join(codir, "git", "")), d)
os.chdir(codir)
bb.msg.note(1, bb.msg.domain.Fetcher, "Creating tarball of git checkout")
runfetchcmd("tar -czf %s %s" % (ud.localpath, os.path.join(".", "*") ), d)
os.chdir(repodir)
bb.utils.prunedir(codir)
def suppports_srcrev(self):
return True
def _contains_ref(self, tag, d):
output = runfetchcmd("git log --pretty=oneline -n 1 %s -- 2> /dev/null | wc -l" % tag, d, quiet=True)
return output.split()[0] != "0"
def _revision_key(self, url, ud, d):
"""
Return a unique key for the url
"""
return "git:" + ud.host + ud.path.replace('/', '.')
def _latest_revision(self, url, ud, d):
"""
Compute the HEAD revision for the url
"""
if ud.user:
username = ud.user + '@'
else:
username = ""
output = runfetchcmd("git ls-remote %s://%s%s%s %s" % (ud.proto, username, ud.host, ud.path, ud.branch), d, True)
return output.split()[0]
def _build_revision(self, url, ud, d):
return ud.tag
def _want_sortable_revision(self, url, ud, d):
return bb.data.getVar("BB_GIT_CLONE_FOR_SRCREV", d, True) or False
def _sortable_revision_disabled(self, url, ud, d):
"""
This is only called when _want_sortable_revision called true
We will have to get the updated revision.
"""
gitsrcname = '%s%s' % (ud.host, ud.path.replace('/', '.'))
repodir = os.path.join(data.expand('${GITDIR}', d), gitsrcname)
key = "GIT_CACHED_REVISION-%s-%s" % (gitsrcname, ud.tag)
if bb.data.getVar(key, d):
return bb.data.getVar(key, d)
# Runtime warning on wrongly configured sources
if ud.tag == "1":
bb.msg.error(1, bb.msg.domain.Fetcher, "SRCREV is '1'. This indicates a configuration error of %s" % url)
return "0+1"
cwd = os.getcwd()
# Check if we have the rev already
if not os.path.exists(repodir):
print "no repo"
self.go(None, ud, d)
os.chdir(repodir)
if not self._contains_ref(ud.tag, d):
self.go(None, ud, d)
output = runfetchcmd("git rev-list %s -- 2> /dev/null | wc -l" % ud.tag, d, quiet=True)
os.chdir(cwd)
sortable_revision = "%s+%s" % (output.split()[0], ud.tag)
bb.data.setVar(key, sortable_revision, d)
return sortable_revision

View File

@@ -0,0 +1,178 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementation for mercurial DRCS (hg).
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2004 Marcin Juszkiewicz
# Copyright (C) 2007 Robert Schuster
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import sys
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import MissingParameterError
from bb.fetch import runfetchcmd
class Hg(Fetch):
"""Class to fetch a from mercurial repositories"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with mercurial.
"""
return ud.type in ['hg']
def localpath(self, url, ud, d):
if not "module" in ud.parm:
raise MissingParameterError("hg method needs a 'module' parameter")
ud.module = ud.parm["module"]
# Create paths to mercurial checkouts
relpath = ud.path
if relpath.startswith('/'):
# Remove leading slash as os.path.join can't cope
relpath = relpath[1:]
ud.pkgdir = os.path.join(data.expand('${HGDIR}', d), ud.host, relpath)
ud.moddir = os.path.join(ud.pkgdir, ud.module)
if 'rev' in ud.parm:
ud.revision = ud.parm['rev']
else:
tag = Fetch.srcrev_internal_helper(ud, d)
if tag is True:
ud.revision = self.latest_revision(url, ud, d)
elif tag:
ud.revision = tag
else:
ud.revision = self.latest_revision(url, ud, d)
ud.localfile = data.expand('%s_%s_%s_%s.tar.gz' % (ud.module.replace('/', '.'), ud.host, ud.path.replace('/', '.'), ud.revision), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def _buildhgcommand(self, ud, d, command):
"""
Build up an hg commandline based on ud
command is "fetch", "update", "info"
"""
basecmd = data.expand('${FETCHCMD_hg}', d)
proto = "http"
if "proto" in ud.parm:
proto = ud.parm["proto"]
host = ud.host
if proto == "file":
host = "/"
ud.host = "localhost"
if not ud.user:
hgroot = host + ud.path
else:
hgroot = ud.user + "@" + host + ud.path
if command is "info":
return "%s identify -i %s://%s/%s" % (basecmd, proto, hgroot, ud.module)
options = [];
if ud.revision:
options.append("-r %s" % ud.revision)
if command is "fetch":
cmd = "%s clone %s %s://%s/%s %s" % (basecmd, " ".join(options), proto, hgroot, ud.module, ud.module)
elif command is "pull":
# do not pass options list; limiting pull to rev causes the local
# repo not to contain it and immediately following "update" command
# will crash
cmd = "%s pull" % (basecmd)
elif command is "update":
cmd = "%s update -C %s" % (basecmd, " ".join(options))
else:
raise FetchError("Invalid hg command %s" % command)
return cmd
def go(self, loc, ud, d):
"""Fetch url"""
# try to use the tarball stash
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping hg checkout." % ud.localpath)
return
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: checking for module directory '" + ud.moddir + "'")
if os.access(os.path.join(ud.moddir, '.hg'), os.R_OK):
updatecmd = self._buildhgcommand(ud, d, "pull")
bb.msg.note(1, bb.msg.domain.Fetcher, "Update " + loc)
# update sources there
os.chdir(ud.moddir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % updatecmd)
runfetchcmd(updatecmd, d)
else:
fetchcmd = self._buildhgcommand(ud, d, "fetch")
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
# check out sources there
bb.mkdirhier(ud.pkgdir)
os.chdir(ud.pkgdir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % fetchcmd)
runfetchcmd(fetchcmd, d)
# Even when we clone (fetch), we still need to update as hg's clone
# won't checkout the specified revision if its on a branch
updatecmd = self._buildhgcommand(ud, d, "update")
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % updatecmd)
runfetchcmd(updatecmd, d)
os.chdir(ud.pkgdir)
try:
runfetchcmd("tar -czf %s %s" % (ud.localpath, ud.module), d)
except:
t, v, tb = sys.exc_info()
try:
os.unlink(ud.localpath)
except OSError:
pass
raise t, v, tb
def suppports_srcrev(self):
return True
def _latest_revision(self, url, ud, d):
"""
Compute tip revision for the url
"""
output = runfetchcmd(self._buildhgcommand(ud, d, "info"), d)
return output.strip()
def _build_revision(self, url, ud, d):
return ud.revision
def _revision_key(self, url, ud, d):
"""
Return a unique key for the url
"""
return "hg:" + ud.moddir

View File

@@ -0,0 +1,72 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
Classes for obtaining upstream sources for the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import bb
from bb import data
from bb.fetch import Fetch
class Local(Fetch):
def supports(self, url, urldata, d):
"""
Check to see if a given url can be fetched with cvs.
"""
return urldata.type in ['file','patch']
def localpath(self, url, urldata, d):
"""
Return the local filename of a given url assuming a successful fetch.
"""
path = url.split("://")[1]
path = path.split(";")[0]
newpath = path
if path[0] != "/":
filespath = data.getVar('FILESPATH', d, 1)
if filespath:
newpath = bb.which(filespath, path)
if not newpath:
filesdir = data.getVar('FILESDIR', d, 1)
if filesdir:
newpath = os.path.join(filesdir, path)
# We don't set localfile as for this fetcher the file is already local!
return newpath
def go(self, url, urldata, d):
"""Fetch urls (no-op for Local method)"""
# no need to fetch local files, we'll deal with them in place.
return 1
def checkstatus(self, url, urldata, d):
"""
Check the status of the url
"""
if urldata.localpath.find("*") != -1:
bb.msg.note(1, bb.msg.domain.Fetcher, "URL %s looks like a glob and was therefore not checked." % url)
return True
if os.path.exists(urldata.localpath):
return True
return False

View File

@@ -0,0 +1,155 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
Bitbake "Fetch" implementation for osc (Opensuse build service client).
Based on the svn "Fetch" implementation.
"""
import os
import sys
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import MissingParameterError
from bb.fetch import runfetchcmd
class Osc(Fetch):
"""Class to fetch a module or modules from Opensuse build server
repositories."""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with osc.
"""
return ud.type in ['osc']
def localpath(self, url, ud, d):
if not "module" in ud.parm:
raise MissingParameterError("osc method needs a 'module' parameter.")
ud.module = ud.parm["module"]
# Create paths to osc checkouts
relpath = ud.path
if relpath.startswith('/'):
# Remove leading slash as os.path.join can't cope
relpath = relpath[1:]
ud.pkgdir = os.path.join(data.expand('${OSCDIR}', d), ud.host)
ud.moddir = os.path.join(ud.pkgdir, relpath, ud.module)
if 'rev' in ud.parm:
ud.revision = ud.parm['rev']
else:
pv = data.getVar("PV", d, 0)
rev = Fetch.srcrev_internal_helper(ud, d)
if rev and rev != True:
ud.revision = rev
else:
ud.revision = ""
ud.localfile = data.expand('%s_%s_%s.tar.gz' % (ud.module.replace('/', '.'), ud.path.replace('/', '.'), ud.revision), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def _buildosccommand(self, ud, d, command):
"""
Build up an ocs commandline based on ud
command is "fetch", "update", "info"
"""
basecmd = data.expand('${FETCHCMD_osc}', d)
proto = "ocs"
if "proto" in ud.parm:
proto = ud.parm["proto"]
options = []
config = "-c %s" % self.generate_config(ud, d)
if ud.revision:
options.append("-r %s" % ud.revision)
coroot = ud.path
if coroot.startswith('/'):
# Remove leading slash as os.path.join can't cope
coroot= coroot[1:]
if command is "fetch":
osccmd = "%s %s co %s/%s %s" % (basecmd, config, coroot, ud.module, " ".join(options))
elif command is "update":
osccmd = "%s %s up %s" % (basecmd, config, " ".join(options))
else:
raise FetchError("Invalid osc command %s" % command)
return osccmd
def go(self, loc, ud, d):
"""
Fetch url
"""
# Try to use the tarball stash
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping osc checkout." % ud.localpath)
return
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: checking for module directory '" + ud.moddir + "'")
if os.access(os.path.join(data.expand('${OSCDIR}', d), ud.path, ud.module), os.R_OK):
oscupdatecmd = self._buildosccommand(ud, d, "update")
bb.msg.note(1, bb.msg.domain.Fetcher, "Update "+ loc)
# update sources there
os.chdir(ud.moddir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % oscupdatecmd)
runfetchcmd(oscupdatecmd, d)
else:
oscfetchcmd = self._buildosccommand(ud, d, "fetch")
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
# check out sources there
bb.mkdirhier(ud.pkgdir)
os.chdir(ud.pkgdir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % oscfetchcmd)
runfetchcmd(oscfetchcmd, d)
os.chdir(os.path.join(ud.pkgdir + ud.path))
# tar them up to a defined filename
try:
runfetchcmd("tar -czf %s %s" % (ud.localpath, ud.module), d)
except:
t, v, tb = sys.exc_info()
try:
os.unlink(ud.localpath)
except OSError:
pass
raise t, v, tb
def supports_srcrev(self):
return False
def generate_config(self, ud, d):
"""
Generate a .oscrc to be used for this run.
"""
config_path = "%s/oscrc" % data.expand('${OSCDIR}', d)
if (os.path.exists(config_path)):
os.remove(config_path)
f = open(config_path, 'w')
f.write("[general]\n")
f.write("apisrv = %s\n" % ud.host)
f.write("scheme = http\n")
f.write("su-wrapper = su -c\n")
f.write("build-root = %s\n" % data.expand('${WORKDIR}', d))
f.write("urllist = http://moblin-obs.jf.intel.com:8888/build/%(project)s/%(repository)s/%(buildarch)s/:full/%(name)s.rpm\n")
f.write("extra-pkgs = gzip\n")
f.write("\n")
f.write("[%s]\n" % ud.host)
f.write("user = %s\n" % ud.parm["user"])
f.write("pass = %s\n" % ud.parm["pswd"])
f.close()
return config_path

View File

@@ -0,0 +1,214 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
Classes for obtaining upstream sources for the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
class Perforce(Fetch):
def supports(self, url, ud, d):
return ud.type in ['p4']
def doparse(url,d):
parm = {}
path = url.split("://")[1]
delim = path.find("@");
if delim != -1:
(user,pswd,host,port) = path.split('@')[0].split(":")
path = path.split('@')[1]
else:
(host,port) = data.getVar('P4PORT', d).split(':')
user = ""
pswd = ""
if path.find(";") != -1:
keys=[]
values=[]
plist = path.split(';')
for item in plist:
if item.count('='):
(key,value) = item.split('=')
keys.append(key)
values.append(value)
parm = dict(zip(keys,values))
path = "//" + path.split(';')[0]
host += ":%s" % (port)
parm["cset"] = Perforce.getcset(d, path, host, user, pswd, parm)
return host,path,user,pswd,parm
doparse = staticmethod(doparse)
def getcset(d, depot,host,user,pswd,parm):
p4opt = ""
if "cset" in parm:
return parm["cset"];
if user:
p4opt += " -u %s" % (user)
if pswd:
p4opt += " -P %s" % (pswd)
if host:
p4opt += " -p %s" % (host)
p4date = data.getVar("P4DATE", d, 1)
if "revision" in parm:
depot += "#%s" % (parm["revision"])
elif "label" in parm:
depot += "@%s" % (parm["label"])
elif p4date:
depot += "@%s" % (p4date)
p4cmd = data.getVar('FETCHCOMMAND_p4', d, 1)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s%s changes -m 1 %s" % (p4cmd, p4opt, depot))
p4file = os.popen("%s%s changes -m 1 %s" % (p4cmd, p4opt, depot))
cset = p4file.readline().strip()
bb.msg.debug(1, bb.msg.domain.Fetcher, "READ %s" % (cset))
if not cset:
return -1
return cset.split(' ')[1]
getcset = staticmethod(getcset)
def localpath(self, url, ud, d):
(host,path,user,pswd,parm) = Perforce.doparse(url,d)
# If a label is specified, we use that as our filename
if "label" in parm:
ud.localfile = "%s.tar.gz" % (parm["label"])
return os.path.join(data.getVar("DL_DIR", d, 1), ud.localfile)
base = path
which = path.find('/...')
if which != -1:
base = path[:which]
if base[0] == "/":
base = base[1:]
cset = Perforce.getcset(d, path, host, user, pswd, parm)
ud.localfile = data.expand('%s+%s+%s.tar.gz' % (host,base.replace('/', '.'), cset), d)
return os.path.join(data.getVar("DL_DIR", d, 1), ud.localfile)
def go(self, loc, ud, d):
"""
Fetch urls
"""
# try to use the tarball stash
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping perforce checkout." % ud.localpath)
return
(host,depot,user,pswd,parm) = Perforce.doparse(loc, d)
if depot.find('/...') != -1:
path = depot[:depot.find('/...')]
else:
path = depot
if "module" in parm:
module = parm["module"]
else:
module = os.path.basename(path)
localdata = data.createCopy(d)
data.setVar('OVERRIDES', "p4:%s" % data.getVar('OVERRIDES', localdata), localdata)
data.update_data(localdata)
# Get the p4 command
p4opt = ""
if user:
p4opt += " -u %s" % (user)
if pswd:
p4opt += " -P %s" % (pswd)
if host:
p4opt += " -p %s" % (host)
p4cmd = data.getVar('FETCHCOMMAND', localdata, 1)
# create temp directory
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: creating temporary directory")
bb.mkdirhier(data.expand('${WORKDIR}', localdata))
data.setVar('TMPBASE', data.expand('${WORKDIR}/oep4.XXXXXX', localdata), localdata)
tmppipe = os.popen(data.getVar('MKTEMPDIRCMD', localdata, 1) or "false")
tmpfile = tmppipe.readline().strip()
if not tmpfile:
bb.error("Fetch: unable to create temporary directory.. make sure 'mktemp' is in the PATH.")
raise FetchError(module)
if "label" in parm:
depot = "%s@%s" % (depot,parm["label"])
else:
cset = Perforce.getcset(d, depot, host, user, pswd, parm)
depot = "%s@%s" % (depot,cset)
os.chdir(tmpfile)
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
bb.msg.note(1, bb.msg.domain.Fetcher, "%s%s files %s" % (p4cmd, p4opt, depot))
p4file = os.popen("%s%s files %s" % (p4cmd, p4opt, depot))
if not p4file:
bb.error("Fetch: unable to get the P4 files from %s" % (depot))
raise FetchError(module)
count = 0
for file in p4file:
list = file.split()
if list[2] == "delete":
continue
dest = list[0][len(path)+1:]
where = dest.find("#")
os.system("%s%s print -o %s/%s %s" % (p4cmd, p4opt, module,dest[:where],list[0]))
count = count + 1
if count == 0:
bb.error("Fetch: No files gathered from the P4 fetch")
raise FetchError(module)
myret = os.system("tar -czf %s %s" % (ud.localpath, module))
if myret != 0:
try:
os.unlink(ud.localpath)
except OSError:
pass
raise FetchError(module)
# cleanup
os.system('rm -rf %s' % tmpfile)

View File

@@ -0,0 +1,118 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
'''
BitBake 'Fetch' implementations
This implementation is for Secure Shell (SSH), and attempts to comply with the
IETF secsh internet draft:
http://tools.ietf.org/wg/secsh/draft-ietf-secsh-scp-sftp-ssh-uri/
Currently does not support the sftp parameters, as this uses scp
Also does not support the 'fingerprint' connection parameter.
'''
# Copyright (C) 2006 OpenedHand Ltd.
#
#
# Based in part on svk.py:
# Copyright (C) 2006 Holger Hans Peter Freyther
# Based on svn.py:
# Copyright (C) 2003, 2004 Chris Larson
# Based on functions from the base bb module:
# Copyright 2003 Holger Schurig
#
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import re, os
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
__pattern__ = re.compile(r'''
\s* # Skip leading whitespace
ssh:// # scheme
( # Optional username/password block
(?P<user>\S+) # username
(:(?P<pass>\S+))? # colon followed by the password (optional)
)?
(?P<cparam>(;[^;]+)*)? # connection parameters block (optional)
@
(?P<host>\S+?) # non-greedy match of the host
(:(?P<port>[0-9]+))? # colon followed by the port (optional)
/
(?P<path>[^;]+) # path on the remote system, may be absolute or relative,
# and may include the use of '~' to reference the remote home
# directory
(?P<sparam>(;[^;]+)*)? # parameters block (optional)
$
''', re.VERBOSE)
class SSH(Fetch):
'''Class to fetch a module or modules via Secure Shell'''
def supports(self, url, urldata, d):
return __pattern__.match(url) != None
def localpath(self, url, urldata, d):
m = __pattern__.match(url)
path = m.group('path')
host = m.group('host')
lpath = os.path.join(data.getVar('DL_DIR', d, True), host, os.path.basename(path))
return lpath
def go(self, url, urldata, d):
dldir = data.getVar('DL_DIR', d, 1)
m = __pattern__.match(url)
path = m.group('path')
host = m.group('host')
port = m.group('port')
user = m.group('user')
password = m.group('pass')
ldir = os.path.join(dldir, host)
lpath = os.path.join(ldir, os.path.basename(path))
if not os.path.exists(ldir):
os.makedirs(ldir)
if port:
port = '-P %s' % port
else:
port = ''
if user:
fr = user
if password:
fr += ':%s' % password
fr += '@%s' % host
else:
fr = host
fr += ':%s' % path
import commands
cmd = 'scp -B -r %s %s %s/' % (
port,
commands.mkarg(fr),
commands.mkarg(ldir)
)
(exitstatus, output) = commands.getstatusoutput(cmd)
if exitstatus != 0:
print output
raise FetchError('Unable to fetch %s' % url)

View File

@@ -0,0 +1,109 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
This implementation is for svk. It is based on the svn implementation
"""
# Copyright (C) 2006 Holger Hans Peter Freyther
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import MissingParameterError
class Svk(Fetch):
"""Class to fetch a module or modules from svk repositories"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with cvs.
"""
return ud.type in ['svk']
def localpath(self, url, ud, d):
if not "module" in ud.parm:
raise MissingParameterError("svk method needs a 'module' parameter")
else:
ud.module = ud.parm["module"]
ud.revision = ""
if 'rev' in ud.parm:
ud.revision = ud.parm['rev']
ud.localfile = data.expand('%s_%s_%s_%s_%s.tar.gz' % (ud.module.replace('/', '.'), ud.host, ud.path.replace('/', '.'), ud.revision, ud.date), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def forcefetch(self, url, ud, d):
if (ud.date == "now"):
return True
return False
def go(self, loc, ud, d):
"""Fetch urls"""
if not self.forcefetch(loc, ud, d) and Fetch.try_mirror(d, ud.localfile):
return
svkroot = ud.host + ud.path
svkcmd = "svk co -r {%s} %s/%s" % (ud.date, svkroot, ud.module)
if ud.revision:
svkcmd = "svk co -r %s %s/%s" % (ud.revision, svkroot, ud.module)
# create temp directory
localdata = data.createCopy(d)
data.update_data(localdata)
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: creating temporary directory")
bb.mkdirhier(data.expand('${WORKDIR}', localdata))
data.setVar('TMPBASE', data.expand('${WORKDIR}/oesvk.XXXXXX', localdata), localdata)
tmppipe = os.popen(data.getVar('MKTEMPDIRCMD', localdata, 1) or "false")
tmpfile = tmppipe.readline().strip()
if not tmpfile:
bb.msg.error(bb.msg.domain.Fetcher, "Fetch: unable to create temporary directory.. make sure 'mktemp' is in the PATH.")
raise FetchError(ud.module)
# check out sources there
os.chdir(tmpfile)
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % svkcmd)
myret = os.system(svkcmd)
if myret != 0:
try:
os.rmdir(tmpfile)
except OSError:
pass
raise FetchError(ud.module)
os.chdir(os.path.join(tmpfile, os.path.dirname(ud.module)))
# tar them up to a defined filename
myret = os.system("tar -czf %s %s" % (ud.localpath, os.path.basename(ud.module)))
if myret != 0:
try:
os.unlink(ud.localpath)
except OSError:
pass
raise FetchError(ud.module)
# cleanup
os.system('rm -rf %s' % tmpfile)

View File

@@ -0,0 +1,206 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementation for svn.
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2004 Marcin Juszkiewicz
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import sys
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import MissingParameterError
from bb.fetch import runfetchcmd
class Svn(Fetch):
"""Class to fetch a module or modules from svn repositories"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with svn.
"""
return ud.type in ['svn']
def localpath(self, url, ud, d):
if not "module" in ud.parm:
raise MissingParameterError("svn method needs a 'module' parameter")
ud.module = ud.parm["module"]
# Create paths to svn checkouts
relpath = ud.path
if relpath.startswith('/'):
# Remove leading slash as os.path.join can't cope
relpath = relpath[1:]
ud.pkgdir = os.path.join(data.expand('${SVNDIR}', d), ud.host, relpath)
ud.moddir = os.path.join(ud.pkgdir, ud.module)
if 'rev' in ud.parm:
ud.date = ""
ud.revision = ud.parm['rev']
elif 'date' in ud.date:
ud.date = ud.parm['date']
ud.revision = ""
else:
#
# ***Nasty hack***
# If DATE in unexpanded PV, use ud.date (which is set from SRCDATE)
# Should warn people to switch to SRCREV here
#
pv = data.getVar("PV", d, 0)
if "DATE" in pv:
ud.revision = ""
else:
rev = Fetch.srcrev_internal_helper(ud, d)
if rev is True:
ud.revision = self.latest_revision(url, ud, d)
ud.date = ""
elif rev:
ud.revision = rev
ud.date = ""
else:
ud.revision = ""
ud.localfile = data.expand('%s_%s_%s_%s_%s.tar.gz' % (ud.module.replace('/', '.'), ud.host, ud.path.replace('/', '.'), ud.revision, ud.date), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def _buildsvncommand(self, ud, d, command):
"""
Build up an svn commandline based on ud
command is "fetch", "update", "info"
"""
basecmd = data.expand('${FETCHCMD_svn}', d)
proto = "svn"
if "proto" in ud.parm:
proto = ud.parm["proto"]
svn_rsh = None
if proto == "svn+ssh" and "rsh" in ud.parm:
svn_rsh = ud.parm["rsh"]
svnroot = ud.host + ud.path
# either use the revision, or SRCDATE in braces,
options = []
if ud.user:
options.append("--username %s" % ud.user)
if ud.pswd:
options.append("--password %s" % ud.pswd)
if command is "info":
svncmd = "%s info %s %s://%s/%s/" % (basecmd, " ".join(options), proto, svnroot, ud.module)
else:
suffix = ""
if ud.revision:
options.append("-r %s" % ud.revision)
suffix = "@%s" % (ud.revision)
elif ud.date:
options.append("-r {%s}" % ud.date)
if command is "fetch":
svncmd = "%s co %s %s://%s/%s%s %s" % (basecmd, " ".join(options), proto, svnroot, ud.module, suffix, ud.module)
elif command is "update":
svncmd = "%s update %s" % (basecmd, " ".join(options))
else:
raise FetchError("Invalid svn command %s" % command)
if svn_rsh:
svncmd = "svn_RSH=\"%s\" %s" % (svn_rsh, svncmd)
return svncmd
def go(self, loc, ud, d):
"""Fetch url"""
# try to use the tarball stash
if Fetch.try_mirror(d, ud.localfile):
bb.msg.debug(1, bb.msg.domain.Fetcher, "%s already exists or was mirrored, skipping svn checkout." % ud.localpath)
return
bb.msg.debug(2, bb.msg.domain.Fetcher, "Fetch: checking for module directory '" + ud.moddir + "'")
if os.access(os.path.join(ud.moddir, '.svn'), os.R_OK):
svnupdatecmd = self._buildsvncommand(ud, d, "update")
bb.msg.note(1, bb.msg.domain.Fetcher, "Update " + loc)
# update sources there
os.chdir(ud.moddir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % svnupdatecmd)
runfetchcmd(svnupdatecmd, d)
else:
svnfetchcmd = self._buildsvncommand(ud, d, "fetch")
bb.msg.note(1, bb.msg.domain.Fetcher, "Fetch " + loc)
# check out sources there
bb.mkdirhier(ud.pkgdir)
os.chdir(ud.pkgdir)
bb.msg.debug(1, bb.msg.domain.Fetcher, "Running %s" % svnfetchcmd)
runfetchcmd(svnfetchcmd, d)
os.chdir(ud.pkgdir)
# tar them up to a defined filename
try:
runfetchcmd("tar -czf %s %s" % (ud.localpath, ud.module), d)
except:
t, v, tb = sys.exc_info()
try:
os.unlink(ud.localpath)
except OSError:
pass
raise t, v, tb
def suppports_srcrev(self):
return True
def _revision_key(self, url, ud, d):
"""
Return a unique key for the url
"""
return "svn:" + ud.moddir
def _latest_revision(self, url, ud, d):
"""
Return the latest upstream revision number
"""
bb.msg.debug(2, bb.msg.domain.Fetcher, "SVN fetcher hitting network for %s" % url)
output = runfetchcmd("LANG=C LC_ALL=C " + self._buildsvncommand(ud, d, "info"), d, True)
revision = None
for line in output.splitlines():
if "Last Changed Rev" in line:
revision = line.split(":")[1].strip()
return revision
def _sortable_revision(self, url, ud, d):
"""
Return a sortable revision number which in our case is the revision number
"""
return self._build_revision(url, ud, d)
def _build_revision(self, url, ud, d):
return ud.revision

View File

@@ -0,0 +1,130 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'Fetch' implementations
Classes for obtaining upstream sources for the
BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
import os
import bb
from bb import data
from bb.fetch import Fetch
from bb.fetch import FetchError
from bb.fetch import uri_replace
class Wget(Fetch):
"""Class to fetch urls via 'wget'"""
def supports(self, url, ud, d):
"""
Check to see if a given url can be fetched with cvs.
"""
return ud.type in ['http','https','ftp']
def localpath(self, url, ud, d):
url = bb.encodeurl([ud.type, ud.host, ud.path, ud.user, ud.pswd, {}])
ud.basename = os.path.basename(ud.path)
ud.localfile = data.expand(os.path.basename(url), d)
return os.path.join(data.getVar("DL_DIR", d, True), ud.localfile)
def go(self, uri, ud, d, checkonly = False):
"""Fetch urls"""
def fetch_uri(uri, ud, d):
if checkonly:
fetchcmd = data.getVar("CHECKCOMMAND", d, 1)
elif os.path.exists(ud.localpath):
# file exists, but we didnt complete it.. trying again..
fetchcmd = data.getVar("RESUMECOMMAND", d, 1)
else:
fetchcmd = data.getVar("FETCHCOMMAND", d, 1)
uri = uri.split(";")[0]
uri_decoded = list(bb.decodeurl(uri))
uri_type = uri_decoded[0]
uri_host = uri_decoded[1]
bb.msg.note(1, bb.msg.domain.Fetcher, "fetch " + uri)
fetchcmd = fetchcmd.replace("${URI}", uri.split(";")[0])
fetchcmd = fetchcmd.replace("${FILE}", ud.basename)
httpproxy = None
ftpproxy = None
if uri_type == 'http':
httpproxy = data.getVar("HTTP_PROXY", d, True)
httpproxy_ignore = (data.getVar("HTTP_PROXY_IGNORE", d, True) or "").split()
for p in httpproxy_ignore:
if uri_host.endswith(p):
httpproxy = None
break
if uri_type == 'ftp':
ftpproxy = data.getVar("FTP_PROXY", d, True)
ftpproxy_ignore = (data.getVar("HTTP_PROXY_IGNORE", d, True) or "").split()
for p in ftpproxy_ignore:
if uri_host.endswith(p):
ftpproxy = None
break
if httpproxy:
fetchcmd = "http_proxy=" + httpproxy + " " + fetchcmd
if ftpproxy:
fetchcmd = "ftp_proxy=" + ftpproxy + " " + fetchcmd
bb.msg.debug(2, bb.msg.domain.Fetcher, "executing " + fetchcmd)
ret = os.system(fetchcmd)
if ret != 0:
return False
# Sanity check since wget can pretend it succeed when it didn't
# Also, this used to happen if sourceforge sent us to the mirror page
if not os.path.exists(ud.localpath) and not checkonly:
bb.msg.debug(2, bb.msg.domain.Fetcher, "The fetch command for %s returned success but %s doesn't exist?..." % (uri, ud.localpath))
return False
return True
localdata = data.createCopy(d)
data.setVar('OVERRIDES', "wget:" + data.getVar('OVERRIDES', localdata), localdata)
data.update_data(localdata)
premirrors = [ i.split() for i in (data.getVar('PREMIRRORS', localdata, 1) or "").split('\n') if i ]
for (find, replace) in premirrors:
newuri = uri_replace(uri, find, replace, d)
if newuri != uri:
if fetch_uri(newuri, ud, localdata):
return True
if fetch_uri(uri, ud, localdata):
return True
# try mirrors
mirrors = [ i.split() for i in (data.getVar('MIRRORS', localdata, 1) or "").split('\n') if i ]
for (find, replace) in mirrors:
newuri = uri_replace(uri, find, replace, d)
if newuri != uri:
if fetch_uri(newuri, ud, localdata):
return True
raise FetchError(uri)
def checkstatus(self, uri, ud, d):
return self.go(uri, ud, d, True)

View File

@@ -0,0 +1,144 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2003, 2004 Chris Larson
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os, sys
import bb, bb.data
def getfields(line):
fields = {}
fieldmap = ( "pkg", "src", "dest", "type", "mode", "uid", "gid", "major", "minor", "start", "inc", "count" )
for f in xrange(len(fieldmap)):
fields[fieldmap[f]] = None
if not line:
return None
splitline = line.split()
if not len(splitline):
return None
try:
for f in xrange(len(fieldmap)):
if splitline[f] == '-':
continue
fields[fieldmap[f]] = splitline[f]
except IndexError:
pass
return fields
def parse (mfile, d):
manifest = []
while 1:
line = mfile.readline()
if not line:
break
if line.startswith("#"):
continue
fields = getfields(line)
if not fields:
continue
manifest.append(fields)
return manifest
def emit (func, manifest, d):
#str = "%s () {\n" % func
str = ""
for line in manifest:
emittedline = emit_line(func, line, d)
if not emittedline:
continue
str += emittedline + "\n"
# str += "}\n"
return str
def mangle (func, line, d):
import copy
newline = copy.copy(line)
src = bb.data.expand(newline["src"], d)
if src:
if not os.path.isabs(src):
src = "${WORKDIR}/" + src
dest = newline["dest"]
if not dest:
return
if dest.startswith("/"):
dest = dest[1:]
if func is "do_install":
dest = "${D}/" + dest
elif func is "do_populate":
dest = "${WORKDIR}/install/" + newline["pkg"] + "/" + dest
elif func is "do_stage":
varmap = {}
varmap["${bindir}"] = "${STAGING_DIR}/${HOST_SYS}/bin"
varmap["${libdir}"] = "${STAGING_DIR}/${HOST_SYS}/lib"
varmap["${includedir}"] = "${STAGING_DIR}/${HOST_SYS}/include"
varmap["${datadir}"] = "${STAGING_DATADIR}"
matched = 0
for key in varmap.keys():
if dest.startswith(key):
dest = varmap[key] + "/" + dest[len(key):]
matched = 1
if not matched:
newline = None
return
else:
newline = None
return
newline["src"] = src
newline["dest"] = dest
return newline
def emit_line (func, line, d):
import copy
newline = copy.deepcopy(line)
newline = mangle(func, newline, d)
if not newline:
return None
str = ""
type = newline["type"]
mode = newline["mode"]
src = newline["src"]
dest = newline["dest"]
if type is "d":
str = "install -d "
if mode:
str += "-m %s " % mode
str += dest
elif type is "f":
if not src:
return None
if dest.endswith("/"):
str = "install -d "
str += dest + "\n"
str += "install "
else:
str = "install -D "
if mode:
str += "-m %s " % mode
str += src + " " + dest
del newline
return str

View File

@@ -0,0 +1,84 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
#
# Copyright (C) 2006 Holger Hans Peter Freyther
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
"""
What is a method pool?
BitBake has a global method scope where .bb, .inc and .bbclass
files can install methods. These methods are parsed from strings.
To avoid recompiling and executing these string we introduce
a method pool to do this task.
This pool will be used to compile and execute the functions. It
will be smart enough to
"""
from bb.utils import better_compile, better_exec
from bb import error
# A dict of modules we have handled
# it is the number of .bbclasses + x in size
_parsed_methods = { }
_parsed_fns = { }
def insert_method(modulename, code, fn):
"""
Add code of a module should be added. The methods
will be simply added, no checking will be done
"""
comp = better_compile(code, "<bb>", fn )
better_exec(comp, __builtins__, code, fn)
# now some instrumentation
code = comp.co_names
for name in code:
if name in ['None', 'False']:
continue
elif name in _parsed_fns and not _parsed_fns[name] == modulename:
error( "Error Method already seen: %s in' %s' now in '%s'" % (name, _parsed_fns[name], modulename))
else:
_parsed_fns[name] = modulename
def check_insert_method(modulename, code, fn):
"""
Add the code if it wasnt added before. The module
name will be used for that
Variables:
@modulename a short name e.g. base.bbclass
@code The actual python code
@fn The filename from the outer file
"""
if not modulename in _parsed_methods:
return insert_method(modulename, code, fn)
_parsed_methods[modulename] = 1
def parsed_module(modulename):
"""
Inform me file xyz was parsed
"""
return modulename in _parsed_methods
def get_parsed_dict():
"""
shortcut
"""
return _parsed_methods

125
bitbake-dev/lib/bb/msg.py Normal file
View File

@@ -0,0 +1,125 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'msg' implementation
Message handling infrastructure for bitbake
"""
# Copyright (C) 2006 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import sys, bb
from bb import event
debug_level = {}
verbose = False
domain = bb.utils.Enum(
'Build',
'Cache',
'Collection',
'Data',
'Depends',
'Fetcher',
'Parsing',
'PersistData',
'Provider',
'RunQueue',
'TaskData',
'Util')
class MsgBase(bb.event.Event):
"""Base class for messages"""
def __init__(self, msg, d ):
self._message = msg
event.Event.__init__(self, d)
class MsgDebug(MsgBase):
"""Debug Message"""
class MsgNote(MsgBase):
"""Note Message"""
class MsgWarn(MsgBase):
"""Warning Message"""
class MsgError(MsgBase):
"""Error Message"""
class MsgFatal(MsgBase):
"""Fatal Message"""
class MsgPlain(MsgBase):
"""General output"""
#
# Message control functions
#
def set_debug_level(level):
bb.msg.debug_level = {}
for domain in bb.msg.domain:
bb.msg.debug_level[domain] = level
bb.msg.debug_level['default'] = level
def set_verbose(level):
bb.msg.verbose = level
def set_debug_domains(domains):
for domain in domains:
found = False
for ddomain in bb.msg.domain:
if domain == str(ddomain):
bb.msg.debug_level[ddomain] = bb.msg.debug_level[ddomain] + 1
found = True
if not found:
bb.msg.warn(None, "Logging domain %s is not valid, ignoring" % domain)
#
# Message handling functions
#
def debug(level, domain, msg, fn = None):
if not domain:
domain = 'default'
if debug_level[domain] >= level:
bb.event.fire(MsgDebug(msg, None))
def note(level, domain, msg, fn = None):
if not domain:
domain = 'default'
if level == 1 or verbose or debug_level[domain] >= 1:
bb.event.fire(MsgNote(msg, None))
def warn(domain, msg, fn = None):
bb.event.fire(MsgWarn(msg, None))
def error(domain, msg, fn = None):
bb.event.fire(MsgError(msg, None))
print 'ERROR: ' + msg
def fatal(domain, msg, fn = None):
bb.event.fire(MsgFatal(msg, None))
print 'FATAL: ' + msg
sys.exit(1)
def plain(msg, fn = None):
bb.event.fire(MsgPlain(msg, None))

View File

@@ -0,0 +1,84 @@
"""
BitBake Parsers
File parsers for the BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
__all__ = [ 'ParseError', 'SkipPackage', 'cached_mtime', 'mark_dependency',
'supports', 'handle', 'init' ]
handlers = []
import bb, os
class ParseError(Exception):
"""Exception raised when parsing fails"""
class SkipPackage(Exception):
"""Exception raised to skip this package"""
__mtime_cache = {}
def cached_mtime(f):
if not __mtime_cache.has_key(f):
__mtime_cache[f] = os.stat(f)[8]
return __mtime_cache[f]
def cached_mtime_noerror(f):
if not __mtime_cache.has_key(f):
try:
__mtime_cache[f] = os.stat(f)[8]
except OSError:
return 0
return __mtime_cache[f]
def update_mtime(f):
__mtime_cache[f] = os.stat(f)[8]
return __mtime_cache[f]
def mark_dependency(d, f):
if f.startswith('./'):
f = "%s/%s" % (os.getcwd(), f[2:])
deps = bb.data.getVar('__depends', d) or []
deps.append( (f, cached_mtime(f)) )
bb.data.setVar('__depends', deps, d)
def supports(fn, data):
"""Returns true if we have a handler for this file, false otherwise"""
for h in handlers:
if h['supports'](fn, data):
return 1
return 0
def handle(fn, data, include = 0):
"""Call the handler that is appropriate for this file"""
for h in handlers:
if h['supports'](fn, data):
return h['handle'](fn, data, include)
raise ParseError("%s is not a BitBake file" % fn)
def init(fn, data):
for h in handlers:
if h['supports'](fn):
return h['init'](data)
from parse_py import __version__, ConfHandler, BBHandler

View File

@@ -0,0 +1,401 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
class for handling .bb files
Reads a .bb file and obtains its metadata
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import re, bb, os, sys, time, string
import bb.fetch, bb.build, bb.utils
from bb import data, fetch, methodpool
from ConfHandler import include, localpath, obtain, init
from bb.parse import ParseError
__func_start_regexp__ = re.compile( r"(((?P<py>python)|(?P<fr>fakeroot))\s*)*(?P<func>[\w\.\-\+\{\}\$]+)?\s*\(\s*\)\s*{$" )
__inherit_regexp__ = re.compile( r"inherit\s+(.+)" )
__export_func_regexp__ = re.compile( r"EXPORT_FUNCTIONS\s+(.+)" )
__addtask_regexp__ = re.compile("addtask\s+(?P<func>\w+)\s*((before\s*(?P<before>((.*(?=after))|(.*))))|(after\s*(?P<after>((.*(?=before))|(.*)))))*")
__addhandler_regexp__ = re.compile( r"addhandler\s+(.+)" )
__def_regexp__ = re.compile( r"def\s+(\w+).*:" )
__python_func_regexp__ = re.compile( r"(\s+.*)|(^$)" )
__word__ = re.compile(r"\S+")
__infunc__ = ""
__inpython__ = False
__body__ = []
__classname__ = ""
classes = [ None, ]
# We need to indicate EOF to the feeder. This code is so messy that
# factoring it out to a close_parse_file method is out of question.
# We will use the IN_PYTHON_EOF as an indicator to just close the method
#
# The two parts using it are tightly integrated anyway
IN_PYTHON_EOF = -9999999999999
__parsed_methods__ = methodpool.get_parsed_dict()
def supports(fn, d):
localfn = localpath(fn, d)
return localfn[-3:] == ".bb" or localfn[-8:] == ".bbclass" or localfn[-4:] == ".inc"
def inherit(files, d):
__inherit_cache = data.getVar('__inherit_cache', d) or []
fn = ""
lineno = 0
files = data.expand(files, d)
for file in files:
if file[0] != "/" and file[-8:] != ".bbclass":
file = os.path.join('classes', '%s.bbclass' % file)
if not file in __inherit_cache:
bb.msg.debug(2, bb.msg.domain.Parsing, "BB %s:%d: inheriting %s" % (fn, lineno, file))
__inherit_cache.append( file )
data.setVar('__inherit_cache', __inherit_cache, d)
include(fn, file, d, "inherit")
__inherit_cache = data.getVar('__inherit_cache', d) or []
def finalise(fn, d):
data.expandKeys(d)
data.update_data(d)
anonqueue = data.getVar("__anonqueue", d, 1) or []
body = [x['content'] for x in anonqueue]
flag = { 'python' : 1, 'func' : 1 }
data.setVar("__anonfunc", "\n".join(body), d)
data.setVarFlags("__anonfunc", flag, d)
from bb import build
try:
t = data.getVar('T', d)
data.setVar('T', '${TMPDIR}/anonfunc/', d)
anonfuncs = data.getVar('__BBANONFUNCS', d) or []
code = ""
for f in anonfuncs:
code = code + " %s(d)\n" % f
data.setVar("__anonfunc", code, d)
build.exec_func("__anonfunc", d)
data.delVar('T', d)
if t:
data.setVar('T', t, d)
except Exception, e:
bb.msg.debug(1, bb.msg.domain.Parsing, "Exception when executing anonymous function: %s" % e)
raise
data.delVar("__anonqueue", d)
data.delVar("__anonfunc", d)
data.update_data(d)
all_handlers = {}
for var in data.getVar('__BBHANDLERS', d) or []:
# try to add the handler
handler = data.getVar(var,d)
bb.event.register(var, handler)
tasklist = data.getVar('__BBTASKS', d) or []
bb.build.add_tasks(tasklist, d)
bb.event.fire(bb.event.RecipeParsed(fn, d))
def handle(fn, d, include = 0):
global __func_start_regexp__, __inherit_regexp__, __export_func_regexp__, __addtask_regexp__, __addhandler_regexp__, __infunc__, __body__, __residue__
__body__ = []
__infunc__ = ""
__classname__ = ""
__residue__ = []
if include == 0:
bb.msg.debug(2, bb.msg.domain.Parsing, "BB " + fn + ": handle(data)")
else:
bb.msg.debug(2, bb.msg.domain.Parsing, "BB " + fn + ": handle(data, include)")
(root, ext) = os.path.splitext(os.path.basename(fn))
base_name = "%s%s" % (root,ext)
init(d)
if ext == ".bbclass":
__classname__ = root
classes.append(__classname__)
__inherit_cache = data.getVar('__inherit_cache', d) or []
if not fn in __inherit_cache:
__inherit_cache.append(fn)
data.setVar('__inherit_cache', __inherit_cache, d)
if include != 0:
oldfile = data.getVar('FILE', d)
else:
oldfile = None
fn = obtain(fn, d)
bbpath = (data.getVar('BBPATH', d, 1) or '').split(':')
if not os.path.isabs(fn):
f = None
for p in bbpath:
j = os.path.join(p, fn)
if os.access(j, os.R_OK):
abs_fn = j
f = open(j, 'r')
break
if f is None:
raise IOError("file %s not found" % fn)
else:
f = open(fn,'r')
abs_fn = fn
if include:
bb.parse.mark_dependency(d, abs_fn)
if ext != ".bbclass":
data.setVar('FILE', fn, d)
lineno = 0
while 1:
lineno = lineno + 1
s = f.readline()
if not s: break
s = s.rstrip()
feeder(lineno, s, fn, base_name, d)
if __inpython__:
# add a blank line to close out any python definition
feeder(IN_PYTHON_EOF, "", fn, base_name, d)
if ext == ".bbclass":
classes.remove(__classname__)
else:
if include == 0:
multi = data.getVar('BBCLASSEXTEND', d, 1)
if multi:
based = bb.data.createCopy(d)
finalise(fn, based)
darray = {"": based}
for cls in multi.split():
pn = data.getVar('PN', d, True)
based = bb.data.createCopy(d)
data.setVar('PN', pn + '-' + cls, based)
inherit([cls], based)
finalise(fn, based)
darray[cls] = based
return darray
else:
finalise(fn, d)
bbpath.pop(0)
if oldfile:
bb.data.setVar("FILE", oldfile, d)
# we have parsed the bb class now
if ext == ".bbclass" or ext == ".inc":
__parsed_methods__[base_name] = 1
return d
def feeder(lineno, s, fn, root, d):
global __func_start_regexp__, __inherit_regexp__, __export_func_regexp__, __addtask_regexp__, __addhandler_regexp__, __def_regexp__, __python_func_regexp__, __inpython__,__infunc__, __body__, classes, bb, __residue__
if __infunc__:
if s == '}':
__body__.append('')
if __infunc__ == "__anonymous":
funcname = ("__anon_%s_%s" % (lineno, fn.translate(string.maketrans('/.+-', '____'))))
if not funcname in methodpool._parsed_fns:
text = "def %s(d):\n" % (funcname) + '\n'.join(__body__)
methodpool.insert_method(funcname, text, fn)
anonfuncs = data.getVar('__BBANONFUNCS', d) or []
anonfuncs.append(funcname)
data.setVar('__BBANONFUNCS', anonfuncs, d)
else:
data.setVarFlag(__infunc__, "func", 1, d)
data.setVar(__infunc__, '\n'.join(__body__), d)
__infunc__ = ""
__body__ = []
else:
__body__.append(s)
return
if __inpython__:
m = __python_func_regexp__.match(s)
if m and lineno != IN_PYTHON_EOF:
__body__.append(s)
return
else:
# Note we will add root to parsedmethods after having parse
# 'this' file. This means we will not parse methods from
# bb classes twice
if not root in __parsed_methods__:
text = '\n'.join(__body__)
methodpool.insert_method( root, text, fn )
__body__ = []
__inpython__ = False
if lineno == IN_PYTHON_EOF:
return
# fall through
if s == '' or s[0] == '#': return # skip comments and empty lines
if s[-1] == '\\':
__residue__.append(s[:-1])
return
s = "".join(__residue__) + s
__residue__ = []
m = __func_start_regexp__.match(s)
if m:
__infunc__ = m.group("func") or "__anonymous"
key = __infunc__
if data.getVar(key, d):
# clean up old version of this piece of metadata, as its
# flags could cause problems
data.setVarFlag(key, 'python', None, d)
data.setVarFlag(key, 'fakeroot', None, d)
if m.group("py") is not None:
data.setVarFlag(key, "python", "1", d)
else:
data.delVarFlag(key, "python", d)
if m.group("fr") is not None:
data.setVarFlag(key, "fakeroot", "1", d)
else:
data.delVarFlag(key, "fakeroot", d)
return
m = __def_regexp__.match(s)
if m:
__body__.append(s)
__inpython__ = True
return
m = __export_func_regexp__.match(s)
if m:
fns = m.group(1)
n = __word__.findall(fns)
for f in n:
allvars = []
allvars.append(f)
allvars.append(classes[-1] + "_" + f)
vars = [[ allvars[0], allvars[1] ]]
if len(classes) > 1 and classes[-2] is not None:
allvars.append(classes[-2] + "_" + f)
vars = []
vars.append([allvars[2], allvars[1]])
vars.append([allvars[0], allvars[2]])
for (var, calledvar) in vars:
if data.getVar(var, d) and not data.getVarFlag(var, 'export_func', d):
continue
if data.getVar(var, d):
data.setVarFlag(var, 'python', None, d)
data.setVarFlag(var, 'func', None, d)
for flag in [ "func", "python" ]:
if data.getVarFlag(calledvar, flag, d):
data.setVarFlag(var, flag, data.getVarFlag(calledvar, flag, d), d)
for flag in [ "dirs" ]:
if data.getVarFlag(var, flag, d):
data.setVarFlag(calledvar, flag, data.getVarFlag(var, flag, d), d)
if data.getVarFlag(calledvar, "python", d):
data.setVar(var, "\tbb.build.exec_func('" + calledvar + "', d)\n", d)
else:
data.setVar(var, "\t" + calledvar + "\n", d)
data.setVarFlag(var, 'export_func', '1', d)
return
m = __addtask_regexp__.match(s)
if m:
func = m.group("func")
before = m.group("before")
after = m.group("after")
if func is None:
return
var = "do_" + func
data.setVarFlag(var, "task", 1, d)
bbtasks = data.getVar('__BBTASKS', d) or []
if not var in bbtasks:
bbtasks.append(var)
data.setVar('__BBTASKS', bbtasks, d)
existing = data.getVarFlag(var, "deps", d) or []
if after is not None:
# set up deps for function
for entry in after.split():
if entry not in existing:
existing.append(entry)
data.setVarFlag(var, "deps", existing, d)
if before is not None:
# set up things that depend on this func
for entry in before.split():
existing = data.getVarFlag(entry, "deps", d) or []
if var not in existing:
data.setVarFlag(entry, "deps", [var] + existing, d)
return
m = __addhandler_regexp__.match(s)
if m:
fns = m.group(1)
hs = __word__.findall(fns)
bbhands = data.getVar('__BBHANDLERS', d) or []
for h in hs:
bbhands.append(h)
data.setVarFlag(h, "handler", 1, d)
data.setVar('__BBHANDLERS', bbhands, d)
return
m = __inherit_regexp__.match(s)
if m:
files = m.group(1)
n = __word__.findall(files)
inherit(n, d)
return
from bb.parse import ConfHandler
return ConfHandler.feeder(lineno, s, fn, d)
__pkgsplit_cache__={}
def vars_from_file(mypkg, d):
if not mypkg:
return (None, None, None)
if mypkg in __pkgsplit_cache__:
return __pkgsplit_cache__[mypkg]
myfile = os.path.splitext(os.path.basename(mypkg))
parts = myfile[0].split('_')
__pkgsplit_cache__[mypkg] = parts
if len(parts) > 3:
raise ParseError("Unable to generate default variables from the filename: %s (too many underscores)" % mypkg)
exp = 3 - len(parts)
tmplist = []
while exp != 0:
exp -= 1
tmplist.append(None)
parts.extend(tmplist)
return parts
# Add us to the handlers list
from bb.parse import handlers
handlers.append({'supports': supports, 'handle': handle, 'init': init})
del handlers

View File

@@ -0,0 +1,234 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
class for handling configuration data files
Reads a .conf file and obtains its metadata
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import re, bb.data, os, sys
from bb.parse import ParseError
#__config_regexp__ = re.compile( r"(?P<exp>export\s*)?(?P<var>[a-zA-Z0-9\-_+.${}]+)\s*(?P<colon>:)?(?P<ques>\?)?=\s*(?P<apo>['\"]?)(?P<value>.*)(?P=apo)$")
__config_regexp__ = re.compile( r"(?P<exp>export\s*)?(?P<var>[a-zA-Z0-9\-_+.${}/]+)(\[(?P<flag>[a-zA-Z0-9\-_+.]+)\])?\s*((?P<colon>:=)|(?P<ques>\?=)|(?P<append>\+=)|(?P<prepend>=\+)|(?P<predot>=\.)|(?P<postdot>\.=)|=)\s*(?P<apo>['\"]?)(?P<value>.*)(?P=apo)$")
__include_regexp__ = re.compile( r"include\s+(.+)" )
__require_regexp__ = re.compile( r"require\s+(.+)" )
__export_regexp__ = re.compile( r"export\s+(.+)" )
def init(data):
if not bb.data.getVar('TOPDIR', data):
bb.data.setVar('TOPDIR', os.getcwd(), data)
if not bb.data.getVar('BBPATH', data):
bb.data.setVar('BBPATH', os.path.join(sys.prefix, 'share', 'bitbake'), data)
def supports(fn, d):
return localpath(fn, d)[-5:] == ".conf"
def localpath(fn, d):
if os.path.exists(fn):
return fn
if "://" not in fn:
return fn
localfn = None
try:
localfn = bb.fetch.localpath(fn, d, False)
except bb.MalformedUrl:
pass
if not localfn:
return fn
return localfn
def obtain(fn, data):
import sys, bb
fn = bb.data.expand(fn, data)
localfn = bb.data.expand(localpath(fn, data), data)
if localfn != fn:
dldir = bb.data.getVar('DL_DIR', data, 1)
if not dldir:
bb.msg.debug(1, bb.msg.domain.Parsing, "obtain: DL_DIR not defined")
return localfn
bb.mkdirhier(dldir)
try:
bb.fetch.init([fn], data)
except bb.fetch.NoMethodError:
(type, value, traceback) = sys.exc_info()
bb.msg.debug(1, bb.msg.domain.Parsing, "obtain: no method: %s" % value)
return localfn
try:
bb.fetch.go(data)
except bb.fetch.MissingParameterError:
(type, value, traceback) = sys.exc_info()
bb.msg.debug(1, bb.msg.domain.Parsing, "obtain: missing parameters: %s" % value)
return localfn
except bb.fetch.FetchError:
(type, value, traceback) = sys.exc_info()
bb.msg.debug(1, bb.msg.domain.Parsing, "obtain: failed: %s" % value)
return localfn
return localfn
def include(oldfn, fn, data, error_out):
"""
error_out If True a ParseError will be reaised if the to be included
"""
if oldfn == fn: # prevent infinate recursion
return None
import bb
fn = bb.data.expand(fn, data)
oldfn = bb.data.expand(oldfn, data)
if not os.path.isabs(fn):
dname = os.path.dirname(oldfn)
bbpath = "%s:%s" % (dname, bb.data.getVar("BBPATH", data, 1))
abs_fn = bb.which(bbpath, fn)
if abs_fn:
fn = abs_fn
from bb.parse import handle
try:
ret = handle(fn, data, True)
except IOError:
if error_out:
raise ParseError("Could not %(error_out)s file %(fn)s" % vars() )
bb.msg.debug(2, bb.msg.domain.Parsing, "CONF file '%s' not found" % fn)
def handle(fn, data, include = 0):
if include:
inc_string = "including"
else:
inc_string = "reading"
init(data)
if include == 0:
oldfile = None
else:
oldfile = bb.data.getVar('FILE', data)
fn = obtain(fn, data)
if not os.path.isabs(fn):
f = None
bbpath = bb.data.getVar("BBPATH", data, 1) or []
for p in bbpath.split(":"):
currname = os.path.join(p, fn)
if os.access(currname, os.R_OK):
f = open(currname, 'r')
abs_fn = currname
bb.msg.debug(2, bb.msg.domain.Parsing, "CONF %s %s" % (inc_string, currname))
break
if f is None:
raise IOError("file '%s' not found" % fn)
else:
f = open(fn,'r')
bb.msg.debug(1, bb.msg.domain.Parsing, "CONF %s %s" % (inc_string,fn))
abs_fn = fn
if include:
bb.parse.mark_dependency(data, abs_fn)
lineno = 0
bb.data.setVar('FILE', fn, data)
while 1:
lineno = lineno + 1
s = f.readline()
if not s: break
w = s.strip()
if not w: continue # skip empty lines
s = s.rstrip()
if s[0] == '#': continue # skip comments
while s[-1] == '\\':
s2 = f.readline()[:-1].strip()
lineno = lineno + 1
s = s[:-1] + s2
feeder(lineno, s, fn, data)
if oldfile:
bb.data.setVar('FILE', oldfile, data)
return data
def feeder(lineno, s, fn, data):
def getFunc(groupd, key, data):
if 'flag' in groupd and groupd['flag'] != None:
return bb.data.getVarFlag(key, groupd['flag'], data)
else:
return bb.data.getVar(key, data)
m = __config_regexp__.match(s)
if m:
groupd = m.groupdict()
key = groupd["var"]
if "exp" in groupd and groupd["exp"] != None:
bb.data.setVarFlag(key, "export", 1, data)
if "ques" in groupd and groupd["ques"] != None:
val = getFunc(groupd, key, data)
if val == None:
val = groupd["value"]
elif "colon" in groupd and groupd["colon"] != None:
e = data.createCopy()
bb.data.update_data(e)
val = bb.data.expand(groupd["value"], e)
elif "append" in groupd and groupd["append"] != None:
val = "%s %s" % ((getFunc(groupd, key, data) or ""), groupd["value"])
elif "prepend" in groupd and groupd["prepend"] != None:
val = "%s %s" % (groupd["value"], (getFunc(groupd, key, data) or ""))
elif "postdot" in groupd and groupd["postdot"] != None:
val = "%s%s" % ((getFunc(groupd, key, data) or ""), groupd["value"])
elif "predot" in groupd and groupd["predot"] != None:
val = "%s%s" % (groupd["value"], (getFunc(groupd, key, data) or ""))
else:
val = groupd["value"]
if 'flag' in groupd and groupd['flag'] != None:
bb.msg.debug(3, bb.msg.domain.Parsing, "setVarFlag(%s, %s, %s, data)" % (key, groupd['flag'], val))
bb.data.setVarFlag(key, groupd['flag'], val, data)
else:
bb.data.setVar(key, val, data)
return
m = __include_regexp__.match(s)
if m:
s = bb.data.expand(m.group(1), data)
bb.msg.debug(3, bb.msg.domain.Parsing, "CONF %s:%d: including %s" % (fn, lineno, s))
include(fn, s, data, False)
return
m = __require_regexp__.match(s)
if m:
s = bb.data.expand(m.group(1), data)
include(fn, s, data, "include required")
return
m = __export_regexp__.match(s)
if m:
bb.data.setVarFlag(m.group(1), "export", 1, data)
return
raise ParseError("%s:%d: unparsed line: '%s'" % (fn, lineno, s));
# Add us to the handlers list
from bb.parse import handlers
handlers.append({'supports': supports, 'handle': handle, 'init': init})
del handlers

View File

@@ -0,0 +1,33 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake Parsers
File parsers for the BitBake build tools.
"""
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# Based on functions from the base bb module, Copyright 2003 Holger Schurig
__version__ = '1.0'
__all__ = [ 'ConfHandler', 'BBHandler']
import ConfHandler
import BBHandler

View File

@@ -0,0 +1,121 @@
# BitBake Persistent Data Store
#
# Copyright (C) 2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import bb, os
try:
import sqlite3
except ImportError:
try:
from pysqlite2 import dbapi2 as sqlite3
except ImportError:
bb.msg.fatal(bb.msg.domain.PersistData, "Importing sqlite3 and pysqlite2 failed, please install one of them. Python 2.5 or a 'python-pysqlite2' like package is likely to be what you need.")
sqlversion = sqlite3.sqlite_version_info
if sqlversion[0] < 3 or (sqlversion[0] == 3 and sqlversion[1] < 3):
bb.msg.fatal(bb.msg.domain.PersistData, "sqlite3 version 3.3.0 or later is required.")
class PersistData:
"""
BitBake Persistent Data Store
Used to store data in a central location such that other threads/tasks can
access them at some future date.
The "domain" is used as a key to isolate each data pool and in this
implementation corresponds to an SQL table. The SQL table consists of a
simple key and value pair.
Why sqlite? It handles all the locking issues for us.
"""
def __init__(self, d):
self.cachedir = bb.data.getVar("PERSISTENT_DIR", d, True) or bb.data.getVar("CACHE", d, True)
if self.cachedir in [None, '']:
bb.msg.fatal(bb.msg.domain.PersistData, "Please set the 'PERSISTENT_DIR' or 'CACHE' variable.")
try:
os.stat(self.cachedir)
except OSError:
bb.mkdirhier(self.cachedir)
self.cachefile = os.path.join(self.cachedir,"bb_persist_data.sqlite3")
bb.msg.debug(1, bb.msg.domain.PersistData, "Using '%s' as the persistent data cache" % self.cachefile)
self.connection = sqlite3.connect(self.cachefile, timeout=5, isolation_level=None)
def addDomain(self, domain):
"""
Should be called before any domain is used
Creates it if it doesn't exist.
"""
self.connection.execute("CREATE TABLE IF NOT EXISTS %s(key TEXT, value TEXT);" % domain)
def delDomain(self, domain):
"""
Removes a domain and all the data it contains
"""
self.connection.execute("DROP TABLE IF EXISTS %s;" % domain)
def getKeyValues(self, domain):
"""
Return a list of key + value pairs for a domain
"""
ret = {}
data = self.connection.execute("SELECT key, value from %s;" % domain)
for row in data:
ret[str(row[0])] = str(row[1])
return ret
def getValue(self, domain, key):
"""
Return the value of a key for a domain
"""
data = self.connection.execute("SELECT * from %s where key=?;" % domain, [key])
for row in data:
return row[1]
def setValue(self, domain, key, value):
"""
Sets the value of a key for a domain
"""
data = self.connection.execute("SELECT * from %s where key=?;" % domain, [key])
rows = 0
for row in data:
rows = rows + 1
if rows:
self._execute("UPDATE %s SET value=? WHERE key=?;" % domain, [value, key])
else:
self._execute("INSERT into %s(key, value) values (?, ?);" % domain, [key, value])
def delValue(self, domain, key):
"""
Deletes a key/value pair
"""
self._execute("DELETE from %s where key=?;" % domain, [key])
def _execute(self, *query):
while True:
try:
self.connection.execute(*query)
return
except sqlite3.OperationalError, e:
if 'database is locked' in str(e):
continue
raise

View File

@@ -0,0 +1,315 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2003, 2004 Chris Larson
# Copyright (C) 2003, 2004 Phil Blundell
# Copyright (C) 2003 - 2005 Michael 'Mickey' Lauer
# Copyright (C) 2005 Holger Hans Peter Freyther
# Copyright (C) 2005 ROAD GmbH
# Copyright (C) 2006 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import re
from bb import data, utils
import bb
class NoProvider(Exception):
"""Exception raised when no provider of a build dependency can be found"""
class NoRProvider(Exception):
"""Exception raised when no provider of a runtime dependency can be found"""
def sortPriorities(pn, dataCache, pkg_pn = None):
"""
Reorder pkg_pn by file priority and default preference
"""
if not pkg_pn:
pkg_pn = dataCache.pkg_pn
files = pkg_pn[pn]
priorities = {}
for f in files:
priority = dataCache.bbfile_priority[f]
preference = dataCache.pkg_dp[f]
if priority not in priorities:
priorities[priority] = {}
if preference not in priorities[priority]:
priorities[priority][preference] = []
priorities[priority][preference].append(f)
pri_list = priorities.keys()
pri_list.sort(lambda a, b: a - b)
tmp_pn = []
for pri in pri_list:
pref_list = priorities[pri].keys()
pref_list.sort(lambda a, b: b - a)
tmp_pref = []
for pref in pref_list:
tmp_pref.extend(priorities[pri][pref])
tmp_pn = [tmp_pref] + tmp_pn
return tmp_pn
def findPreferredProvider(pn, cfgData, dataCache, pkg_pn = None, item = None):
"""
Find the first provider in pkg_pn with a PREFERRED_VERSION set.
"""
preferred_file = None
preferred_ver = None
localdata = data.createCopy(cfgData)
bb.data.setVar('OVERRIDES', "pn-%s:%s:%s" % (pn, pn, data.getVar('OVERRIDES', localdata)), localdata)
bb.data.update_data(localdata)
preferred_v = bb.data.getVar('PREFERRED_VERSION_%s' % pn, localdata, True)
if preferred_v:
m = re.match('(\d+:)*(.*)(_.*)*', preferred_v)
if m:
if m.group(1):
preferred_e = int(m.group(1)[:-1])
else:
preferred_e = None
preferred_v = m.group(2)
if m.group(3):
preferred_r = m.group(3)[1:]
else:
preferred_r = None
else:
preferred_e = None
preferred_r = None
for file_set in pkg_pn:
for f in file_set:
pe,pv,pr = dataCache.pkg_pepvpr[f]
if preferred_v == pv and (preferred_r == pr or preferred_r == None) and (preferred_e == pe or preferred_e == None):
preferred_file = f
preferred_ver = (pe, pv, pr)
break
if preferred_file:
break;
if preferred_r:
pv_str = '%s-%s' % (preferred_v, preferred_r)
else:
pv_str = preferred_v
if not (preferred_e is None):
pv_str = '%s:%s' % (preferred_e, pv_str)
itemstr = ""
if item:
itemstr = " (for item %s)" % item
if preferred_file is None:
bb.msg.note(1, bb.msg.domain.Provider, "preferred version %s of %s not available%s" % (pv_str, pn, itemstr))
else:
bb.msg.debug(1, bb.msg.domain.Provider, "selecting %s as PREFERRED_VERSION %s of package %s%s" % (preferred_file, pv_str, pn, itemstr))
return (preferred_ver, preferred_file)
def findLatestProvider(pn, cfgData, dataCache, file_set):
"""
Return the highest version of the providers in file_set.
Take default preferences into account.
"""
latest = None
latest_p = 0
latest_f = None
for file_name in file_set:
pe,pv,pr = dataCache.pkg_pepvpr[file_name]
dp = dataCache.pkg_dp[file_name]
if (latest is None) or ((latest_p == dp) and (utils.vercmp(latest, (pe, pv, pr)) < 0)) or (dp > latest_p):
latest = (pe, pv, pr)
latest_f = file_name
latest_p = dp
return (latest, latest_f)
def findBestProvider(pn, cfgData, dataCache, pkg_pn = None, item = None):
"""
If there is a PREFERRED_VERSION, find the highest-priority bbfile
providing that version. If not, find the latest version provided by
an bbfile in the highest-priority set.
"""
sortpkg_pn = sortPriorities(pn, dataCache, pkg_pn)
# Find the highest priority provider with a PREFERRED_VERSION set
(preferred_ver, preferred_file) = findPreferredProvider(pn, cfgData, dataCache, sortpkg_pn, item)
# Find the latest version of the highest priority provider
(latest, latest_f) = findLatestProvider(pn, cfgData, dataCache, sortpkg_pn[0])
if preferred_file is None:
preferred_file = latest_f
preferred_ver = latest
return (latest, latest_f, preferred_ver, preferred_file)
def _filterProviders(providers, item, cfgData, dataCache):
"""
Take a list of providers and filter/reorder according to the
environment variables and previous build results
"""
eligible = []
preferred_versions = {}
sortpkg_pn = {}
# The order of providers depends on the order of the files on the disk
# up to here. Sort pkg_pn to make dependency issues reproducible rather
# than effectively random.
providers.sort()
# Collate providers by PN
pkg_pn = {}
for p in providers:
pn = dataCache.pkg_fn[p]
if pn not in pkg_pn:
pkg_pn[pn] = []
pkg_pn[pn].append(p)
bb.msg.debug(1, bb.msg.domain.Provider, "providers for %s are: %s" % (item, pkg_pn.keys()))
# First add PREFERRED_VERSIONS
for pn in pkg_pn.keys():
sortpkg_pn[pn] = sortPriorities(pn, dataCache, pkg_pn)
preferred_versions[pn] = findPreferredProvider(pn, cfgData, dataCache, sortpkg_pn[pn], item)
if preferred_versions[pn][1]:
eligible.append(preferred_versions[pn][1])
# Now add latest verisons
for pn in pkg_pn.keys():
if pn in preferred_versions and preferred_versions[pn][1]:
continue
preferred_versions[pn] = findLatestProvider(pn, cfgData, dataCache, sortpkg_pn[pn][0])
eligible.append(preferred_versions[pn][1])
if len(eligible) == 0:
bb.msg.error(bb.msg.domain.Provider, "no eligible providers for %s" % item)
return 0
# If pn == item, give it a slight default preference
# This means PREFERRED_PROVIDER_foobar defaults to foobar if available
for p in providers:
pn = dataCache.pkg_fn[p]
if pn != item:
continue
(newvers, fn) = preferred_versions[pn]
if not fn in eligible:
continue
eligible.remove(fn)
eligible = [fn] + eligible
return eligible
def filterProviders(providers, item, cfgData, dataCache):
"""
Take a list of providers and filter/reorder according to the
environment variables and previous build results
Takes a "normal" target item
"""
eligible = _filterProviders(providers, item, cfgData, dataCache)
prefervar = bb.data.getVar('PREFERRED_PROVIDER_%s' % item, cfgData, 1)
if prefervar:
dataCache.preferred[item] = prefervar
foundUnique = False
if item in dataCache.preferred:
for p in eligible:
pn = dataCache.pkg_fn[p]
if dataCache.preferred[item] == pn:
bb.msg.note(2, bb.msg.domain.Provider, "selecting %s to satisfy %s due to PREFERRED_PROVIDERS" % (pn, item))
eligible.remove(p)
eligible = [p] + eligible
foundUnique = True
break
bb.msg.debug(1, bb.msg.domain.Provider, "sorted providers for %s are: %s" % (item, eligible))
return eligible, foundUnique
def filterProvidersRunTime(providers, item, cfgData, dataCache):
"""
Take a list of providers and filter/reorder according to the
environment variables and previous build results
Takes a "runtime" target item
"""
eligible = _filterProviders(providers, item, cfgData, dataCache)
# Should use dataCache.preferred here?
preferred = []
preferred_vars = []
for p in eligible:
pn = dataCache.pkg_fn[p]
provides = dataCache.pn_provides[pn]
for provide in provides:
bb.msg.note(2, bb.msg.domain.Provider, "checking PREFERRED_PROVIDER_%s" % (provide))
prefervar = bb.data.getVar('PREFERRED_PROVIDER_%s' % provide, cfgData, 1)
if prefervar == pn:
var = "PREFERRED_PROVIDER_%s = %s" % (provide, prefervar)
bb.msg.note(2, bb.msg.domain.Provider, "selecting %s to satisfy runtime %s due to %s" % (pn, item, var))
preferred_vars.append(var)
eligible.remove(p)
eligible = [p] + eligible
preferred.append(p)
break
numberPreferred = len(preferred)
if numberPreferred > 1:
bb.msg.error(bb.msg.domain.Provider, "Conflicting PREFERRED_PROVIDER entries were found which resulted in an attempt to select multiple providers (%s) for runtime dependecy %s\nThe entries resulting in this conflict were: %s" % (preferred, item, preferred_vars))
bb.msg.debug(1, bb.msg.domain.Provider, "sorted providers for %s are: %s" % (item, eligible))
return eligible, numberPreferred
regexp_cache = {}
def getRuntimeProviders(dataCache, rdepend):
"""
Return any providers of runtime dependency
"""
rproviders = []
if rdepend in dataCache.rproviders:
rproviders += dataCache.rproviders[rdepend]
if rdepend in dataCache.packages:
rproviders += dataCache.packages[rdepend]
if rproviders:
return rproviders
# Only search dynamic packages if we can't find anything in other variables
for pattern in dataCache.packages_dynamic:
pattern = pattern.replace('+', "\+")
if pattern in regexp_cache:
regexp = regexp_cache[pattern]
else:
try:
regexp = re.compile(pattern)
except:
bb.msg.error(bb.msg.domain.Provider, "Error parsing re expression: %s" % pattern)
raise
regexp_cache[pattern] = regexp
if regexp.match(rdepend):
rproviders += dataCache.packages_dynamic[pattern]
return rproviders

File diff suppressed because it is too large Load Diff

824
bitbake-dev/lib/bb/shell.py Normal file
View File

@@ -0,0 +1,824 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
##########################################################################
#
# Copyright (C) 2005-2006 Michael 'Mickey' Lauer <mickey@Vanille.de>
# Copyright (C) 2005-2006 Vanille Media
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
##########################################################################
#
# Thanks to:
# * Holger Freyther <zecke@handhelds.org>
# * Justin Patrin <papercrane@reversefold.com>
#
##########################################################################
"""
BitBake Shell
IDEAS:
* list defined tasks per package
* list classes
* toggle force
* command to reparse just one (or more) bbfile(s)
* automatic check if reparsing is necessary (inotify?)
* frontend for bb file manipulation
* more shell-like features:
- output control, i.e. pipe output into grep, sort, etc.
- job control, i.e. bring running commands into background and foreground
* start parsing in background right after startup
* ncurses interface
PROBLEMS:
* force doesn't always work
* readline completion for commands with more than one parameters
"""
##########################################################################
# Import and setup global variables
##########################################################################
try:
set
except NameError:
from sets import Set as set
import sys, os, readline, socket, httplib, urllib, commands, popen2, copy, shlex, Queue, fnmatch
from bb import data, parse, build, fatal, cache, taskdata, runqueue, providers as Providers
__version__ = "0.5.3.1"
__credits__ = """BitBake Shell Version %s (C) 2005 Michael 'Mickey' Lauer <mickey@Vanille.de>
Type 'help' for more information, press CTRL-D to exit.""" % __version__
cmds = {}
leave_mainloop = False
last_exception = None
cooker = None
parsed = False
debug = os.environ.get( "BBSHELL_DEBUG", "" )
##########################################################################
# Class BitBakeShellCommands
##########################################################################
class BitBakeShellCommands:
"""This class contains the valid commands for the shell"""
def __init__( self, shell ):
"""Register all the commands"""
self._shell = shell
for attr in BitBakeShellCommands.__dict__:
if not attr.startswith( "_" ):
if attr.endswith( "_" ):
command = attr[:-1].lower()
else:
command = attr[:].lower()
method = getattr( BitBakeShellCommands, attr )
debugOut( "registering command '%s'" % command )
# scan number of arguments
usage = getattr( method, "usage", "" )
if usage != "<...>":
numArgs = len( usage.split() )
else:
numArgs = -1
shell.registerCommand( command, method, numArgs, "%s %s" % ( command, usage ), method.__doc__ )
def _checkParsed( self ):
if not parsed:
print "SHELL: This command needs to parse bbfiles..."
self.parse( None )
def _findProvider( self, item ):
self._checkParsed()
# Need to use taskData for this information
preferred = data.getVar( "PREFERRED_PROVIDER_%s" % item, cooker.configuration.data, 1 )
if not preferred: preferred = item
try:
lv, lf, pv, pf = Providers.findBestProvider(preferred, cooker.configuration.data, cooker.status)
except KeyError:
if item in cooker.status.providers:
pf = cooker.status.providers[item][0]
else:
pf = None
return pf
def alias( self, params ):
"""Register a new name for a command"""
new, old = params
if not old in cmds:
print "ERROR: Command '%s' not known" % old
else:
cmds[new] = cmds[old]
print "OK"
alias.usage = "<alias> <command>"
def buffer( self, params ):
"""Dump specified output buffer"""
index = params[0]
print self._shell.myout.buffer( int( index ) )
buffer.usage = "<index>"
def buffers( self, params ):
"""Show the available output buffers"""
commands = self._shell.myout.bufferedCommands()
if not commands:
print "SHELL: No buffered commands available yet. Start doing something."
else:
print "="*35, "Available Output Buffers", "="*27
for index, cmd in enumerate( commands ):
print "| %s %s" % ( str( index ).ljust( 3 ), cmd )
print "="*88
def build( self, params, cmd = "build" ):
"""Build a providee"""
global last_exception
globexpr = params[0]
self._checkParsed()
names = globfilter( cooker.status.pkg_pn.keys(), globexpr )
if len( names ) == 0: names = [ globexpr ]
print "SHELL: Building %s" % ' '.join( names )
td = taskdata.TaskData(cooker.configuration.abort)
localdata = data.createCopy(cooker.configuration.data)
data.update_data(localdata)
data.expandKeys(localdata)
try:
tasks = []
for name in names:
td.add_provider(localdata, cooker.status, name)
providers = td.get_provider(name)
if len(providers) == 0:
raise Providers.NoProvider
tasks.append([name, "do_%s" % cmd])
td.add_unresolved(localdata, cooker.status)
rq = runqueue.RunQueue(cooker, localdata, cooker.status, td, tasks)
rq.prepare_runqueue()
rq.execute_runqueue()
except Providers.NoProvider:
print "ERROR: No Provider"
last_exception = Providers.NoProvider
except runqueue.TaskFailure, fnids:
for fnid in fnids:
print "ERROR: '%s' failed" % td.fn_index[fnid]
last_exception = runqueue.TaskFailure
except build.EventException, e:
print "ERROR: Couldn't build '%s'" % names
last_exception = e
build.usage = "<providee>"
def clean( self, params ):
"""Clean a providee"""
self.build( params, "clean" )
clean.usage = "<providee>"
def compile( self, params ):
"""Execute 'compile' on a providee"""
self.build( params, "compile" )
compile.usage = "<providee>"
def configure( self, params ):
"""Execute 'configure' on a providee"""
self.build( params, "configure" )
configure.usage = "<providee>"
def install( self, params ):
"""Execute 'install' on a providee"""
self.build( params, "install" )
install.usage = "<providee>"
def edit( self, params ):
"""Call $EDITOR on a providee"""
name = params[0]
bbfile = self._findProvider( name )
if bbfile is not None:
os.system( "%s %s" % ( os.environ.get( "EDITOR", "vi" ), bbfile ) )
else:
print "ERROR: Nothing provides '%s'" % name
edit.usage = "<providee>"
def environment( self, params ):
"""Dump out the outer BitBake environment"""
cooker.showEnvironment()
def exit_( self, params ):
"""Leave the BitBake Shell"""
debugOut( "setting leave_mainloop to true" )
global leave_mainloop
leave_mainloop = True
def fetch( self, params ):
"""Fetch a providee"""
self.build( params, "fetch" )
fetch.usage = "<providee>"
def fileBuild( self, params, cmd = "build" ):
"""Parse and build a .bb file"""
global last_exception
name = params[0]
bf = completeFilePath( name )
print "SHELL: Calling '%s' on '%s'" % ( cmd, bf )
try:
cooker.buildFile(bf, cmd)
except parse.ParseError:
print "ERROR: Unable to open or parse '%s'" % bf
except build.EventException, e:
print "ERROR: Couldn't build '%s'" % name
last_exception = e
fileBuild.usage = "<bbfile>"
def fileClean( self, params ):
"""Clean a .bb file"""
self.fileBuild( params, "clean" )
fileClean.usage = "<bbfile>"
def fileEdit( self, params ):
"""Call $EDITOR on a .bb file"""
name = params[0]
os.system( "%s %s" % ( os.environ.get( "EDITOR", "vi" ), completeFilePath( name ) ) )
fileEdit.usage = "<bbfile>"
def fileRebuild( self, params ):
"""Rebuild (clean & build) a .bb file"""
self.fileBuild( params, "rebuild" )
fileRebuild.usage = "<bbfile>"
def fileReparse( self, params ):
"""(re)Parse a bb file"""
bbfile = params[0]
print "SHELL: Parsing '%s'" % bbfile
parse.update_mtime( bbfile )
cooker.bb_cache.cacheValidUpdate(bbfile)
fromCache = cooker.bb_cache.loadData(bbfile, cooker.configuration.data, cooker.status)
cooker.bb_cache.sync()
if False: #fromCache:
print "SHELL: File has not been updated, not reparsing"
else:
print "SHELL: Parsed"
fileReparse.usage = "<bbfile>"
def abort( self, params ):
"""Toggle abort task execution flag (see bitbake -k)"""
cooker.configuration.abort = not cooker.configuration.abort
print "SHELL: Abort Flag is now '%s'" % repr( cooker.configuration.abort )
def force( self, params ):
"""Toggle force task execution flag (see bitbake -f)"""
cooker.configuration.force = not cooker.configuration.force
print "SHELL: Force Flag is now '%s'" % repr( cooker.configuration.force )
def help( self, params ):
"""Show a comprehensive list of commands and their purpose"""
print "="*30, "Available Commands", "="*30
allcmds = cmds.keys()
allcmds.sort()
for cmd in allcmds:
function,numparams,usage,helptext = cmds[cmd]
print "| %s | %s" % (usage.ljust(30), helptext)
print "="*78
def lastError( self, params ):
"""Show the reason or log that was produced by the last BitBake event exception"""
if last_exception is None:
print "SHELL: No Errors yet (Phew)..."
else:
reason, event = last_exception.args
print "SHELL: Reason for the last error: '%s'" % reason
if ':' in reason:
msg, filename = reason.split( ':' )
filename = filename.strip()
print "SHELL: Dumping log file for last error:"
try:
print open( filename ).read()
except IOError:
print "ERROR: Couldn't open '%s'" % filename
def match( self, params ):
"""Dump all files or providers matching a glob expression"""
what, globexpr = params
if what == "files":
self._checkParsed()
for key in globfilter( cooker.status.pkg_fn.keys(), globexpr ): print key
elif what == "providers":
self._checkParsed()
for key in globfilter( cooker.status.pkg_pn.keys(), globexpr ): print key
else:
print "Usage: match %s" % self.print_.usage
match.usage = "<files|providers> <glob>"
def new( self, params ):
"""Create a new .bb file and open the editor"""
dirname, filename = params
packages = '/'.join( data.getVar( "BBFILES", cooker.configuration.data, 1 ).split('/')[:-2] )
fulldirname = "%s/%s" % ( packages, dirname )
if not os.path.exists( fulldirname ):
print "SHELL: Creating '%s'" % fulldirname
os.mkdir( fulldirname )
if os.path.exists( fulldirname ) and os.path.isdir( fulldirname ):
if os.path.exists( "%s/%s" % ( fulldirname, filename ) ):
print "SHELL: ERROR: %s/%s already exists" % ( fulldirname, filename )
return False
print "SHELL: Creating '%s/%s'" % ( fulldirname, filename )
newpackage = open( "%s/%s" % ( fulldirname, filename ), "w" )
print >>newpackage,"""DESCRIPTION = ""
SECTION = ""
AUTHOR = ""
HOMEPAGE = ""
MAINTAINER = ""
LICENSE = "GPL"
PR = "r0"
SRC_URI = ""
#inherit base
#do_configure() {
#
#}
#do_compile() {
#
#}
#do_stage() {
#
#}
#do_install() {
#
#}
"""
newpackage.close()
os.system( "%s %s/%s" % ( os.environ.get( "EDITOR" ), fulldirname, filename ) )
new.usage = "<directory> <filename>"
def package( self, params ):
"""Execute 'package' on a providee"""
self.build( params, "package" )
package.usage = "<providee>"
def pasteBin( self, params ):
"""Send a command + output buffer to the pastebin at http://rafb.net/paste"""
index = params[0]
contents = self._shell.myout.buffer( int( index ) )
sendToPastebin( "output of " + params[0], contents )
pasteBin.usage = "<index>"
def pasteLog( self, params ):
"""Send the last event exception error log (if there is one) to http://rafb.net/paste"""
if last_exception is None:
print "SHELL: No Errors yet (Phew)..."
else:
reason, event = last_exception.args
print "SHELL: Reason for the last error: '%s'" % reason
if ':' in reason:
msg, filename = reason.split( ':' )
filename = filename.strip()
print "SHELL: Pasting log file to pastebin..."
file = open( filename ).read()
sendToPastebin( "contents of " + filename, file )
def patch( self, params ):
"""Execute 'patch' command on a providee"""
self.build( params, "patch" )
patch.usage = "<providee>"
def parse( self, params ):
"""(Re-)parse .bb files and calculate the dependency graph"""
cooker.status = cache.CacheData()
ignore = data.getVar("ASSUME_PROVIDED", cooker.configuration.data, 1) or ""
cooker.status.ignored_dependencies = set( ignore.split() )
cooker.handleCollections( data.getVar("BBFILE_COLLECTIONS", cooker.configuration.data, 1) )
(filelist, masked) = cooker.collect_bbfiles()
cooker.parse_bbfiles(filelist, masked, cooker.myProgressCallback)
cooker.buildDepgraph()
global parsed
parsed = True
print
def reparse( self, params ):
"""(re)Parse a providee's bb file"""
bbfile = self._findProvider( params[0] )
if bbfile is not None:
print "SHELL: Found bbfile '%s' for '%s'" % ( bbfile, params[0] )
self.fileReparse( [ bbfile ] )
else:
print "ERROR: Nothing provides '%s'" % params[0]
reparse.usage = "<providee>"
def getvar( self, params ):
"""Dump the contents of an outer BitBake environment variable"""
var = params[0]
value = data.getVar( var, cooker.configuration.data, 1 )
print value
getvar.usage = "<variable>"
def peek( self, params ):
"""Dump contents of variable defined in providee's metadata"""
name, var = params
bbfile = self._findProvider( name )
if bbfile is not None:
the_data = cooker.bb_cache.loadDataFull(bbfile, cooker.configuration.data)
value = the_data.getVar( var, 1 )
print value
else:
print "ERROR: Nothing provides '%s'" % name
peek.usage = "<providee> <variable>"
def poke( self, params ):
"""Set contents of variable defined in providee's metadata"""
name, var, value = params
bbfile = self._findProvider( name )
if bbfile is not None:
print "ERROR: Sorry, this functionality is currently broken"
#d = cooker.pkgdata[bbfile]
#data.setVar( var, value, d )
# mark the change semi persistant
#cooker.pkgdata.setDirty(bbfile, d)
#print "OK"
else:
print "ERROR: Nothing provides '%s'" % name
poke.usage = "<providee> <variable> <value>"
def print_( self, params ):
"""Dump all files or providers"""
what = params[0]
if what == "files":
self._checkParsed()
for key in cooker.status.pkg_fn.keys(): print key
elif what == "providers":
self._checkParsed()
for key in cooker.status.providers.keys(): print key
else:
print "Usage: print %s" % self.print_.usage
print_.usage = "<files|providers>"
def python( self, params ):
"""Enter the expert mode - an interactive BitBake Python Interpreter"""
sys.ps1 = "EXPERT BB>>> "
sys.ps2 = "EXPERT BB... "
import code
interpreter = code.InteractiveConsole( dict( globals() ) )
interpreter.interact( "SHELL: Expert Mode - BitBake Python %s\nType 'help' for more information, press CTRL-D to switch back to BBSHELL." % sys.version )
def showdata( self, params ):
"""Execute 'showdata' on a providee"""
cooker.showEnvironment(None, params)
showdata.usage = "<providee>"
def setVar( self, params ):
"""Set an outer BitBake environment variable"""
var, value = params
data.setVar( var, value, cooker.configuration.data )
print "OK"
setVar.usage = "<variable> <value>"
def rebuild( self, params ):
"""Clean and rebuild a .bb file or a providee"""
self.build( params, "clean" )
self.build( params, "build" )
rebuild.usage = "<providee>"
def shell( self, params ):
"""Execute a shell command and dump the output"""
if params != "":
print commands.getoutput( " ".join( params ) )
shell.usage = "<...>"
def stage( self, params ):
"""Execute 'stage' on a providee"""
self.build( params, "populate_staging" )
stage.usage = "<providee>"
def status( self, params ):
"""<just for testing>"""
print "-" * 78
print "building list = '%s'" % cooker.building_list
print "build path = '%s'" % cooker.build_path
print "consider_msgs_cache = '%s'" % cooker.consider_msgs_cache
print "build stats = '%s'" % cooker.stats
if last_exception is not None: print "last_exception = '%s'" % repr( last_exception.args )
print "memory output contents = '%s'" % self._shell.myout._buffer
def test( self, params ):
"""<just for testing>"""
print "testCommand called with '%s'" % params
def unpack( self, params ):
"""Execute 'unpack' on a providee"""
self.build( params, "unpack" )
unpack.usage = "<providee>"
def which( self, params ):
"""Computes the providers for a given providee"""
# Need to use taskData for this information
item = params[0]
self._checkParsed()
preferred = data.getVar( "PREFERRED_PROVIDER_%s" % item, cooker.configuration.data, 1 )
if not preferred: preferred = item
try:
lv, lf, pv, pf = Providers.findBestProvider(preferred, cooker.configuration.data, cooker.status)
except KeyError:
lv, lf, pv, pf = (None,)*4
try:
providers = cooker.status.providers[item]
except KeyError:
print "SHELL: ERROR: Nothing provides", preferred
else:
for provider in providers:
if provider == pf: provider = " (***) %s" % provider
else: provider = " %s" % provider
print provider
which.usage = "<providee>"
##########################################################################
# Common helper functions
##########################################################################
def completeFilePath( bbfile ):
"""Get the complete bbfile path"""
if not cooker.status: return bbfile
if not cooker.status.pkg_fn: return bbfile
for key in cooker.status.pkg_fn.keys():
if key.endswith( bbfile ):
return key
return bbfile
def sendToPastebin( desc, content ):
"""Send content to http://oe.pastebin.com"""
mydata = {}
mydata["lang"] = "Plain Text"
mydata["desc"] = desc
mydata["cvt_tabs"] = "No"
mydata["nick"] = "%s@%s" % ( os.environ.get( "USER", "unknown" ), socket.gethostname() or "unknown" )
mydata["text"] = content
params = urllib.urlencode( mydata )
headers = {"Content-type": "application/x-www-form-urlencoded","Accept": "text/plain"}
host = "rafb.net"
conn = httplib.HTTPConnection( "%s:80" % host )
conn.request("POST", "/paste/paste.php", params, headers )
response = conn.getresponse()
conn.close()
if response.status == 302:
location = response.getheader( "location" ) or "unknown"
print "SHELL: Pasted to http://%s%s" % ( host, location )
else:
print "ERROR: %s %s" % ( response.status, response.reason )
def completer( text, state ):
"""Return a possible readline completion"""
debugOut( "completer called with text='%s', state='%d'" % ( text, state ) )
if state == 0:
line = readline.get_line_buffer()
if " " in line:
line = line.split()
# we are in second (or more) argument
if line[0] in cmds and hasattr( cmds[line[0]][0], "usage" ): # known command and usage
u = getattr( cmds[line[0]][0], "usage" ).split()[0]
if u == "<variable>":
allmatches = cooker.configuration.data.keys()
elif u == "<bbfile>":
if cooker.status.pkg_fn is None: allmatches = [ "(No Matches Available. Parsed yet?)" ]
else: allmatches = [ x.split("/")[-1] for x in cooker.status.pkg_fn.keys() ]
elif u == "<providee>":
if cooker.status.pkg_fn is None: allmatches = [ "(No Matches Available. Parsed yet?)" ]
else: allmatches = cooker.status.providers.iterkeys()
else: allmatches = [ "(No tab completion available for this command)" ]
else: allmatches = [ "(No tab completion available for this command)" ]
else:
# we are in first argument
allmatches = cmds.iterkeys()
completer.matches = [ x for x in allmatches if x[:len(text)] == text ]
#print "completer.matches = '%s'" % completer.matches
if len( completer.matches ) > state:
return completer.matches[state]
else:
return None
def debugOut( text ):
if debug:
sys.stderr.write( "( %s )\n" % text )
def columnize( alist, width = 80 ):
"""
A word-wrap function that preserves existing line breaks
and most spaces in the text. Expects that existing line
breaks are posix newlines (\n).
"""
return reduce(lambda line, word, width=width: '%s%s%s' %
(line,
' \n'[(len(line[line.rfind('\n')+1:])
+ len(word.split('\n',1)[0]
) >= width)],
word),
alist
)
def globfilter( names, pattern ):
return fnmatch.filter( names, pattern )
##########################################################################
# Class MemoryOutput
##########################################################################
class MemoryOutput:
"""File-like output class buffering the output of the last 10 commands"""
def __init__( self, delegate ):
self.delegate = delegate
self._buffer = []
self.text = []
self._command = None
def startCommand( self, command ):
self._command = command
self.text = []
def endCommand( self ):
if self._command is not None:
if len( self._buffer ) == 10: del self._buffer[0]
self._buffer.append( ( self._command, self.text ) )
def removeLast( self ):
if self._buffer:
del self._buffer[ len( self._buffer ) - 1 ]
self.text = []
self._command = None
def lastBuffer( self ):
if self._buffer:
return self._buffer[ len( self._buffer ) -1 ][1]
def bufferedCommands( self ):
return [ cmd for cmd, output in self._buffer ]
def buffer( self, i ):
if i < len( self._buffer ):
return "BB>> %s\n%s" % ( self._buffer[i][0], "".join( self._buffer[i][1] ) )
else: return "ERROR: Invalid buffer number. Buffer needs to be in (0, %d)" % ( len( self._buffer ) - 1 )
def write( self, text ):
if self._command is not None and text != "BB>> ": self.text.append( text )
if self.delegate is not None: self.delegate.write( text )
def flush( self ):
return self.delegate.flush()
def fileno( self ):
return self.delegate.fileno()
def isatty( self ):
return self.delegate.isatty()
##########################################################################
# Class BitBakeShell
##########################################################################
class BitBakeShell:
def __init__( self ):
"""Register commands and set up readline"""
self.commandQ = Queue.Queue()
self.commands = BitBakeShellCommands( self )
self.myout = MemoryOutput( sys.stdout )
self.historyfilename = os.path.expanduser( "~/.bbsh_history" )
self.startupfilename = os.path.expanduser( "~/.bbsh_startup" )
readline.set_completer( completer )
readline.set_completer_delims( " " )
readline.parse_and_bind("tab: complete")
try:
readline.read_history_file( self.historyfilename )
except IOError:
pass # It doesn't exist yet.
print __credits__
def cleanup( self ):
"""Write readline history and clean up resources"""
debugOut( "writing command history" )
try:
readline.write_history_file( self.historyfilename )
except:
print "SHELL: Unable to save command history"
def registerCommand( self, command, function, numparams = 0, usage = "", helptext = "" ):
"""Register a command"""
if usage == "": usage = command
if helptext == "": helptext = function.__doc__ or "<not yet documented>"
cmds[command] = ( function, numparams, usage, helptext )
def processCommand( self, command, params ):
"""Process a command. Check number of params and print a usage string, if appropriate"""
debugOut( "processing command '%s'..." % command )
try:
function, numparams, usage, helptext = cmds[command]
except KeyError:
print "SHELL: ERROR: '%s' command is not a valid command." % command
self.myout.removeLast()
else:
if (numparams != -1) and (not len( params ) == numparams):
print "Usage: '%s'" % usage
return
result = function( self.commands, params )
debugOut( "result was '%s'" % result )
def processStartupFile( self ):
"""Read and execute all commands found in $HOME/.bbsh_startup"""
if os.path.exists( self.startupfilename ):
startupfile = open( self.startupfilename, "r" )
for cmdline in startupfile:
debugOut( "processing startup line '%s'" % cmdline )
if not cmdline:
continue
if "|" in cmdline:
print "ERROR: '|' in startup file is not allowed. Ignoring line"
continue
self.commandQ.put( cmdline.strip() )
def main( self ):
"""The main command loop"""
while not leave_mainloop:
try:
if self.commandQ.empty():
sys.stdout = self.myout.delegate
cmdline = raw_input( "BB>> " )
sys.stdout = self.myout
else:
cmdline = self.commandQ.get()
if cmdline:
allCommands = cmdline.split( ';' )
for command in allCommands:
pipecmd = None
#
# special case for expert mode
if command == 'python':
sys.stdout = self.myout.delegate
self.processCommand( command, "" )
sys.stdout = self.myout
else:
self.myout.startCommand( command )
if '|' in command: # disable output
command, pipecmd = command.split( '|' )
delegate = self.myout.delegate
self.myout.delegate = None
tokens = shlex.split( command, True )
self.processCommand( tokens[0], tokens[1:] or "" )
self.myout.endCommand()
if pipecmd is not None: # restore output
self.myout.delegate = delegate
pipe = popen2.Popen4( pipecmd )
pipe.tochild.write( "\n".join( self.myout.lastBuffer() ) )
pipe.tochild.close()
sys.stdout.write( pipe.fromchild.read() )
#
except EOFError:
print
return
except KeyboardInterrupt:
print
##########################################################################
# Start function - called from the BitBake command line utility
##########################################################################
def start( aCooker ):
global cooker
cooker = aCooker
bbshell = BitBakeShell()
bbshell.processStartupFile()
bbshell.main()
bbshell.cleanup()
if __name__ == "__main__":
print "SHELL: Sorry, this program should only be called by BitBake."

View File

@@ -0,0 +1,595 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake 'TaskData' implementation
Task data collection and handling
"""
# Copyright (C) 2006 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import bb
class TaskData:
"""
BitBake Task Data implementation
"""
def __init__(self, abort = True, tryaltconfigs = False):
self.build_names_index = []
self.run_names_index = []
self.fn_index = []
self.build_targets = {}
self.run_targets = {}
self.external_targets = []
self.tasks_fnid = []
self.tasks_name = []
self.tasks_tdepends = []
self.tasks_idepends = []
# Cache to speed up task ID lookups
self.tasks_lookup = {}
self.depids = {}
self.rdepids = {}
self.consider_msgs_cache = []
self.failed_deps = []
self.failed_rdeps = []
self.failed_fnids = []
self.abort = abort
self.tryaltconfigs = tryaltconfigs
def getbuild_id(self, name):
"""
Return an ID number for the build target name.
If it doesn't exist, create one.
"""
if not name in self.build_names_index:
self.build_names_index.append(name)
return len(self.build_names_index) - 1
return self.build_names_index.index(name)
def getrun_id(self, name):
"""
Return an ID number for the run target name.
If it doesn't exist, create one.
"""
if not name in self.run_names_index:
self.run_names_index.append(name)
return len(self.run_names_index) - 1
return self.run_names_index.index(name)
def getfn_id(self, name):
"""
Return an ID number for the filename.
If it doesn't exist, create one.
"""
if not name in self.fn_index:
self.fn_index.append(name)
return len(self.fn_index) - 1
return self.fn_index.index(name)
def gettask_ids(self, fnid):
"""
Return an array of the ID numbers matching a given fnid.
"""
ids = []
if fnid in self.tasks_lookup:
for task in self.tasks_lookup[fnid]:
ids.append(self.tasks_lookup[fnid][task])
return ids
def gettask_id(self, fn, task, create = True):
"""
Return an ID number for the task matching fn and task.
If it doesn't exist, create one by default.
Optionally return None instead.
"""
fnid = self.getfn_id(fn)
if fnid in self.tasks_lookup:
if task in self.tasks_lookup[fnid]:
return self.tasks_lookup[fnid][task]
if not create:
return None
self.tasks_name.append(task)
self.tasks_fnid.append(fnid)
self.tasks_tdepends.append([])
self.tasks_idepends.append([])
listid = len(self.tasks_name) - 1
if fnid not in self.tasks_lookup:
self.tasks_lookup[fnid] = {}
self.tasks_lookup[fnid][task] = listid
return listid
def add_tasks(self, fn, dataCache):
"""
Add tasks for a given fn to the database
"""
task_deps = dataCache.task_deps[fn]
fnid = self.getfn_id(fn)
if fnid in self.failed_fnids:
bb.msg.fatal(bb.msg.domain.TaskData, "Trying to re-add a failed file? Something is broken...")
# Check if we've already seen this fn
if fnid in self.tasks_fnid:
return
for task in task_deps['tasks']:
# Work out task dependencies
parentids = []
for dep in task_deps['parents'][task]:
parentid = self.gettask_id(fn, dep)
parentids.append(parentid)
taskid = self.gettask_id(fn, task)
self.tasks_tdepends[taskid].extend(parentids)
# Touch all intertask dependencies
if 'depends' in task_deps and task in task_deps['depends']:
ids = []
for dep in task_deps['depends'][task].split():
if dep:
ids.append(((self.getbuild_id(dep.split(":")[0])), dep.split(":")[1]))
self.tasks_idepends[taskid].extend(ids)
# Work out build dependencies
if not fnid in self.depids:
dependids = {}
for depend in dataCache.deps[fn]:
bb.msg.debug(2, bb.msg.domain.TaskData, "Added dependency %s for %s" % (depend, fn))
dependids[self.getbuild_id(depend)] = None
self.depids[fnid] = dependids.keys()
# Work out runtime dependencies
if not fnid in self.rdepids:
rdependids = {}
rdepends = dataCache.rundeps[fn]
rrecs = dataCache.runrecs[fn]
for package in rdepends:
for rdepend in bb.utils.explode_deps(rdepends[package]):
bb.msg.debug(2, bb.msg.domain.TaskData, "Added runtime dependency %s for %s" % (rdepend, fn))
rdependids[self.getrun_id(rdepend)] = None
for package in rrecs:
for rdepend in bb.utils.explode_deps(rrecs[package]):
bb.msg.debug(2, bb.msg.domain.TaskData, "Added runtime recommendation %s for %s" % (rdepend, fn))
rdependids[self.getrun_id(rdepend)] = None
self.rdepids[fnid] = rdependids.keys()
for dep in self.depids[fnid]:
if dep in self.failed_deps:
self.fail_fnid(fnid)
return
for dep in self.rdepids[fnid]:
if dep in self.failed_rdeps:
self.fail_fnid(fnid)
return
def have_build_target(self, target):
"""
Have we a build target matching this name?
"""
targetid = self.getbuild_id(target)
if targetid in self.build_targets:
return True
return False
def have_runtime_target(self, target):
"""
Have we a runtime target matching this name?
"""
targetid = self.getrun_id(target)
if targetid in self.run_targets:
return True
return False
def add_build_target(self, fn, item):
"""
Add a build target.
If already present, append the provider fn to the list
"""
targetid = self.getbuild_id(item)
fnid = self.getfn_id(fn)
if targetid in self.build_targets:
if fnid in self.build_targets[targetid]:
return
self.build_targets[targetid].append(fnid)
return
self.build_targets[targetid] = [fnid]
def add_runtime_target(self, fn, item):
"""
Add a runtime target.
If already present, append the provider fn to the list
"""
targetid = self.getrun_id(item)
fnid = self.getfn_id(fn)
if targetid in self.run_targets:
if fnid in self.run_targets[targetid]:
return
self.run_targets[targetid].append(fnid)
return
self.run_targets[targetid] = [fnid]
def mark_external_target(self, item):
"""
Mark a build target as being externally requested
"""
targetid = self.getbuild_id(item)
if targetid not in self.external_targets:
self.external_targets.append(targetid)
def get_unresolved_build_targets(self, dataCache):
"""
Return a list of build targets who's providers
are unknown.
"""
unresolved = []
for target in self.build_names_index:
if target in dataCache.ignored_dependencies:
continue
if self.build_names_index.index(target) in self.failed_deps:
continue
if not self.have_build_target(target):
unresolved.append(target)
return unresolved
def get_unresolved_run_targets(self, dataCache):
"""
Return a list of runtime targets who's providers
are unknown.
"""
unresolved = []
for target in self.run_names_index:
if target in dataCache.ignored_dependencies:
continue
if self.run_names_index.index(target) in self.failed_rdeps:
continue
if not self.have_runtime_target(target):
unresolved.append(target)
return unresolved
def get_provider(self, item):
"""
Return a list of providers of item
"""
targetid = self.getbuild_id(item)
return self.build_targets[targetid]
def get_dependees(self, itemid):
"""
Return a list of targets which depend on item
"""
dependees = []
for fnid in self.depids:
if itemid in self.depids[fnid]:
dependees.append(fnid)
return dependees
def get_dependees_str(self, item):
"""
Return a list of targets which depend on item as a user readable string
"""
itemid = self.getbuild_id(item)
dependees = []
for fnid in self.depids:
if itemid in self.depids[fnid]:
dependees.append(self.fn_index[fnid])
return dependees
def get_rdependees(self, itemid):
"""
Return a list of targets which depend on runtime item
"""
dependees = []
for fnid in self.rdepids:
if itemid in self.rdepids[fnid]:
dependees.append(fnid)
return dependees
def get_rdependees_str(self, item):
"""
Return a list of targets which depend on runtime item as a user readable string
"""
itemid = self.getrun_id(item)
dependees = []
for fnid in self.rdepids:
if itemid in self.rdepids[fnid]:
dependees.append(self.fn_index[fnid])
return dependees
def add_provider(self, cfgData, dataCache, item):
try:
self.add_provider_internal(cfgData, dataCache, item)
except bb.providers.NoProvider:
if self.abort:
if self.get_rdependees_str(item):
bb.msg.error(bb.msg.domain.Provider, "Nothing PROVIDES '%s' (but '%s' DEPENDS on or otherwise requires it)" % (item, self.get_dependees_str(item)))
else:
bb.msg.error(bb.msg.domain.Provider, "Nothing PROVIDES '%s'" % (item))
raise
targetid = self.getbuild_id(item)
self.remove_buildtarget(targetid)
self.mark_external_target(item)
def add_provider_internal(self, cfgData, dataCache, item):
"""
Add the providers of item to the task data
Mark entries were specifically added externally as against dependencies
added internally during dependency resolution
"""
if item in dataCache.ignored_dependencies:
return
if not item in dataCache.providers:
if self.get_rdependees_str(item):
bb.msg.note(2, bb.msg.domain.Provider, "Nothing PROVIDES '%s' (but '%s' DEPENDS on or otherwise requires it)" % (item, self.get_dependees_str(item)))
else:
bb.msg.note(2, bb.msg.domain.Provider, "Nothing PROVIDES '%s'" % (item))
bb.event.fire(bb.event.NoProvider(item, cfgData))
raise bb.providers.NoProvider(item)
if self.have_build_target(item):
return
all_p = dataCache.providers[item]
eligible, foundUnique = bb.providers.filterProviders(all_p, item, cfgData, dataCache)
eligible = [p for p in eligible if not self.getfn_id(p) in self.failed_fnids]
if not eligible:
bb.msg.note(2, bb.msg.domain.Provider, "No buildable provider PROVIDES '%s' but '%s' DEPENDS on or otherwise requires it. Enable debugging and see earlier logs to find unbuildable providers." % (item, self.get_dependees_str(item)))
bb.event.fire(bb.event.NoProvider(item, cfgData))
raise bb.providers.NoProvider(item)
if len(eligible) > 1 and foundUnique == False:
if item not in self.consider_msgs_cache:
providers_list = []
for fn in eligible:
providers_list.append(dataCache.pkg_fn[fn])
bb.msg.note(1, bb.msg.domain.Provider, "multiple providers are available for %s (%s);" % (item, ", ".join(providers_list)))
bb.msg.note(1, bb.msg.domain.Provider, "consider defining PREFERRED_PROVIDER_%s" % item)
bb.event.fire(bb.event.MultipleProviders(item, providers_list, cfgData))
self.consider_msgs_cache.append(item)
for fn in eligible:
fnid = self.getfn_id(fn)
if fnid in self.failed_fnids:
continue
bb.msg.debug(2, bb.msg.domain.Provider, "adding %s to satisfy %s" % (fn, item))
self.add_build_target(fn, item)
self.add_tasks(fn, dataCache)
#item = dataCache.pkg_fn[fn]
def add_rprovider(self, cfgData, dataCache, item):
"""
Add the runtime providers of item to the task data
(takes item names from RDEPENDS/PACKAGES namespace)
"""
if item in dataCache.ignored_dependencies:
return
if self.have_runtime_target(item):
return
all_p = bb.providers.getRuntimeProviders(dataCache, item)
if not all_p:
bb.msg.error(bb.msg.domain.Provider, "'%s' RDEPENDS/RRECOMMENDS or otherwise requires the runtime entity '%s' but it wasn't found in any PACKAGE or RPROVIDES variables" % (self.get_rdependees_str(item), item))
bb.event.fire(bb.event.NoProvider(item, cfgData, runtime=True))
raise bb.providers.NoRProvider(item)
eligible, numberPreferred = bb.providers.filterProvidersRunTime(all_p, item, cfgData, dataCache)
eligible = [p for p in eligible if not self.getfn_id(p) in self.failed_fnids]
if not eligible:
bb.msg.error(bb.msg.domain.Provider, "'%s' RDEPENDS/RRECOMMENDS or otherwise requires the runtime entity '%s' but it wasn't found in any PACKAGE or RPROVIDES variables of any buildable targets.\nEnable debugging and see earlier logs to find unbuildable targets." % (self.get_rdependees_str(item), item))
bb.event.fire(bb.event.NoProvider(item, cfgData, runtime=True))
raise bb.providers.NoRProvider(item)
if len(eligible) > 1 and numberPreferred == 0:
if item not in self.consider_msgs_cache:
providers_list = []
for fn in eligible:
providers_list.append(dataCache.pkg_fn[fn])
bb.msg.note(2, bb.msg.domain.Provider, "multiple providers are available for runtime %s (%s);" % (item, ", ".join(providers_list)))
bb.msg.note(2, bb.msg.domain.Provider, "consider defining a PREFERRED_PROVIDER entry to match runtime %s" % item)
bb.event.fire(bb.event.MultipleProviders(item,providers_list, cfgData, runtime=True))
self.consider_msgs_cache.append(item)
if numberPreferred > 1:
if item not in self.consider_msgs_cache:
providers_list = []
for fn in eligible:
providers_list.append(dataCache.pkg_fn[fn])
bb.msg.note(2, bb.msg.domain.Provider, "multiple providers are available for runtime %s (top %s entries preferred) (%s);" % (item, numberPreferred, ", ".join(providers_list)))
bb.msg.note(2, bb.msg.domain.Provider, "consider defining only one PREFERRED_PROVIDER entry to match runtime %s" % item)
bb.event.fire(bb.event.MultipleProviders(item,providers_list, cfgData, runtime=True))
self.consider_msgs_cache.append(item)
# run through the list until we find one that we can build
for fn in eligible:
fnid = self.getfn_id(fn)
if fnid in self.failed_fnids:
continue
bb.msg.debug(2, bb.msg.domain.Provider, "adding '%s' to satisfy runtime '%s'" % (fn, item))
self.add_runtime_target(fn, item)
self.add_tasks(fn, dataCache)
def fail_fnid(self, fnid, missing_list = []):
"""
Mark a file as failed (unbuildable)
Remove any references from build and runtime provider lists
missing_list, A list of missing requirements for this target
"""
if fnid in self.failed_fnids:
return
bb.msg.debug(1, bb.msg.domain.Provider, "File '%s' is unbuildable, removing..." % self.fn_index[fnid])
self.failed_fnids.append(fnid)
for target in self.build_targets:
if fnid in self.build_targets[target]:
self.build_targets[target].remove(fnid)
if len(self.build_targets[target]) == 0:
self.remove_buildtarget(target, missing_list)
for target in self.run_targets:
if fnid in self.run_targets[target]:
self.run_targets[target].remove(fnid)
if len(self.run_targets[target]) == 0:
self.remove_runtarget(target, missing_list)
def remove_buildtarget(self, targetid, missing_list = []):
"""
Mark a build target as failed (unbuildable)
Trigger removal of any files that have this as a dependency
"""
if not missing_list:
missing_list = [self.build_names_index[targetid]]
else:
missing_list = [self.build_names_index[targetid]] + missing_list
bb.msg.note(2, bb.msg.domain.Provider, "Target '%s' is unbuildable, removing...\nMissing or unbuildable dependency chain was: %s" % (self.build_names_index[targetid], missing_list))
self.failed_deps.append(targetid)
dependees = self.get_dependees(targetid)
for fnid in dependees:
self.fail_fnid(fnid, missing_list)
for taskid in range(len(self.tasks_idepends)):
idepends = self.tasks_idepends[taskid]
for (idependid, idependtask) in idepends:
if idependid == targetid:
self.fail_fnid(self.tasks_fnid[taskid], missing_list)
if self.abort and targetid in self.external_targets:
bb.msg.error(bb.msg.domain.Provider, "Required build target '%s' has no buildable providers.\nMissing or unbuildable dependency chain was: %s" % (self.build_names_index[targetid], missing_list))
raise bb.providers.NoProvider
def remove_runtarget(self, targetid, missing_list = []):
"""
Mark a run target as failed (unbuildable)
Trigger removal of any files that have this as a dependency
"""
if not missing_list:
missing_list = [self.run_names_index[targetid]]
else:
missing_list = [self.run_names_index[targetid]] + missing_list
bb.msg.note(1, bb.msg.domain.Provider, "Runtime target '%s' is unbuildable, removing...\nMissing or unbuildable dependency chain was: %s" % (self.run_names_index[targetid], missing_list))
self.failed_rdeps.append(targetid)
dependees = self.get_rdependees(targetid)
for fnid in dependees:
self.fail_fnid(fnid, missing_list)
def add_unresolved(self, cfgData, dataCache):
"""
Resolve all unresolved build and runtime targets
"""
bb.msg.note(1, bb.msg.domain.TaskData, "Resolving any missing task queue dependencies")
while 1:
added = 0
for target in self.get_unresolved_build_targets(dataCache):
try:
self.add_provider_internal(cfgData, dataCache, target)
added = added + 1
except bb.providers.NoProvider:
targetid = self.getbuild_id(target)
if self.abort and targetid in self.external_targets:
if self.get_rdependees_str(target):
bb.msg.error(bb.msg.domain.Provider, "Nothing PROVIDES '%s' (but '%s' DEPENDS on or otherwise requires it)" % (target, self.get_dependees_str(target)))
else:
bb.msg.error(bb.msg.domain.Provider, "Nothing PROVIDES '%s'" % (target))
raise
self.remove_buildtarget(targetid)
for target in self.get_unresolved_run_targets(dataCache):
try:
self.add_rprovider(cfgData, dataCache, target)
added = added + 1
except bb.providers.NoRProvider:
self.remove_runtarget(self.getrun_id(target))
bb.msg.debug(1, bb.msg.domain.TaskData, "Resolved " + str(added) + " extra dependecies")
if added == 0:
break
# self.dump_data()
def dump_data(self):
"""
Dump some debug information on the internal data structures
"""
bb.msg.debug(3, bb.msg.domain.TaskData, "build_names:")
bb.msg.debug(3, bb.msg.domain.TaskData, ", ".join(self.build_names_index))
bb.msg.debug(3, bb.msg.domain.TaskData, "run_names:")
bb.msg.debug(3, bb.msg.domain.TaskData, ", ".join(self.run_names_index))
bb.msg.debug(3, bb.msg.domain.TaskData, "build_targets:")
for buildid in range(len(self.build_names_index)):
target = self.build_names_index[buildid]
targets = "None"
if buildid in self.build_targets:
targets = self.build_targets[buildid]
bb.msg.debug(3, bb.msg.domain.TaskData, " (%s)%s: %s" % (buildid, target, targets))
bb.msg.debug(3, bb.msg.domain.TaskData, "run_targets:")
for runid in range(len(self.run_names_index)):
target = self.run_names_index[runid]
targets = "None"
if runid in self.run_targets:
targets = self.run_targets[runid]
bb.msg.debug(3, bb.msg.domain.TaskData, " (%s)%s: %s" % (runid, target, targets))
bb.msg.debug(3, bb.msg.domain.TaskData, "tasks:")
for task in range(len(self.tasks_name)):
bb.msg.debug(3, bb.msg.domain.TaskData, " (%s)%s - %s: %s" % (
task,
self.fn_index[self.tasks_fnid[task]],
self.tasks_name[task],
self.tasks_tdepends[task]))
bb.msg.debug(3, bb.msg.domain.TaskData, "dependency ids (per fn):")
for fnid in self.depids:
bb.msg.debug(3, bb.msg.domain.TaskData, " %s %s: %s" % (fnid, self.fn_index[fnid], self.depids[fnid]))
bb.msg.debug(3, bb.msg.domain.TaskData, "runtime dependency ids (per fn):")
for fnid in self.rdepids:
bb.msg.debug(3, bb.msg.domain.TaskData, " %s %s: %s" % (fnid, self.fn_index[fnid], self.rdepids[fnid]))

View File

@@ -0,0 +1,18 @@
#
# BitBake UI Implementation
#
# Copyright (C) 2006-2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.

View File

@@ -0,0 +1,18 @@
#
# BitBake UI Implementation
#
# Copyright (C) 2006-2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.

View File

@@ -28,7 +28,7 @@ import time
class BuildConfiguration:
""" Represents a potential *or* historic *or* concrete build. It
encompasses all the things that we need to tell bitbake to do to make it
build what we want it to build.
build what we want it to build.
It also stored the metadata URL and the set of possible machines (and the
distros / images / uris for these. Apart from the metdata URL these are
@@ -73,33 +73,34 @@ class BuildConfiguration:
return self.urls
# It might be a lot lot better if we stored these in like, bitbake conf
# file format.
@staticmethod
# file format.
@staticmethod
def load_from_file (filename):
f = open (filename, "r")
conf = BuildConfiguration()
with open(filename, "r") as f:
for line in f:
data = line.split (";")[1]
if (line.startswith ("metadata-url;")):
conf.metadata_url = data.strip()
continue
if (line.startswith ("url;")):
conf.urls += [data.strip()]
continue
if (line.startswith ("extra-url;")):
conf.extra_urls += [data.strip()]
continue
if (line.startswith ("machine;")):
conf.machine = data.strip()
continue
if (line.startswith ("distribution;")):
conf.distro = data.strip()
continue
if (line.startswith ("image;")):
conf.image = data.strip()
continue
for line in f.readlines():
data = line.split (";")[1]
if (line.startswith ("metadata-url;")):
conf.metadata_url = data.strip()
continue
if (line.startswith ("url;")):
conf.urls += [data.strip()]
continue
if (line.startswith ("extra-url;")):
conf.extra_urls += [data.strip()]
continue
if (line.startswith ("machine;")):
conf.machine = data.strip()
continue
if (line.startswith ("distribution;")):
conf.distro = data.strip()
continue
if (line.startswith ("image;")):
conf.image = data.strip()
continue
f.close ()
return conf
# Serialise to a file. This is part of the build process and we use this
@@ -139,13 +140,13 @@ class BuildResult(gobject.GObject):
".conf" in the directory for the build.
This is GObject so that it can be included in the TreeStore."""
(STATE_COMPLETE, STATE_FAILED, STATE_ONGOING) = \
(0, 1, 2)
def __init__ (self, parent, identifier):
gobject.GObject.__init__ (self)
self.date = None
self.date = None
self.files = []
self.status = None
@@ -156,8 +157,8 @@ class BuildResult(gobject.GObject):
# format build-<year><month><day>-<ordinal> we can easily
# pull it out.
# TODO: Better to stat a file?
(_, date, revision) = identifier.split ("-")
print(date)
(_ , date, revision) = identifier.split ("-")
print date
year = int (date[0:4])
month = int (date[4:6])
@@ -180,7 +181,7 @@ class BuildResult(gobject.GObject):
self.add_file (file)
def add_file (self, file):
# Just add the file for now. Don't care about the type.
# Just add the file for now. Don't care about the type.
self.files += [(file, None)]
class BuildManagerModel (gtk.TreeStore):
@@ -193,7 +194,7 @@ class BuildManagerModel (gtk.TreeStore):
def __init__ (self):
gtk.TreeStore.__init__ (self,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
@@ -206,7 +207,7 @@ class BuildManager (gobject.GObject):
"results" directory but is also used for starting a new build."""
__gsignals__ = {
'population-finished' : (gobject.SIGNAL_RUN_LAST,
'population-finished' : (gobject.SIGNAL_RUN_LAST,
gobject.TYPE_NONE,
()),
'populate-error' : (gobject.SIGNAL_RUN_LAST,
@@ -219,13 +220,13 @@ class BuildManager (gobject.GObject):
date = long (time.mktime (result.date.timetuple()))
# Add a top level entry for the build
self.model.set (iter,
self.model.set (iter,
BuildManagerModel.COL_IDENT, result.identifier,
BuildManagerModel.COL_DESC, result.conf.image,
BuildManagerModel.COL_MACHINE, result.conf.machine,
BuildManagerModel.COL_DISTRO, result.conf.distro,
BuildManagerModel.COL_BUILD_RESULT, result,
BuildManagerModel.COL_MACHINE, result.conf.machine,
BuildManagerModel.COL_DISTRO, result.conf.distro,
BuildManagerModel.COL_BUILD_RESULT, result,
BuildManagerModel.COL_DATE, date,
BuildManagerModel.COL_STATE, result.state)
@@ -256,7 +257,7 @@ class BuildManager (gobject.GObject):
while (iter):
(ident, state) = self.model.get(iter,
BuildManagerModel.COL_IDENT,
BuildManagerModel.COL_IDENT,
BuildManagerModel.COL_STATE)
if state == BuildResult.STATE_ONGOING:
@@ -384,8 +385,8 @@ class BuildManager (gobject.GObject):
build_directory])
server.runCommand(["buildTargets", [conf.image], "rootfs"])
except Exception as e:
print(e)
except Exception, e:
print e
class BuildManagerTreeView (gtk.TreeView):
""" The tree view for the build manager. This shows the historic builds
@@ -421,29 +422,29 @@ class BuildManagerTreeView (gtk.TreeView):
# Misc descriptiony thing
renderer = gtk.CellRendererText ()
col = gtk.TreeViewColumn (None, renderer,
col = gtk.TreeViewColumn (None, renderer,
text=BuildManagerModel.COL_DESC)
self.append_column (col)
# Machine
renderer = gtk.CellRendererText ()
col = gtk.TreeViewColumn ("Machine", renderer,
col = gtk.TreeViewColumn ("Machine", renderer,
text=BuildManagerModel.COL_MACHINE)
self.append_column (col)
# distro
renderer = gtk.CellRendererText ()
col = gtk.TreeViewColumn ("Distribution", renderer,
col = gtk.TreeViewColumn ("Distribution", renderer,
text=BuildManagerModel.COL_DISTRO)
self.append_column (col)
# date (using a custom function for formatting the cell contents it
# takes epoch -> human readable string)
renderer = gtk.CellRendererText ()
col = gtk.TreeViewColumn ("Date", renderer,
col = gtk.TreeViewColumn ("Date", renderer,
text=BuildManagerModel.COL_DATE)
self.append_column (col)
col.set_cell_data_func (renderer,
col.set_cell_data_func (renderer,
self.date_format_custom_cell_data_func)
# For status.
@@ -453,3 +454,4 @@ class BuildManagerTreeView (gtk.TreeView):
self.append_column (col)
col.set_cell_data_func (renderer,
self.state_format_custom_cell_data_fun)

View File

@@ -0,0 +1,180 @@
#
# BitBake Graphical GTK User Interface
#
# Copyright (C) 2008 Intel Corporation
#
# Authored by Rob Bradford <rob@linux.intel.com>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import gtk
import gobject
class RunningBuildModel (gtk.TreeStore):
(COL_TYPE, COL_PACKAGE, COL_TASK, COL_MESSAGE, COL_ICON, COL_ACTIVE) = (0, 1, 2, 3, 4, 5)
def __init__ (self):
gtk.TreeStore.__init__ (self,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_STRING,
gobject.TYPE_BOOLEAN)
class RunningBuild (gobject.GObject):
__gsignals__ = {
'build-succeeded' : (gobject.SIGNAL_RUN_LAST,
gobject.TYPE_NONE,
()),
'build-failed' : (gobject.SIGNAL_RUN_LAST,
gobject.TYPE_NONE,
())
}
pids_to_task = {}
tasks_to_iter = {}
def __init__ (self):
gobject.GObject.__init__ (self)
self.model = RunningBuildModel()
def handle_event (self, event):
# Handle an event from the event queue, this may result in updating
# the model and thus the UI. Or it may be to tell us that the build
# has finished successfully (or not, as the case may be.)
parent = None
pid = 0
package = None
task = None
# If we have a pid attached to this message/event try and get the
# (package, task) pair for it. If we get that then get the parent iter
# for the message.
if event[1].has_key ('pid'):
pid = event[1]['pid']
if self.pids_to_task.has_key(pid):
(package, task) = self.pids_to_task[pid]
parent = self.tasks_to_iter[(package, task)]
if event[0].startswith('bb.msg.Msg'):
# Set a pretty icon for the message based on it's type.
if (event[0].startswith ('bb.msg.MsgWarn')):
icon = "dialog-warning"
elif (event[0].startswith ('bb.msg.MsgErr')):
icon = "dialog-error"
else:
icon = None
# Ignore the "Running task i of n .." messages
if (event[1]['_message'].startswith ("Running task")):
return
# Add the message to the tree either at the top level if parent is
# None otherwise as a descendent of a task.
self.model.append (parent,
(event[0].split()[-1], # e.g. MsgWarn, MsgError
package,
task,
event[1]['_message'],
icon,
False))
elif event[0].startswith('bb.build.TaskStarted'):
(package, task) = (event[1]['_package'], event[1]['_task'])
# Save out this PID.
self.pids_to_task[pid] = (package,task)
# Check if we already have this package in our model. If so then
# that can be the parent for the task. Otherwise we create a new
# top level for the package.
if (self.tasks_to_iter.has_key ((package, None))):
parent = self.tasks_to_iter[(package, None)]
else:
parent = self.model.append (None, (None,
package,
None,
"Package: %s" % (package),
None,
False))
self.tasks_to_iter[(package, None)] = parent
# Because this parent package now has an active child mark it as
# such.
self.model.set(parent, self.model.COL_ICON, "gtk-execute")
# Add an entry in the model for this task
i = self.model.append (parent, (None,
package,
task,
"Task: %s" % (task),
None,
False))
# Save out the iter so that we can find it when we have a message
# that we need to attach to a task.
self.tasks_to_iter[(package, task)] = i
# Mark this task as active.
self.model.set(i, self.model.COL_ICON, "gtk-execute")
elif event[0].startswith('bb.build.Task'):
if event[0].startswith('bb.build.TaskFailed'):
# Mark the task as failed
i = self.tasks_to_iter[(package, task)]
self.model.set(i, self.model.COL_ICON, "dialog-error")
# Mark the parent package as failed
i = self.tasks_to_iter[(package, None)]
self.model.set(i, self.model.COL_ICON, "dialog-error")
else:
# Mark the task as inactive
i = self.tasks_to_iter[(package, task)]
self.model.set(i, self.model.COL_ICON, None)
# Mark the parent package as inactive
i = self.tasks_to_iter[(package, None)]
self.model.set(i, self.model.COL_ICON, None)
# Clear the iters and the pids since when the task goes away the
# pid will no longer be used for messages
del self.tasks_to_iter[(package, task)]
del self.pids_to_task[pid]
elif event[0].startswith('bb.event.BuildCompleted'):
failures = int (event[1]['_failures'])
# Emit the appropriate signal depending on the number of failures
if (failures > 1):
self.emit ("build-failed")
else:
self.emit ("build-succeeded")
class RunningBuildTreeView (gtk.TreeView):
def __init__ (self):
gtk.TreeView.__init__ (self)
# The icon that indicates whether we're building or failed.
renderer = gtk.CellRendererPixbuf ()
col = gtk.TreeViewColumn ("Status", renderer)
col.add_attribute (renderer, "icon-name", 4)
self.append_column (col)
# The message of the build.
renderer = gtk.CellRendererText ()
col = gtk.TreeViewColumn ("Message", renderer, text=3)
self.append_column (col)

View File

@@ -0,0 +1,272 @@
#
# BitBake Graphical GTK based Dependency Explorer
#
# Copyright (C) 2007 Ross Burton
# Copyright (C) 2007 - 2008 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import gobject
import gtk
import threading
import xmlrpclib
# Package Model
(COL_PKG_NAME) = (0)
# Dependency Model
(TYPE_DEP, TYPE_RDEP) = (0, 1)
(COL_DEP_TYPE, COL_DEP_PARENT, COL_DEP_PACKAGE) = (0, 1, 2)
class PackageDepView(gtk.TreeView):
def __init__(self, model, dep_type, label):
gtk.TreeView.__init__(self)
self.current = None
self.dep_type = dep_type
self.filter_model = model.filter_new()
self.filter_model.set_visible_func(self._filter)
self.set_model(self.filter_model)
#self.connect("row-activated", self.on_package_activated, COL_DEP_PACKAGE)
self.append_column(gtk.TreeViewColumn(label, gtk.CellRendererText(), text=COL_DEP_PACKAGE))
def _filter(self, model, iter):
(this_type, package) = model.get(iter, COL_DEP_TYPE, COL_DEP_PARENT)
if this_type != self.dep_type: return False
return package == self.current
def set_current_package(self, package):
self.current = package
self.filter_model.refilter()
class PackageReverseDepView(gtk.TreeView):
def __init__(self, model, label):
gtk.TreeView.__init__(self)
self.current = None
self.filter_model = model.filter_new()
self.filter_model.set_visible_func(self._filter)
self.set_model(self.filter_model)
self.append_column(gtk.TreeViewColumn(label, gtk.CellRendererText(), text=COL_DEP_PARENT))
def _filter(self, model, iter):
package = model.get_value(iter, COL_DEP_PACKAGE)
return package == self.current
def set_current_package(self, package):
self.current = package
self.filter_model.refilter()
class DepExplorer(gtk.Window):
def __init__(self):
gtk.Window.__init__(self)
self.set_title("Dependency Explorer")
self.set_default_size(500, 500)
self.connect("delete-event", gtk.main_quit)
# Create the data models
self.pkg_model = gtk.ListStore(gobject.TYPE_STRING)
self.depends_model = gtk.ListStore(gobject.TYPE_INT, gobject.TYPE_STRING, gobject.TYPE_STRING)
pane = gtk.HPaned()
pane.set_position(250)
self.add(pane)
# The master list of packages
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.set_shadow_type(gtk.SHADOW_IN)
self.pkg_treeview = gtk.TreeView(self.pkg_model)
self.pkg_treeview.get_selection().connect("changed", self.on_cursor_changed)
self.pkg_treeview.append_column(gtk.TreeViewColumn("Package", gtk.CellRendererText(), text=COL_PKG_NAME))
pane.add1(scrolled)
scrolled.add(self.pkg_treeview)
box = gtk.VBox(homogeneous=True, spacing=4)
# Runtime Depends
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.set_shadow_type(gtk.SHADOW_IN)
self.rdep_treeview = PackageDepView(self.depends_model, TYPE_RDEP, "Runtime Depends")
self.rdep_treeview.connect("row-activated", self.on_package_activated, COL_DEP_PACKAGE)
scrolled.add(self.rdep_treeview)
box.add(scrolled)
# Build Depends
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.set_shadow_type(gtk.SHADOW_IN)
self.dep_treeview = PackageDepView(self.depends_model, TYPE_DEP, "Build Depends")
self.dep_treeview.connect("row-activated", self.on_package_activated, COL_DEP_PACKAGE)
scrolled.add(self.dep_treeview)
box.add(scrolled)
pane.add2(box)
# Reverse Depends
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.set_shadow_type(gtk.SHADOW_IN)
self.revdep_treeview = PackageReverseDepView(self.depends_model, "Reverse Depends")
self.revdep_treeview.connect("row-activated", self.on_package_activated, COL_DEP_PARENT)
scrolled.add(self.revdep_treeview)
box.add(scrolled)
pane.add2(box)
self.show_all()
def on_package_activated(self, treeview, path, column, data_col):
model = treeview.get_model()
package = model.get_value(model.get_iter(path), data_col)
pkg_path = []
def finder(model, path, iter, needle):
package = model.get_value(iter, COL_PKG_NAME)
if package == needle:
pkg_path.append(path)
return True
else:
return False
self.pkg_model.foreach(finder, package)
if pkg_path:
self.pkg_treeview.get_selection().select_path(pkg_path[0])
self.pkg_treeview.scroll_to_cell(pkg_path[0])
def on_cursor_changed(self, selection):
(model, it) = selection.get_selected()
if iter is None:
current_package = None
else:
current_package = model.get_value(it, COL_PKG_NAME)
self.rdep_treeview.set_current_package(current_package)
self.dep_treeview.set_current_package(current_package)
self.revdep_treeview.set_current_package(current_package)
def parse(depgraph, pkg_model, depends_model):
for package in depgraph["pn"]:
pkg_model.set(pkg_model.append(), COL_PKG_NAME, package)
for package in depgraph["depends"]:
for depend in depgraph["depends"][package]:
depends_model.set (depends_model.append(),
COL_DEP_TYPE, TYPE_DEP,
COL_DEP_PARENT, package,
COL_DEP_PACKAGE, depend)
for package in depgraph["rdepends-pn"]:
for rdepend in depgraph["rdepends-pn"][package]:
depends_model.set (depends_model.append(),
COL_DEP_TYPE, TYPE_RDEP,
COL_DEP_PARENT, package,
COL_DEP_PACKAGE, rdepend)
class ProgressBar(gtk.Window):
def __init__(self):
gtk.Window.__init__(self)
self.set_title("Parsing .bb files, please wait...")
self.set_default_size(500, 0)
self.connect("delete-event", gtk.main_quit)
self.progress = gtk.ProgressBar()
self.add(self.progress)
self.show_all()
class gtkthread(threading.Thread):
quit = threading.Event()
def __init__(self, shutdown):
threading.Thread.__init__(self)
self.setDaemon(True)
self.shutdown = shutdown
def run(self):
gobject.threads_init()
gtk.gdk.threads_init()
gtk.main()
gtkthread.quit.set()
def init(server, eventHandler):
try:
cmdline = server.runCommand(["getCmdLineAction"])
if not cmdline or cmdline[0] != "generateDotGraph":
print "This UI is only compatible with the -g option"
return
ret = server.runCommand(["generateDepTreeEvent", cmdline[1], cmdline[2]])
if ret != True:
print "Couldn't run command! %s" % ret
return
except xmlrpclib.Fault, x:
print "XMLRPC Fault getting commandline:\n %s" % x
return
shutdown = 0
gtkgui = gtkthread(shutdown)
gtkgui.start()
gtk.gdk.threads_enter()
pbar = ProgressBar()
dep = DepExplorer()
gtk.gdk.threads_leave()
while True:
try:
event = eventHandler.waitEvent(0.25)
if gtkthread.quit.isSet():
break
if event is None:
continue
if event[0].startswith('bb.event.ParseProgress'):
x = event[1]['sofar']
y = event[1]['total']
if x == y:
print("\nParsing finished. %d cached, %d parsed, %d skipped, %d masked, %d errors."
% ( event[1]['cached'], event[1]['parsed'], event[1]['skipped'], event[1]['masked'], event[1]['errors']))
pbar.hide()
gtk.gdk.threads_enter()
pbar.progress.set_fraction(float(x)/float(y))
pbar.progress.set_text("%d/%d (%2d %%)" % (x, y, x*100/y))
gtk.gdk.threads_leave()
continue
if event[0] == "bb.event.DepTreeGenerated":
gtk.gdk.threads_enter()
parse(event[1]['_depgraph'], dep.pkg_model, dep.depends_model)
gtk.gdk.threads_leave()
if event[0] == 'bb.command.CookerCommandCompleted':
continue
if event[0] == 'bb.command.CookerCommandFailed':
print "Command execution failed: %s" % event[1]['error']
break
if event[0] == 'bb.cooker.CookerExit':
break
continue
except KeyboardInterrupt:
if shutdown == 2:
print "\nThird Keyboard Interrupt, exit.\n"
break
if shutdown == 1:
print "\nSecond Keyboard Interrupt, stopping...\n"
server.runCommand(["stateStop"])
if shutdown == 0:
print "\nKeyboard Interrupt, closing down...\n"
server.runCommand(["stateShutdown"])
shutdown = shutdown + 1
pass

View File

@@ -0,0 +1,77 @@
#
# BitBake Graphical GTK User Interface
#
# Copyright (C) 2008 Intel Corporation
#
# Authored by Rob Bradford <rob@linux.intel.com>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import gobject
import gtk
import xmlrpclib
from bb.ui.crumbs.runningbuild import RunningBuildTreeView, RunningBuild
def event_handle_idle_func (eventHandler, build):
# Consume as many messages as we can in the time available to us
event = eventHandler.getEvent()
while event:
build.handle_event (event)
event = eventHandler.getEvent()
return True
class MainWindow (gtk.Window):
def __init__ (self):
gtk.Window.__init__ (self, gtk.WINDOW_TOPLEVEL)
# Setup tree view and the scrolled window
scrolled_window = gtk.ScrolledWindow ()
self.add (scrolled_window)
self.cur_build_tv = RunningBuildTreeView()
scrolled_window.add (self.cur_build_tv)
def init (server, eventHandler):
gobject.threads_init()
gtk.gdk.threads_init()
window = MainWindow ()
window.show_all ()
# Create the object for the current build
running_build = RunningBuild ()
window.cur_build_tv.set_model (running_build.model)
try:
cmdline = server.runCommand(["getCmdLineAction"])
print cmdline
if not cmdline:
return 1
ret = server.runCommand(cmdline)
if ret != True:
print "Couldn't get default commandline! %s" % ret
return 1
except xmlrpclib.Fault, x:
print "XMLRPC Fault getting commandline:\n %s" % x
return 1
# Use a timeout function for probing the event queue to find out if we
# have a message waiting for us.
gobject.timeout_add (200,
event_handle_idle_func,
eventHandler,
running_build)
gtk.main()

View File

@@ -0,0 +1,162 @@
#
# BitBake (No)TTY UI Implementation
#
# Handling output to TTYs or files (no TTY)
#
# Copyright (C) 2006-2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys
import itertools
import xmlrpclib
parsespin = itertools.cycle( r'|/-\\' )
def init(server, eventHandler):
# Get values of variables which control our output
includelogs = server.runCommand(["getVariable", "BBINCLUDELOGS"])
loglines = server.runCommand(["getVariable", "BBINCLUDELOGS_LINES"])
try:
cmdline = server.runCommand(["getCmdLineAction"])
#print cmdline
if not cmdline:
return 1
ret = server.runCommand(cmdline)
if ret != True:
print "Couldn't get default commandline! %s" % ret
return 1
except xmlrpclib.Fault, x:
print "XMLRPC Fault getting commandline:\n %s" % x
return 1
shutdown = 0
return_value = 0
while True:
try:
event = eventHandler.waitEvent(0.25)
if event is None:
continue
#print event
if event[0].startswith('bb.msg.MsgPlain'):
print event[1]['_message']
continue
if event[0].startswith('bb.msg.MsgDebug'):
print 'DEBUG: ' + event[1]['_message']
continue
if event[0].startswith('bb.msg.MsgNote'):
print 'NOTE: ' + event[1]['_message']
continue
if event[0].startswith('bb.msg.MsgWarn'):
print 'WARNING: ' + event[1]['_message']
continue
if event[0].startswith('bb.msg.MsgError'):
return_value = 1
print 'ERROR: ' + event[1]['_message']
continue
if event[0].startswith('bb.msg.MsgFatal'):
return_value = 1
print 'FATAL: ' + event[1]['_message']
break
if event[0].startswith('bb.build.TaskFailed'):
return_value = 1
logfile = event[1]['logfile']
if logfile:
print "ERROR: Logfile of failure stored in %s." % logfile
if 1 or includelogs:
print "Log data follows:"
f = open(logfile, "r")
lines = []
while True:
l = f.readline()
if l == '':
break
l = l.rstrip()
if loglines:
lines.append(' | %s' % l)
if len(lines) > int(loglines):
lines.pop(0)
else:
print '| %s' % l
f.close()
if lines:
for line in lines:
print line
if event[0].startswith('bb.build.Task'):
print "NOTE: %s" % event[1]['_message']
continue
if event[0].startswith('bb.event.ParseProgress'):
x = event[1]['sofar']
y = event[1]['total']
if os.isatty(sys.stdout.fileno()):
sys.stdout.write("\rNOTE: Handling BitBake files: %s (%04d/%04d) [%2d %%]" % ( parsespin.next(), x, y, x*100/y ) )
sys.stdout.flush()
else:
if x == 1:
sys.stdout.write("Parsing .bb files, please wait...")
sys.stdout.flush()
if x == y:
sys.stdout.write("done.")
sys.stdout.flush()
if x == y:
print("\nParsing finished. %d cached, %d parsed, %d skipped, %d masked, %d errors."
% ( event[1]['cached'], event[1]['parsed'], event[1]['skipped'], event[1]['masked'], event[1]['errors']))
continue
if event[0] == 'bb.command.CookerCommandCompleted':
break
if event[0] == 'bb.command.CookerCommandSetExitCode':
return_value = event[1]['exitcode']
continue
if event[0] == 'bb.command.CookerCommandFailed':
return_value = 1
print "Command execution failed: %s" % event[1]['error']
break
if event[0] == 'bb.cooker.CookerExit':
break
# ignore
if event[0].startswith('bb.event.BuildStarted'):
continue
if event[0].startswith('bb.event.BuildCompleted'):
continue
if event[0].startswith('bb.event.MultipleProviders'):
continue
if event[0].startswith('bb.runqueue.runQueue'):
continue
if event[0].startswith('bb.event.StampUpdate'):
continue
if event[0].startswith('bb.event.ConfigParsed'):
continue
if event[0].startswith('bb.event.RecipeParsed'):
continue
print "Unknown Event: %s" % event
except KeyboardInterrupt:
if shutdown == 2:
print "\nThird Keyboard Interrupt, exit.\n"
break
if shutdown == 1:
print "\nSecond Keyboard Interrupt, stopping...\n"
server.runCommand(["stateStop"])
if shutdown == 0:
print "\nKeyboard Interrupt, closing down...\n"
server.runCommand(["stateShutdown"])
shutdown = shutdown + 1
pass
return return_value

View File

@@ -0,0 +1,335 @@
#
# BitBake Curses UI Implementation
#
# Implements an ncurses frontend for the BitBake utility.
#
# Copyright (C) 2006 Michael 'Mickey' Lauer
# Copyright (C) 2006-2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
"""
We have the following windows:
1.) Main Window: Shows what we are ultimately building and how far we are. Includes status bar
2.) Thread Activity Window: Shows one status line for every concurrent bitbake thread.
3.) Command Line Window: Contains an interactive command line where you can interact w/ Bitbake.
Basic window layout is like that:
|---------------------------------------------------------|
| <Main Window> | <Thread Activity Window> |
| | 0: foo do_compile complete|
| Building Gtk+-2.6.10 | 1: bar do_patch complete |
| Status: 60% | ... |
| | ... |
| | ... |
|---------------------------------------------------------|
|<Command Line Window> |
|>>> which virtual/kernel |
|openzaurus-kernel |
|>>> _ |
|---------------------------------------------------------|
"""
import os, sys, curses, itertools, time
import bb
import xmlrpclib
from bb import ui
from bb.ui import uihelper
parsespin = itertools.cycle( r'|/-\\' )
X = 0
Y = 1
WIDTH = 2
HEIGHT = 3
MAXSTATUSLENGTH = 32
class NCursesUI:
"""
NCurses UI Class
"""
class Window:
"""Base Window Class"""
def __init__( self, x, y, width, height, fg=curses.COLOR_BLACK, bg=curses.COLOR_WHITE ):
self.win = curses.newwin( height, width, y, x )
self.dimensions = ( x, y, width, height )
"""
if curses.has_colors():
color = 1
curses.init_pair( color, fg, bg )
self.win.bkgdset( ord(' '), curses.color_pair(color) )
else:
self.win.bkgdset( ord(' '), curses.A_BOLD )
"""
self.erase()
self.setScrolling()
self.win.noutrefresh()
def erase( self ):
self.win.erase()
def setScrolling( self, b = True ):
self.win.scrollok( b )
self.win.idlok( b )
def setBoxed( self ):
self.boxed = True
self.win.box()
self.win.noutrefresh()
def setText( self, x, y, text, *args ):
self.win.addstr( y, x, text, *args )
self.win.noutrefresh()
def appendText( self, text, *args ):
self.win.addstr( text, *args )
self.win.noutrefresh()
def drawHline( self, y ):
self.win.hline( y, 0, curses.ACS_HLINE, self.dimensions[WIDTH] )
self.win.noutrefresh()
class DecoratedWindow( Window ):
"""Base class for windows with a box and a title bar"""
def __init__( self, title, x, y, width, height, fg=curses.COLOR_BLACK, bg=curses.COLOR_WHITE ):
NCursesUI.Window.__init__( self, x+1, y+3, width-2, height-4, fg, bg )
self.decoration = NCursesUI.Window( x, y, width, height, fg, bg )
self.decoration.setBoxed()
self.decoration.win.hline( 2, 1, curses.ACS_HLINE, width-2 )
self.setTitle( title )
def setTitle( self, title ):
self.decoration.setText( 1, 1, title.center( self.dimensions[WIDTH]-2 ), curses.A_BOLD )
#-------------------------------------------------------------------------#
# class TitleWindow( Window ):
#-------------------------------------------------------------------------#
# """Title Window"""
# def __init__( self, x, y, width, height ):
# NCursesUI.Window.__init__( self, x, y, width, height )
# version = bb.__version__
# title = "BitBake %s" % version
# credit = "(C) 2003-2007 Team BitBake"
# #self.win.hline( 2, 1, curses.ACS_HLINE, width-2 )
# self.win.border()
# self.setText( 1, 1, title.center( self.dimensions[WIDTH]-2 ), curses.A_BOLD )
# self.setText( 1, 2, credit.center( self.dimensions[WIDTH]-2 ), curses.A_BOLD )
#-------------------------------------------------------------------------#
class ThreadActivityWindow( DecoratedWindow ):
#-------------------------------------------------------------------------#
"""Thread Activity Window"""
def __init__( self, x, y, width, height ):
NCursesUI.DecoratedWindow.__init__( self, "Thread Activity", x, y, width, height )
def setStatus( self, thread, text ):
line = "%02d: %s" % ( thread, text )
width = self.dimensions[WIDTH]
if ( len(line) > width ):
line = line[:width-3] + "..."
else:
line = line.ljust( width )
self.setText( 0, thread, line )
#-------------------------------------------------------------------------#
class MainWindow( DecoratedWindow ):
#-------------------------------------------------------------------------#
"""Main Window"""
def __init__( self, x, y, width, height ):
self.StatusPosition = width - MAXSTATUSLENGTH
NCursesUI.DecoratedWindow.__init__( self, None, x, y, width, height )
curses.nl()
def setTitle( self, title ):
title = "BitBake %s" % bb.__version__
self.decoration.setText( 2, 1, title, curses.A_BOLD )
self.decoration.setText( self.StatusPosition - 8, 1, "Status:", curses.A_BOLD )
def setStatus(self, status):
while len(status) < MAXSTATUSLENGTH:
status = status + " "
self.decoration.setText( self.StatusPosition, 1, status, curses.A_BOLD )
#-------------------------------------------------------------------------#
class ShellOutputWindow( DecoratedWindow ):
#-------------------------------------------------------------------------#
"""Interactive Command Line Output"""
def __init__( self, x, y, width, height ):
NCursesUI.DecoratedWindow.__init__( self, "Command Line Window", x, y, width, height )
#-------------------------------------------------------------------------#
class ShellInputWindow( Window ):
#-------------------------------------------------------------------------#
"""Interactive Command Line Input"""
def __init__( self, x, y, width, height ):
NCursesUI.Window.__init__( self, x, y, width, height )
# put that to the top again from curses.textpad import Textbox
# self.textbox = Textbox( self.win )
# t = threading.Thread()
# t.run = self.textbox.edit
# t.start()
#-------------------------------------------------------------------------#
def main(self, stdscr, server, eventHandler):
#-------------------------------------------------------------------------#
height, width = stdscr.getmaxyx()
# for now split it like that:
# MAIN_y + THREAD_y = 2/3 screen at the top
# MAIN_x = 2/3 left, THREAD_y = 1/3 right
# CLI_y = 1/3 of screen at the bottom
# CLI_x = full
main_left = 0
main_top = 0
main_height = ( height / 3 * 2 )
main_width = ( width / 3 ) * 2
clo_left = main_left
clo_top = main_top + main_height
clo_height = height - main_height - main_top - 1
clo_width = width
cli_left = main_left
cli_top = clo_top + clo_height
cli_height = 1
cli_width = width
thread_left = main_left + main_width
thread_top = main_top
thread_height = main_height
thread_width = width - main_width
#tw = self.TitleWindow( 0, 0, width, main_top )
mw = self.MainWindow( main_left, main_top, main_width, main_height )
taw = self.ThreadActivityWindow( thread_left, thread_top, thread_width, thread_height )
clo = self.ShellOutputWindow( clo_left, clo_top, clo_width, clo_height )
cli = self.ShellInputWindow( cli_left, cli_top, cli_width, cli_height )
cli.setText( 0, 0, "BB>" )
mw.setStatus("Idle")
helper = uihelper.BBUIHelper()
shutdown = 0
try:
cmdline = server.runCommand(["getCmdLineAction"])
if not cmdline:
return
ret = server.runCommand(cmdline)
if ret != True:
print "Couldn't get default commandlind! %s" % ret
return
except xmlrpclib.Fault, x:
print "XMLRPC Fault getting commandline:\n %s" % x
return
exitflag = False
while not exitflag:
try:
event = eventHandler.waitEvent(0.25)
if not event:
continue
helper.eventHandler(event)
#mw.appendText("%s\n" % event[0])
if event[0].startswith('bb.build.Task'):
mw.appendText("NOTE: %s\n" % event[1]['_message'])
if event[0].startswith('bb.msg.MsgDebug'):
mw.appendText('DEBUG: ' + event[1]['_message'] + '\n')
if event[0].startswith('bb.msg.MsgNote'):
mw.appendText('NOTE: ' + event[1]['_message'] + '\n')
if event[0].startswith('bb.msg.MsgWarn'):
mw.appendText('WARNING: ' + event[1]['_message'] + '\n')
if event[0].startswith('bb.msg.MsgError'):
mw.appendText('ERROR: ' + event[1]['_message'] + '\n')
if event[0].startswith('bb.msg.MsgFatal'):
mw.appendText('FATAL: ' + event[1]['_message'] + '\n')
if event[0].startswith('bb.event.ParseProgress'):
x = event[1]['sofar']
y = event[1]['total']
if x == y:
mw.setStatus("Idle")
mw.appendText("Parsing finished. %d cached, %d parsed, %d skipped, %d masked."
% ( event[1]['cached'], event[1]['parsed'], event[1]['skipped'], event[1]['masked'] ))
else:
mw.setStatus("Parsing: %s (%04d/%04d) [%2d %%]" % ( parsespin.next(), x, y, x*100/y ) )
# if event[0].startswith('bb.build.TaskFailed'):
# if event[1]['logfile']:
# if data.getVar("BBINCLUDELOGS", d):
# bb.msg.error(bb.msg.domain.Build, "log data follows (%s)" % logfile)
# number_of_lines = data.getVar("BBINCLUDELOGS_LINES", d)
# if number_of_lines:
# os.system('tail -n%s %s' % (number_of_lines, logfile))
# else:
# f = open(logfile, "r")
# while True:
# l = f.readline()
# if l == '':
# break
# l = l.rstrip()
# print '| %s' % l
# f.close()
# else:
# bb.msg.error(bb.msg.domain.Build, "see log in %s" % logfile)
if event[0] == 'bb.command.CookerCommandCompleted':
exitflag = True
if event[0] == 'bb.command.CookerCommandFailed':
mw.appendText("Command execution failed: %s" % event[1]['error'])
time.sleep(2)
exitflag = True
if event[0] == 'bb.cooker.CookerExit':
exitflag = True
if helper.needUpdate:
activetasks, failedtasks = helper.getTasks()
taw.erase()
taw.setText(0, 0, "")
if activetasks:
taw.appendText("Active Tasks:\n")
for task in activetasks:
taw.appendText(task)
if failedtasks:
taw.appendText("Failed Tasks:\n")
for task in failedtasks:
taw.appendText(task)
curses.doupdate()
except KeyboardInterrupt:
if shutdown == 2:
mw.appendText("Third Keyboard Interrupt, exit.\n")
exitflag = True
if shutdown == 1:
mw.appendText("Second Keyboard Interrupt, stopping...\n")
server.runCommand(["stateStop"])
if shutdown == 0:
mw.appendText("Keyboard Interrupt, closing down...\n")
server.runCommand(["stateShutdown"])
shutdown = shutdown + 1
pass
def init(server, eventHandler):
if not os.isatty(sys.stdout.fileno()):
print "FATAL: Unable to run 'ncurses' UI without a TTY."
return
ui = NCursesUI()
try:
curses.wrapper(ui.main, server, eventHandler)
except:
import traceback
traceback.print_exc()

View File

@@ -24,7 +24,6 @@ import gtk.glade
import threading
import urllib2
import os
import contextlib
from bb.ui.crumbs.buildmanager import BuildManager, BuildConfiguration
from bb.ui.crumbs.buildmanager import BuildManagerTreeView
@@ -39,7 +38,7 @@ class MetaDataLoader(gobject.GObject):
on what machines are available. The distribution and images available for
the machine and the the uris to use for building the given machine."""
__gsignals__ = {
'success' : (gobject.SIGNAL_RUN_LAST,
'success' : (gobject.SIGNAL_RUN_LAST,
gobject.TYPE_NONE,
()),
'error' : (gobject.SIGNAL_RUN_LAST,
@@ -78,19 +77,20 @@ class MetaDataLoader(gobject.GObject):
def run (self):
result = {}
try:
with contextlib.closing (urllib2.urlopen (self.url)) as f:
# Parse the metadata format. The format is....
# <machine>;<default distro>|<distro>...;<default image>|<image>...;<type##url>|...
for line in f:
components = line.split(";")
if (len (components) < 4):
raise MetaDataLoader.LoaderThread.LoaderImportException
machine = components[0]
distros = components[1].split("|")
images = components[2].split("|")
urls = components[3].split("|")
f = urllib2.urlopen (self.url)
result[machine] = (distros, images, urls)
# Parse the metadata format. The format is....
# <machine>;<default distro>|<distro>...;<default image>|<image>...;<type##url>|...
for line in f.readlines():
components = line.split(";")
if (len (components) < 4):
raise MetaDataLoader.LoaderThread.LoaderImportException
machine = components[0]
distros = components[1].split("|")
images = components[2].split("|")
urls = components[3].split("|")
result[machine] = (distros, images, urls)
# Create an object representing this *potential*
# configuration. It can become concrete if the machine, distro
@@ -104,13 +104,13 @@ class MetaDataLoader(gobject.GObject):
gobject.idle_add (MetaDataLoader.emit_success_signal,
self.loader)
except MetaDataLoader.LoaderThread.LoaderImportException as e:
except MetaDataLoader.LoaderThread.LoaderImportException, e:
gobject.idle_add (MetaDataLoader.emit_error_signal, self.loader,
"Repository metadata corrupt")
except Exception as e:
except Exception, e:
gobject.idle_add (MetaDataLoader.emit_error_signal, self.loader,
"Unable to download repository metadata")
print(e)
print e
def try_fetch_from_url (self, url):
# Try and download the metadata. Firing a signal if successful
@@ -211,7 +211,7 @@ class BuildSetupDialog (gtk.Dialog):
# Build
button = gtk.Button ("_Build", None, True)
image = gtk.Image ()
image.set_from_stock (gtk.STOCK_EXECUTE, gtk.ICON_SIZE_BUTTON)
image.set_from_stock (gtk.STOCK_EXECUTE,gtk.ICON_SIZE_BUTTON)
button.set_image (image)
self.add_action_widget (button, BuildSetupDialog.RESPONSE_BUILD)
button.show_all ()
@@ -293,7 +293,7 @@ class BuildSetupDialog (gtk.Dialog):
if (active_iter):
self.configuration.machine = model.get(active_iter, 0)[0]
# Extract the chosen distro from the combo
# Extract the chosen distro from the combo
model = self.distribution_combo.get_model()
active_iter = self.distribution_combo.get_active_iter()
if (active_iter):
@@ -311,62 +311,62 @@ class BuildSetupDialog (gtk.Dialog):
#
# TODO: Should be a method on the RunningBuild class
def event_handle_timeout (eventHandler, build):
# Consume as many messages as we can ...
event = eventHandler.getEvent()
while event:
build.handle_event (event)
event = eventHandler.getEvent()
return True
# Consume as many messages as we can ...
event = eventHandler.getEvent()
while event:
build.handle_event (event)
event = eventHandler.getEvent()
return True
class MainWindow (gtk.Window):
# Callback that gets fired when the user hits a button in the
# BuildSetupDialog.
def build_dialog_box_response_cb (self, dialog, response_id):
conf = None
if (response_id == BuildSetupDialog.RESPONSE_BUILD):
dialog.update_configuration()
print(dialog.configuration.machine, dialog.configuration.distro, \
dialog.configuration.image)
conf = dialog.configuration
# Callback that gets fired when the user hits a button in the
# BuildSetupDialog.
def build_dialog_box_response_cb (self, dialog, response_id):
conf = None
if (response_id == BuildSetupDialog.RESPONSE_BUILD):
dialog.update_configuration()
print dialog.configuration.machine, dialog.configuration.distro, \
dialog.configuration.image
conf = dialog.configuration
dialog.destroy()
dialog.destroy()
if conf:
self.manager.do_build (conf)
if conf:
self.manager.do_build (conf)
def build_button_clicked_cb (self, button):
dialog = BuildSetupDialog ()
def build_button_clicked_cb (self, button):
dialog = BuildSetupDialog ()
# For some unknown reason Dialog.run causes nice little deadlocks ... :-(
dialog.connect ("response", self.build_dialog_box_response_cb)
dialog.show()
# For some unknown reason Dialog.run causes nice little deadlocks ... :-(
dialog.connect ("response", self.build_dialog_box_response_cb)
dialog.show()
def __init__ (self):
gtk.Window.__init__ (self)
def __init__ (self):
gtk.Window.__init__ (self)
# Pull in *just* the main vbox from the Glade XML data and then pack
# that inside the window
gxml = gtk.glade.XML (os.path.dirname(__file__) + "/crumbs/puccho.glade",
root = "main_window_vbox")
vbox = gxml.get_widget ("main_window_vbox")
self.add (vbox)
# Pull in *just* the main vbox from the Glade XML data and then pack
# that inside the window
gxml = gtk.glade.XML (os.path.dirname(__file__) + "/crumbs/puccho.glade",
root = "main_window_vbox")
vbox = gxml.get_widget ("main_window_vbox")
self.add (vbox)
# Create the tree views for the build manager view and the progress view
self.build_manager_view = BuildManagerTreeView()
self.running_build_view = RunningBuildTreeView()
# Create the tree views for the build manager view and the progress view
self.build_manager_view = BuildManagerTreeView()
self.running_build_view = RunningBuildTreeView()
# Grab the scrolled windows that we put the tree views into
self.results_scrolledwindow = gxml.get_widget ("results_scrolledwindow")
self.progress_scrolledwindow = gxml.get_widget ("progress_scrolledwindow")
# Grab the scrolled windows that we put the tree views into
self.results_scrolledwindow = gxml.get_widget ("results_scrolledwindow")
self.progress_scrolledwindow = gxml.get_widget ("progress_scrolledwindow")
# Put the tree views inside ...
self.results_scrolledwindow.add (self.build_manager_view)
self.progress_scrolledwindow.add (self.running_build_view)
# Put the tree views inside ...
self.results_scrolledwindow.add (self.build_manager_view)
self.progress_scrolledwindow.add (self.running_build_view)
# Hook up the build button...
self.build_button = gxml.get_widget ("main_toolbutton_build")
self.build_button.connect ("clicked", self.build_button_clicked_cb)
# Hook up the build button...
self.build_button = gxml.get_widget ("main_toolbutton_build")
self.build_button.connect ("clicked", self.build_button_clicked_cb)
# I'm not very happy about the current ownership of the RunningBuild. I have
# my suspicions that this object should be held by the BuildManager since we
@@ -383,14 +383,14 @@ def running_build_succeeded_cb (running_build, manager):
# BuildManager. It can then hook onto the signals directly and drive
# interesting things it cares about.
manager.notify_build_succeeded ()
print("build succeeded")
print "build succeeded"
def running_build_failed_cb (running_build, manager):
# As above
print("build failed")
print "build failed"
manager.notify_build_failed ()
def main (server, eventHandler):
def init (server, eventHandler):
# Initialise threading...
gobject.threads_init()
gtk.gdk.threads_init()

View File

@@ -19,37 +19,22 @@
"""
Use this class to fork off a thread to recieve event callbacks from the bitbake
Use this class to fork off a thread to recieve event callbacks from the bitbake
server and queue them for the UI to process. This process must be used to avoid
client/server deadlocks.
"""
import socket, threading, pickle
import socket, threading
from SimpleXMLRPCServer import SimpleXMLRPCServer, SimpleXMLRPCRequestHandler
class BBUIEventQueue:
def __init__(self, BBServer, clientinfo=("localhost, 0")):
def __init__(self, BBServer):
self.eventQueue = []
self.eventQueueLock = threading.Lock()
self.eventQueueNotify = threading.Event()
self.BBServer = BBServer
self.clientinfo = clientinfo
server = UIXMLRPCServer(self.clientinfo)
self.host, self.port = server.socket.getsockname()
server.register_function( self.system_quit, "event.quit" )
server.register_function( self.send_event, "event.sendpickle" )
server.socket.settimeout(1)
self.EventHandle = self.BBServer.registerEventHandler(self.host, self.port)
if (self.EventHandle == None):
bb.fatal("Could not register UI event handler")
self.server = server
self.t = threading.Thread()
self.t.setDaemon(True)
@@ -70,6 +55,7 @@ class BBUIEventQueue:
self.eventQueueNotify.clear()
self.eventQueueLock.release()
return item
def waitEvent(self, delay):
@@ -77,20 +63,27 @@ class BBUIEventQueue:
return self.getEvent()
def queue_event(self, event):
self.eventQueueLock.acquire()
self.eventQueue.append(event)
self.eventQueueNotify.set()
self.eventQueueLock.release()
def send_event(self, event):
self.queue_event(pickle.loads(event))
def startCallbackHandler(self):
self.server.timeout = 1
while not self.server.quit:
self.server.handle_request()
self.server.server_close()
server = UIXMLRPCServer()
self.host, self.port = server.socket.getsockname()
server.register_function( self.system_quit, "event.quit" )
server.register_function( self.queue_event, "event.send" )
server.socket.settimeout(1)
self.EventHandle = self.BBServer.registerEventHandler(self.host, self.port)
self.server = server
while not server.quit:
server.handle_request()
server.server_close()
def system_quit( self ):
"""
@@ -104,7 +97,7 @@ class BBUIEventQueue:
class UIXMLRPCServer (SimpleXMLRPCServer):
def __init__( self, interface ):
def __init__( self, interface = ("localhost", 0) ):
self.quit = False
SimpleXMLRPCServer.__init__( self,
interface,
@@ -119,15 +112,16 @@ class UIXMLRPCServer (SimpleXMLRPCServer):
return (sock, addr)
except socket.timeout:
pass
return (None, None)
return (None,None)
def close_request(self, request):
if request is None:
return
SimpleXMLRPCServer.close_request(self, request)
def process_request(self, request, client_address):
if request is None:
return
SimpleXMLRPCServer.process_request(self, request, client_address)

View File

@@ -0,0 +1,49 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2006 - 2007 Michael 'Mickey' Lauer
# Copyright (C) 2006 - 2007 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
class BBUIHelper:
def __init__(self):
self.needUpdate = False
self.running_tasks = {}
self.failed_tasks = {}
def eventHandler(self, event):
if event[0].startswith('bb.build.TaskStarted'):
self.running_tasks["%s %s\n" % (event[1]['_package'], event[1]['_task'])] = ""
self.needUpdate = True
if event[0].startswith('bb.build.TaskSucceeded'):
del self.running_tasks["%s %s\n" % (event[1]['_package'], event[1]['_task'])]
self.needUpdate = True
if event[0].startswith('bb.build.TaskFailed'):
del self.running_tasks["%s %s\n" % (event[1]['_package'], event[1]['_task'])]
self.failed_tasks["%s %s\n" % (event[1]['_package'], event[1]['_task'])] = ""
self.needUpdate = True
# Add runqueue event handling
#if event[0].startswith('bb.runqueue.runQueueTaskCompleted'):
# a = 1
#if event[0].startswith('bb.runqueue.runQueueTaskStarted'):
# a = 1
#if event[0].startswith('bb.runqueue.runQueueTaskFailed'):
# a = 1
#if event[0].startswith('bb.runqueue.runQueueExitWait'):
# a = 1
def getTasks(self):
return (self.running_tasks, self.failed_tasks)

417
bitbake-dev/lib/bb/utils.py Normal file
View File

@@ -0,0 +1,417 @@
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
"""
BitBake Utility Functions
"""
# Copyright (C) 2004 Michael Lauer
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
digits = "0123456789"
ascii_letters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"
import re, fcntl, os
def explode_version(s):
r = []
alpha_regexp = re.compile('^([a-zA-Z]+)(.*)$')
numeric_regexp = re.compile('^(\d+)(.*)$')
while (s != ''):
if s[0] in digits:
m = numeric_regexp.match(s)
r.append(int(m.group(1)))
s = m.group(2)
continue
if s[0] in ascii_letters:
m = alpha_regexp.match(s)
r.append(m.group(1))
s = m.group(2)
continue
s = s[1:]
return r
def vercmp_part(a, b):
va = explode_version(a)
vb = explode_version(b)
while True:
if va == []:
ca = None
else:
ca = va.pop(0)
if vb == []:
cb = None
else:
cb = vb.pop(0)
if ca == None and cb == None:
return 0
if ca > cb:
return 1
if ca < cb:
return -1
def vercmp(ta, tb):
(ea, va, ra) = ta
(eb, vb, rb) = tb
r = int(ea)-int(eb)
if (r == 0):
r = vercmp_part(va, vb)
if (r == 0):
r = vercmp_part(ra, rb)
return r
def explode_deps(s):
"""
Take an RDEPENDS style string of format:
"DEPEND1 (optional version) DEPEND2 (optional version) ..."
and return a list of dependencies.
Version information is ignored.
"""
r = []
l = s.split()
flag = False
for i in l:
if i[0] == '(':
flag = True
#j = []
if not flag:
r.append(i)
#else:
# j.append(i)
if flag and i.endswith(')'):
flag = False
# Ignore version
#r[-1] += ' ' + ' '.join(j)
return r
def explode_dep_versions(s):
"""
Take an RDEPENDS style string of format:
"DEPEND1 (optional version) DEPEND2 (optional version) ..."
and return a dictonary of dependencies and versions.
"""
r = {}
l = s.split()
lastdep = None
lastver = ""
inversion = False
for i in l:
if i[0] == '(':
inversion = True
lastver = i[1:] or ""
#j = []
elif inversion and i.endswith(')'):
inversion = False
lastver = lastver + " " + (i[:-1] or "")
r[lastdep] = lastver
elif not inversion:
r[i] = None
lastdep = i
lastver = ""
elif inversion:
lastver = lastver + " " + i
return r
def _print_trace(body, line):
"""
Print the Environment of a Text Body
"""
import bb
# print the environment of the method
bb.msg.error(bb.msg.domain.Util, "Printing the environment of the function")
min_line = max(1,line-4)
max_line = min(line+4,len(body)-1)
for i in range(min_line,max_line+1):
bb.msg.error(bb.msg.domain.Util, "\t%.4d:%s" % (i, body[i-1]) )
def better_compile(text, file, realfile):
"""
A better compile method. This method
will print the offending lines.
"""
try:
return compile(text, file, "exec")
except Exception, e:
import bb,sys
# split the text into lines again
body = text.split('\n')
bb.msg.error(bb.msg.domain.Util, "Error in compiling python function in: ", realfile)
bb.msg.error(bb.msg.domain.Util, "The lines resulting into this error were:")
bb.msg.error(bb.msg.domain.Util, "\t%d:%s:'%s'" % (e.lineno, e.__class__.__name__, body[e.lineno-1]))
_print_trace(body, e.lineno)
# exit now
sys.exit(1)
def better_exec(code, context, text, realfile):
"""
Similiar to better_compile, better_exec will
print the lines that are responsible for the
error.
"""
import bb,sys
try:
exec code in context
except:
(t,value,tb) = sys.exc_info()
if t in [bb.parse.SkipPackage, bb.build.FuncFailed]:
raise
# print the Header of the Error Message
bb.msg.error(bb.msg.domain.Util, "Error in executing python function in: %s" % realfile)
bb.msg.error(bb.msg.domain.Util, "Exception:%s Message:%s" % (t,value) )
# let us find the line number now
while tb.tb_next:
tb = tb.tb_next
import traceback
line = traceback.tb_lineno(tb)
_print_trace( text.split('\n'), line )
raise
def Enum(*names):
"""
A simple class to give Enum support
"""
assert names, "Empty enums are not supported"
class EnumClass(object):
__slots__ = names
def __iter__(self): return iter(constants)
def __len__(self): return len(constants)
def __getitem__(self, i): return constants[i]
def __repr__(self): return 'Enum' + str(names)
def __str__(self): return 'enum ' + str(constants)
class EnumValue(object):
__slots__ = ('__value')
def __init__(self, value): self.__value = value
Value = property(lambda self: self.__value)
EnumType = property(lambda self: EnumType)
def __hash__(self): return hash(self.__value)
def __cmp__(self, other):
# C fans might want to remove the following assertion
# to make all enums comparable by ordinal value {;))
assert self.EnumType is other.EnumType, "Only values from the same enum are comparable"
return cmp(self.__value, other.__value)
def __invert__(self): return constants[maximum - self.__value]
def __nonzero__(self): return bool(self.__value)
def __repr__(self): return str(names[self.__value])
maximum = len(names) - 1
constants = [None] * len(names)
for i, each in enumerate(names):
val = EnumValue(i)
setattr(EnumClass, each, val)
constants[i] = val
constants = tuple(constants)
EnumType = EnumClass()
return EnumType
def lockfile(name):
"""
Use the file fn as a lock file, return when the lock has been acquired.
Returns a variable to pass to unlockfile().
"""
path = os.path.dirname(name)
if not os.path.isdir(path):
import bb, sys
bb.msg.error(bb.msg.domain.Util, "Error, lockfile path does not exist!: %s" % path)
sys.exit(1)
while True:
# If we leave the lockfiles lying around there is no problem
# but we should clean up after ourselves. This gives potential
# for races though. To work around this, when we acquire the lock
# we check the file we locked was still the lock file on disk.
# by comparing inode numbers. If they don't match or the lockfile
# no longer exists, we start again.
# This implementation is unfair since the last person to request the
# lock is the most likely to win it.
try:
lf = open(name, "a+")
fcntl.flock(lf.fileno(), fcntl.LOCK_EX)
statinfo = os.fstat(lf.fileno())
if os.path.exists(lf.name):
statinfo2 = os.stat(lf.name)
if statinfo.st_ino == statinfo2.st_ino:
return lf
# File no longer exists or changed, retry
lf.close
except Exception, e:
continue
def unlockfile(lf):
"""
Unlock a file locked using lockfile()
"""
os.unlink(lf.name)
fcntl.flock(lf.fileno(), fcntl.LOCK_UN)
lf.close
def md5_file(filename):
"""
Return the hex string representation of the MD5 checksum of filename.
"""
try:
import hashlib
m = hashlib.md5()
except ImportError:
import md5
m = md5.new()
for line in open(filename):
m.update(line)
return m.hexdigest()
def sha256_file(filename):
"""
Return the hex string representation of the 256-bit SHA checksum of
filename. On Python 2.4 this will return None, so callers will need to
handle that by either skipping SHA checks, or running a standalone sha256sum
binary.
"""
try:
import hashlib
except ImportError:
return None
s = hashlib.sha256()
for line in open(filename):
s.update(line)
return s.hexdigest()
def preserved_envvars_list():
return [
'BBPATH',
'BB_PRESERVE_ENV',
'BB_ENV_WHITELIST',
'BB_ENV_EXTRAWHITE',
'COLORTERM',
'DBUS_SESSION_BUS_ADDRESS',
'DESKTOP_SESSION',
'DESKTOP_STARTUP_ID',
'DISPLAY',
'GNOME_KEYRING_PID',
'GNOME_KEYRING_SOCKET',
'GPG_AGENT_INFO',
'GTK_RC_FILES',
'HOME',
'LANG',
'LOGNAME',
'PATH',
'PWD',
'SESSION_MANAGER',
'SHELL',
'SSH_AUTH_SOCK',
'TERM',
'USER',
'USERNAME',
'_',
'XAUTHORITY',
'XDG_DATA_DIRS',
'XDG_SESSION_COOKIE',
]
def filter_environment(good_vars):
"""
Create a pristine environment for bitbake. This will remove variables that
are not known and may influence the build in a negative way.
"""
import bb
removed_vars = []
for key in os.environ.keys():
if key in good_vars:
continue
removed_vars.append(key)
os.unsetenv(key)
del os.environ[key]
if len(removed_vars):
bb.debug(1, "Removed the following variables from the environment:", ",".join(removed_vars))
return removed_vars
def clean_environment():
"""
Clean up any spurious environment variables. This will remove any
variables the user hasn't chose to preserve.
"""
if 'BB_PRESERVE_ENV' not in os.environ:
if 'BB_ENV_WHITELIST' in os.environ:
good_vars = os.environ['BB_ENV_WHITELIST'].split()
else:
good_vars = preserved_envvars_list()
if 'BB_ENV_EXTRAWHITE' in os.environ:
good_vars.extend(os.environ['BB_ENV_EXTRAWHITE'].split())
filter_environment(good_vars)
def empty_environment():
"""
Remove all variable from the environment.
"""
for s in os.environ.keys():
os.unsetenv(s)
del os.environ[s]
def build_environment(d):
"""
Build an environment from all exported variables.
"""
import bb
for var in bb.data.keys(d):
export = bb.data.getVarFlag(var, "export", d)
if export:
os.environ[var] = bb.data.getVar(var, d, True)
def prunedir(topdir):
# Delete everything reachable from the directory named in 'topdir'.
# CAUTION: This is dangerous!
for root, dirs, files in os.walk(topdir, topdown=False):
for name in files:
os.remove(os.path.join(root, name))
for name in dirs:
if os.path.islink(os.path.join(root, name)):
os.remove(os.path.join(root, name))
else:
os.rmdir(os.path.join(root, name))
os.rmdir(topdir)
#
# Could also use return re.compile("(%s)" % "|".join(map(re.escape, suffixes))).sub(lambda mo: "", var)
# but thats possibly insane and suffixes is probably going to be small
#
def prune_suffix(var, suffixes, d):
# See if var ends with any of the suffixes listed and
# remove it if found
for suffix in suffixes:
if var.endswith(suffix):
return var.replace(suffix, "")
return var

View File

@@ -0,0 +1,145 @@
#
# BitBake XMLRPC Server
#
# Copyright (C) 2006 - 2007 Michael 'Mickey' Lauer
# Copyright (C) 2006 - 2008 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
"""
This module implements an xmlrpc server for BitBake.
Use this by deriving a class from BitBakeXMLRPCServer and then adding
methods which you want to "export" via XMLRPC. If the methods have the
prefix xmlrpc_, then registering those function will happen automatically,
if not, you need to call register_function.
Use register_idle_function() to add a function which the xmlrpc server
calls from within server_forever when no requests are pending. Make sure
that those functions are non-blocking or else you will introduce latency
in the server's main loop.
"""
import bb
import xmlrpclib
DEBUG = False
from SimpleXMLRPCServer import SimpleXMLRPCServer, SimpleXMLRPCRequestHandler
import inspect, select
class BitBakeServerCommands():
def __init__(self, server, cooker):
self.cooker = cooker
self.server = server
def registerEventHandler(self, host, port):
"""
Register a remote UI Event Handler
"""
s = xmlrpclib.Server("http://%s:%d" % (host, port), allow_none=True)
return bb.event.register_UIHhandler(s)
def unregisterEventHandler(self, handlerNum):
"""
Unregister a remote UI Event Handler
"""
return bb.event.unregister_UIHhandler(handlerNum)
def runCommand(self, command):
"""
Run a cooker command on the server
"""
return self.cooker.command.runCommand(command)
def terminateServer(self):
"""
Trigger the server to quit
"""
self.server.quit = True
print "Server (cooker) exitting"
return
def ping(self):
"""
Dummy method which can be used to check the server is still alive
"""
return True
class BitBakeXMLRPCServer(SimpleXMLRPCServer):
# remove this when you're done with debugging
# allow_reuse_address = True
def __init__(self, cooker, interface = ("localhost", 0)):
"""
Constructor
"""
SimpleXMLRPCServer.__init__(self, interface,
requestHandler=SimpleXMLRPCRequestHandler,
logRequests=False, allow_none=True)
self._idlefuns = {}
self.host, self.port = self.socket.getsockname()
#self.register_introspection_functions()
commands = BitBakeServerCommands(self, cooker)
self.autoregister_all_functions(commands, "")
def autoregister_all_functions(self, context, prefix):
"""
Convenience method for registering all functions in the scope
of this class that start with a common prefix
"""
methodlist = inspect.getmembers(context, inspect.ismethod)
for name, method in methodlist:
if name.startswith(prefix):
self.register_function(method, name[len(prefix):])
def register_idle_function(self, function, data):
"""Register a function to be called while the server is idle"""
assert callable(function)
self._idlefuns[function] = data
def serve_forever(self):
"""
Serve Requests. Overloaded to honor a quit command
"""
self.quit = False
while not self.quit:
#print "Idle queue length %s" % len(self._idlefuns)
if len(self._idlefuns) == 0:
self.timeout = None
else:
self.timeout = 0
self.handle_request()
#print "Idle timeout, running idle functions"
for function, data in self._idlefuns.items():
try:
retval = function(self, data, False)
if not retval:
del self._idlefuns[function]
except SystemExit:
raise
except:
import traceback
traceback.print_exc()
pass
# Tell idle functions we're exiting
for function, data in self._idlefuns.items():
try:
retval = function(self, data, True)
except:
pass
self.server_close()
return

View File

@@ -1,8 +1,8 @@
Tim Ansell <mithro@mithis.net>
Phil Blundell <pb@handhelds.org>
Seb Frankengul <seb@frankengul.org>
Holger Freyther <holger@moiji-mobile.com>
Marcin Juszkiewicz <marcin@juszkiewicz.com.pl>
Holger Freyther <zecke@handhelds.org>
Marcin Juszkiewicz <marcin@haerwu.biz>
Chris Larson <kergoth@handhelds.org>
Ulrich Luckas <luckas@musoft.de>
Mickey Lauer <mickey@Vanille.de>

View File

@@ -1,99 +1,8 @@
Changes in Bitbake 1.9.x:
- Add PE (Package Epoch) support from Philipp Zabel (pH5)
- Treat python functions the same as shell functions for logging
- Use TMPDIR/anonfunc as a __anonfunc temp directory (T)
- Catch truncated cache file errors
- Allow operations other than assignment on flag variables
- Add code to handle inter-task dependencies
- Fix cache errors when generation dotGraphs
- Make sure __inherit_cache is updated before calling include() (from Michael Krelin)
- Fix bug when target was in ASSUME_PROVIDED (#2236)
- Raise ParseError for filenames with multiple underscores instead of infinitely looping (#2062)
- Fix invalid regexp in BBMASK error handling (missing import) (#1124)
- Promote certain warnings from debug to note 2 level
- Update manual
- Correctly redirect stdin when forking
- If parsing errors are found, exit, too many users miss the errors
- Remove supriours PREFERRED_PROVIDER warnings
- svn fetcher: Add _buildsvncommand function
- Improve certain error messages
- Rewrite svn fetcher to make adding extra operations easier
as part of future SRCDATE="now" fixes
(requires new FETCHCMD_svn definition in bitbake.conf)
- Change SVNDIR layout to be more unique (fixes #2644 and #2624)
- Add ConfigParsed Event after configuration parsing is complete
- Add SRCREV support for svn fetcher
- data.emit_var() - only call getVar if we need the variable
- Stop generating the A variable (seems to be legacy code)
- Make sure intertask depends get processed correcting in recursive depends
- Add pn-PN to overrides when evaluating PREFERRED_VERSION
- Improve the progress indicator by skipping tasks that have
already run before starting the build rather than during it
- Add profiling option (-P)
- Add BB_SRCREV_POLICY variable (clear or cache) to control SRCREV cache
- Add SRCREV_FORMAT support
- Fix local fetcher's localpath return values
- Apply OVERRIDES before performing immediate expansions
- Allow the -b -e option combination to take regular expressions
- Fix handling of variables with expansion in the name using _append/_prepend
e.g. RRECOMMENDS_${PN}_append_xyz = "abc"
- Add plain message function to bb.msg
- Sort the list of providers before processing so dependency problems are
reproducible rather than effectively random
- Fix/improve bitbake -s output
- Add locking for fetchers so only one tries to fetch a given file at a given time
- Fix int(0)/None confusion in runqueue.py which causes random gaps in dependency chains
- Expand data in addtasks
- Print the list of missing DEPENDS,RDEPENDS for the "No buildable providers available for required...."
error message.
- Rework add_task to be more efficient (6% speedup, 7% number of function calls reduction)
- Sort digraph output to make builds more reproducible
- Split expandKeys into two for loops to benefit from the expand_cache (12% speedup)
- runqueue.py: Fix idepends handling to avoid dependency errors
- Clear the terminal TOSTOP flag if set (and warn the user)
- Fix regression from r653 and make SRCDATE/CVSDATE work for packages again
- Fix a bug in bb.decodeurl where http://some.where.com/somefile.tgz decoded to host="" (#1530)
- Warn about malformed PREFERRED_PROVIDERS (#1072)
- Add support for BB_NICE_LEVEL option (#1627)
- Psyco is used only on x86 as there is no support for other architectures.
- Sort initial providers list by default preference (#1145, #2024)
- Improve provider sorting so prefered versions have preference over latest versions (#768)
- Detect builds of tasks with overlapping providers and warn (will become a fatal error) (#1359)
- Add MULTI_PROVIDER_WHITELIST variable to allow known safe multiple providers to be listed
- Handle paths in svn fetcher module parameter
- Support the syntax "export VARIABLE"
- Add bzr fetcher
- Add support for cleaning directories before a task in the form:
do_taskname[cleandirs] = "dir"
- bzr fetcher tweaks from Robert Schuster (#2913)
- Add mercurial (hg) fetcher from Robert Schuster (#2913)
- Don't add duplicates to BBPATH
- Fix preferred_version return values (providers.py)
- Fix 'depends' flag splitting
- Fix unexport handling (#3135)
- Add bb.copyfile function similar to bb.movefile (and improve movefile error reporting)
- Allow multiple options for deptask flag
- Use git-fetch instead of git-pull removing any need for merges when
fetching (we don't care about the index). Fixes fetch errors.
- Add BB_GENERATE_MIRROR_TARBALLS option, set to 0 to make git fetches
faster at the expense of not creating mirror tarballs.
- SRCREV handling updates, improvements and fixes from Poky
- Add bb.utils.lockfile() and bb.utils.unlockfile() from Poky
- Add support for task selfstamp and lockfiles flags
- Disable task number acceleration since it can allow the tasks to run
out of sequence
- Improve runqueue code comments
- Add task scheduler abstraction and some example schedulers
- Improve circular dependency chain debugging code and user feedback
- Don't give a stacktrace for invalid tasks, have a user friendly message (#3431)
- Add support for "-e target" (#3432)
- Fix shell showdata command (#3259)
- Fix shell data updating problems (#1880)
- Properly raise errors for invalid source URI protocols
- Change the wget fetcher failure handling to avoid lockfile problems
- Add support for branches in git fetcher (Otavio Salvador, Michael Lauer)
- Make taskdata and runqueue errors more user friendly
- Add norecurse and fullpath options to cvs fetcher
Changes in BitBake 1.8.x:
- Add bb.utils.prune_suffix function
Changes in BitBake 1.8.12:
- Fix -f (force) in conjunction with -b
- Fix exit code for build failures in --continue mode
- Fix git branch tags fetching
- Change parseConfigurationFile so it works on real data, not a copy
@@ -118,10 +27,8 @@ Changes in Bitbake 1.9.x:
how extensively stamps are looked at for validity
- When handling build target failures make sure idepends are checked and
failed where needed. Fixes --continue mode crashes.
- Fix -f (force) in conjunction with -b
- Fix problems with recrdeptask handling where some idepends weren't handled
correctly.
- Handle exit codes correctly (from pH5)
- Work around refs/HEAD issues with git over http (#3410)
- Add proxy support to the CVS fetcher (from Cyril Chemparathy)
- Improve runfetchcmd so errors are seen and various GIT variables are exported
@@ -137,8 +44,9 @@ Changes in Bitbake 1.9.x:
- Add PERSISTENT_DIR to store the PersistData in a persistent
directory != the cache dir.
- Add md5 and sha256 checksum generation functions to utils.py
- Make sure Build Completed events are generated even when tasks fail
- Correctly handle '-' characters in class names (#2958)
- Make sure expandKeys has been called on the data dictionary before running tasks
- Make sure expandKeys has been called on the data dictonary before running tasks
- Correctly add a task override in the form task-TASKNAME.
- Revert the '-' character fix in class names since it breaks things
- When a regexp fails to compile for PACKAGES_DYNAMIC, print a more useful error (#4444)
@@ -155,7 +63,6 @@ Changes in Bitbake 1.9.x:
used instead of the internal bitbake one. Alternatively, BB_ENV_EXTRAWHITE can be used
to extend the internal whitelist.
- Perforce fetcher fix to use commandline options instead of being overriden by the environment
- bb.utils.prunedir can cope with symlinks to directoriees without exceptions
- use @rev when doing a svn checkout
- Add osc fetcher (from Joshua Lock in Poky)
- When SRCREV autorevisioning for a recipe is in use, don't cache the recipe
@@ -169,15 +76,109 @@ Changes in Bitbake 1.9.x:
proxies to work better. (from Poky)
- Also allow user and pswd options in SRC_URIs globally (from Poky)
- Improve proxy handling when using mirrors (from Poky)
- Add bb.utils.prune_suffix function
- Fix hg checkouts of specific revisions (from Poky)
- Fix wget fetching of urls with parameters specified (from Poky)
- Add username handling to git fetcher (from Poky)
- Set HOME environmental variable when running fetcher commands (from Poky)
- Make sure allowed variables inherited from the environment are exported again (from Poky)
- When running a stage task in bbshell, run populate_staging, not the stage task (from Poky)
- Fix + character escaping from PACKAGES_DYNAMIC (thanks Otavio Salvador)
- Addition of BBCLASSEXTEND support for allowing one recipe to provide multiple targets (from Poky)
Changes in BitBake 1.8.10:
- Psyco is available only for x86 - do not use it on other architectures.
- Fix a bug in bb.decodeurl where http://some.where.com/somefile.tgz decoded to host="" (#1530)
- Warn about malformed PREFERRED_PROVIDERS (#1072)
- Add support for BB_NICE_LEVEL option (#1627)
- Sort initial providers list by default preference (#1145, #2024)
- Improve provider sorting so prefered versions have preference over latest versions (#768)
- Detect builds of tasks with overlapping providers and warn (will become a fatal error) (#1359)
- Add MULTI_PROVIDER_WHITELIST variable to allow known safe multiple providers to be listed
- Handle paths in svn fetcher module parameter
- Support the syntax "export VARIABLE"
- Add bzr fetcher
- Add support for cleaning directories before a task in the form:
do_taskname[cleandirs] = "dir"
- bzr fetcher tweaks from Robert Schuster (#2913)
- Add mercurial (hg) fetcher from Robert Schuster (#2913)
- Fix bogus preferred_version return values
- Fix 'depends' flag splitting
- Fix unexport handling (#3135)
- Add bb.copyfile function similar to bb.movefile (and improve movefile error reporting)
- Allow multiple options for deptask flag
- Use git-fetch instead of git-pull removing any need for merges when
fetching (we don't care about the index). Fixes fetch errors.
- Add BB_GENERATE_MIRROR_TARBALLS option, set to 0 to make git fetches
faster at the expense of not creating mirror tarballs.
- SRCREV handling updates, improvements and fixes from Poky
- Add bb.utils.lockfile() and bb.utils.unlockfile() from Poky
- Add support for task selfstamp and lockfiles flags
- Disable task number acceleration since it can allow the tasks to run
out of sequence
- Improve runqueue code comments
- Add task scheduler abstraction and some example schedulers
- Improve circular dependency chain debugging code and user feedback
- Don't give a stacktrace for invalid tasks, have a user friendly message (#3431)
- Add support for "-e target" (#3432)
- Fix shell showdata command (#3259)
- Fix shell data updating problems (#1880)
- Properly raise errors for invalid source URI protocols
- Change the wget fetcher failure handling to avoid lockfile problems
- Add git branch support
- Add support for branches in git fetcher (Otavio Salvador, Michael Lauer)
- Make taskdata and runqueue errors more user friendly
- Add norecurse and fullpath options to cvs fetcher
- bb.utils.prunedir can cope with symlinks to directories without exceptions
Changes in Bitbake 1.8.8:
- Rewrite svn fetcher to make adding extra operations easier
as part of future SRCDATE="now" fixes
(requires new FETCHCMD_svn definition in bitbake.conf)
- Change SVNDIR layout to be more unique (fixes #2644 and #2624)
- Import persistent data store from trunk
- Sync fetcher code with that in trunk, adding SRCREV support for svn
- Add ConfigParsed Event after configuration parsing is complete
- data.emit_var() - only call getVar if we need the variable
- Stop generating the A variable (seems to be legacy code)
- Make sure intertask depends get processed correcting in recursive depends
- Add pn-PN to overrides when evaluating PREFERRED_VERSION
- Improve the progress indicator by skipping tasks that have
already run before starting the build rather than during it
- Add profiling option (-P)
- Add BB_SRCREV_POLICY variable (clear or cache) to control SRCREV cache
- Add SRCREV_FORMAT support
- Fix local fetcher's localpath return values
- Apply OVERRIDES before performing immediate expansions
- Allow the -b -e option combination to take regular expressions
- Add plain message function to bb.msg
- Sort the list of providers before processing so dependency problems are
reproducible rather than effectively random
- Add locking for fetchers so only one tries to fetch a given file at a given time
- Fix int(0)/None confusion in runqueue.py which causes random gaps in dependency chains
- Fix handling of variables with expansion in the name using _append/_prepend
e.g. RRECOMMENDS_${PN}_append_xyz = "abc"
- Expand data in addtasks
- Print the list of missing DEPENDS,RDEPENDS for the "No buildable providers available for required...."
error message.
- Rework add_task to be more efficient (6% speedup, 7% number of function calls reduction)
- Sort digraph output to make builds more reproducible
- Split expandKeys into two for loops to benefit from the expand_cache (12% speedup)
- runqueue.py: Fix idepends handling to avoid dependency errors
- Clear the terminal TOSTOP flag if set (and warn the user)
- Fix regression from r653 and make SRCDATE/CVSDATE work for packages again
Changes in Bitbake 1.8.6:
- Correctly redirect stdin when forking
- If parsing errors are found, exit, too many users miss the errors
- Remove supriours PREFERRED_PROVIDER warnings
Changes in Bitbake 1.8.4:
- Make sure __inherit_cache is updated before calling include() (from Michael Krelin)
- Fix bug when target was in ASSUME_PROVIDED (#2236)
- Raise ParseError for filenames with multiple underscores instead of infinitely looping (#2062)
- Fix invalid regexp in BBMASK error handling (missing import) (#1124)
- Don't run build sanity checks on incomplete builds
- Promote certain warnings from debug to note 2 level
- Update manual
Changes in Bitbake 1.8.2:
- Catch truncated cache file errors
- Add PE (Package Epoch) support from Philipp Zabel (pH5)
- Add code to handle inter-task dependencies
- Allow operations other than assignment on flag variables
- Fix cache errors when generation dotGraphs
Changes in Bitbake 1.8.0:
- Release 1.7.x as a stable series

View File

@@ -1,10 +0,0 @@
BitBake is licensed under the GNU General Public License version 2.0. See COPYING for further details.
The following external components are distributed with this software:
* The Toaster Simple UI application is based upon the Django project template, the files of which are covered by the BSD license and are copyright (c) Django Software
Foundation and individual contributors.
* Twitter Bootstrap (including Glyphicons), redistributed under the Apache License 2.0.
* jQuery is redistributed under the MIT license.

53
bitbake/MANIFEST Normal file
View File

@@ -0,0 +1,53 @@
AUTHORS
COPYING
ChangeLog
MANIFEST
setup.py
bin/bitdoc
bin/bbimage
bin/bitbake
lib/bb/__init__.py
lib/bb/build.py
lib/bb/cache.py
lib/bb/cooker.py
lib/bb/COW.py
lib/bb/data.py
lib/bb/data_smart.py
lib/bb/event.py
lib/bb/fetch/__init__.py
lib/bb/fetch/bzr.py
lib/bb/fetch/cvs.py
lib/bb/fetch/git.py
lib/bb/fetch/hg.py
lib/bb/fetch/local.py
lib/bb/fetch/osc.py
lib/bb/fetch/perforce.py
lib/bb/fetch/ssh.py
lib/bb/fetch/svk.py
lib/bb/fetch/svn.py
lib/bb/fetch/wget.py
lib/bb/manifest.py
lib/bb/methodpool.py
lib/bb/msg.py
lib/bb/parse/__init__.py
lib/bb/parse/parse_py/__init__.py
lib/bb/parse/parse_py/BBHandler.py
lib/bb/parse/parse_py/ConfHandler.py
lib/bb/persist_data.py
lib/bb/providers.py
lib/bb/runqueue.py
lib/bb/shell.py
lib/bb/taskdata.py
lib/bb/utils.py
setup.py
doc/COPYING.GPL
doc/COPYING.MIT
doc/bitbake.1
doc/manual/html.css
doc/manual/Makefile
doc/manual/usermanual.xml
contrib/bbdev.sh
contrib/vim/syntax/bitbake.vim
contrib/vim/ftdetect/bitbake.vim
conf/bitbake.conf
classes/base.bbclass

View File

@@ -22,341 +22,125 @@
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys, logging
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(__file__)),
'lib'))
import optparse
import warnings
from traceback import format_exception
try:
import bb
except RuntimeError as exc:
sys.exit(str(exc))
from bb import event
import bb.msg
import sys, os, getopt, re, time, optparse
sys.path.insert(0,os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb
from bb import cooker
from bb import ui
from bb import server
from bb import cookerdata
__version__ = "1.22.0"
logger = logging.getLogger("BitBake")
__version__ = "1.8.13"
# Python multiprocessing requires /dev/shm
if not os.access('/dev/shm', os.W_OK | os.X_OK):
sys.exit("FATAL: /dev/shm does not exist or is not writable")
# Unbuffer stdout to avoid log truncation in the event
# of an unorderly exit as well as to provide timely
# updates to log files for use with tail
try:
if sys.stdout.name == '<stdout>':
sys.stdout = os.fdopen(sys.stdout.fileno(), 'w', 0)
except:
pass
def get_ui(config):
if not config.ui:
# modify 'ui' attribute because it is also read by cooker
config.ui = os.environ.get('BITBAKE_UI', 'knotty')
interface = config.ui
try:
# Dynamically load the UI based on the ui name. Although we
# suggest a fixed set this allows you to have flexibility in which
# ones are available.
module = __import__("bb.ui", fromlist = [interface])
return getattr(module, interface)
except AttributeError:
sys.exit("FATAL: Invalid user interface '%s' specified.\n"
"Valid interfaces: depexp, goggle, ncurses, hob, knotty [default]." % interface)
# Display bitbake/OE warnings via the BitBake.Warnings logger, ignoring others"""
warnlog = logging.getLogger("BitBake.Warnings")
_warnings_showwarning = warnings.showwarning
def _showwarning(message, category, filename, lineno, file=None, line=None):
if file is not None:
if _warnings_showwarning is not None:
_warnings_showwarning(message, category, filename, lineno, file, line)
else:
s = warnings.formatwarning(message, category, filename, lineno)
warnlog.warn(s)
warnings.showwarning = _showwarning
warnings.filterwarnings("ignore")
warnings.filterwarnings("default", module="(<string>$|(oe|bb)\.)")
warnings.filterwarnings("ignore", category=PendingDeprecationWarning)
warnings.filterwarnings("ignore", category=ImportWarning)
warnings.filterwarnings("ignore", category=DeprecationWarning, module="<string>$")
warnings.filterwarnings("ignore", message="With-statements now directly support multiple context managers")
class BitBakeConfigParameters(cookerdata.ConfigParameters):
def parseCommandLine(self):
parser = optparse.OptionParser(
version = "BitBake Build Tool Core version %s, %%prog version %s" % (bb.__version__, __version__),
usage = """%prog [options] [recipename/target ...]
Executes the specified task (default is 'build') for a given set of target recipes (.bb files).
It is assumed there is a conf/bblayers.conf available in cwd or in BBPATH which
will provide the layer, BBFILES and other configuration information.""")
parser.add_option("-b", "--buildfile", help = "Execute tasks from a specific .bb recipe directly. WARNING: Does not handle any dependencies from other recipes.",
action = "store", dest = "buildfile", default = None)
parser.add_option("-k", "--continue", help = "Continue as much as possible after an error. While the target that failed and anything depending on it cannot be built, as much as possible will be built before stopping.",
action = "store_false", dest = "abort", default = True)
parser.add_option("-a", "--tryaltconfigs", help = "Continue with builds by trying to use alternative providers where possible.",
action = "store_true", dest = "tryaltconfigs", default = False)
parser.add_option("-f", "--force", help = "Force the specified targets/task to run (invalidating any existing stamp file).",
action = "store_true", dest = "force", default = False)
parser.add_option("-c", "--cmd", help = "Specify the task to execute. The exact options available depend on the metadata. Some examples might be 'compile' or 'populate_sysroot' or 'listtasks' may give a list of the tasks available.",
action = "store", dest = "cmd")
parser.add_option("-C", "--clear-stamp", help = "Invalidate the stamp for the specified task such as 'compile' and then run the default task for the specified target(s).",
action = "store", dest = "invalidate_stamp")
parser.add_option("-r", "--read", help = "Read the specified file before bitbake.conf.",
action = "append", dest = "prefile", default = [])
parser.add_option("-R", "--postread", help = "Read the specified file after bitbake.conf.",
action = "append", dest = "postfile", default = [])
parser.add_option("-v", "--verbose", help = "Output more log message data to the terminal.",
action = "store_true", dest = "verbose", default = False)
parser.add_option("-D", "--debug", help = "Increase the debug level. You can specify this more than once.",
action = "count", dest="debug", default = 0)
parser.add_option("-n", "--dry-run", help = "Don't execute, just go through the motions.",
action = "store_true", dest = "dry_run", default = False)
parser.add_option("-S", "--dump-signatures", help = "Dump out the signature construction information, with no task execution. Parameters are passed to the signature handling code, use 'none' if no specific handler is required.",
action = "append", dest = "dump_signatures", default = [])
parser.add_option("-p", "--parse-only", help = "Quit after parsing the BB recipes.",
action = "store_true", dest = "parse_only", default = False)
parser.add_option("-s", "--show-versions", help = "Show current and preferred versions of all recipes.",
action = "store_true", dest = "show_versions", default = False)
parser.add_option("-e", "--environment", help = "Show the global or per-package environment complete with information about where variables were set/changed.",
action = "store_true", dest = "show_environment", default = False)
parser.add_option("-g", "--graphviz", help = "Save dependency tree information for the specified targets in the dot syntax.",
action = "store_true", dest = "dot_graph", default = False)
parser.add_option("-I", "--ignore-deps", help = """Assume these dependencies don't exist and are already provided (equivalent to ASSUME_PROVIDED). Useful to make dependency graphs more appealing""",
action = "append", dest = "extra_assume_provided", default = [])
parser.add_option("-l", "--log-domains", help = """Show debug logging for the specified logging domains""",
action = "append", dest = "debug_domains", default = [])
parser.add_option("-P", "--profile", help = "Profile the command and save reports.",
action = "store_true", dest = "profile", default = False)
parser.add_option("-u", "--ui", help = "The user interface to use (e.g. knotty, hob, depexp).",
action = "store", dest = "ui")
parser.add_option("-t", "--servertype", help = "Choose which server to use, process or xmlrpc.",
action = "store", dest = "servertype")
parser.add_option("", "--revisions-changed", help = "Set the exit code depending on whether upstream floating revisions have changed or not.",
action = "store_true", dest = "revisions_changed", default = False)
parser.add_option("", "--server-only", help = "Run bitbake without a UI, only starting a server (cooker) process.",
action = "store_true", dest = "server_only", default = False)
parser.add_option("-B", "--bind", help = "The name/address for the bitbake server to bind to.",
action = "store", dest = "bind", default = False)
parser.add_option("", "--no-setscene", help = "Do not run any setscene tasks. sstate will be ignored and everything needed, built.",
action = "store_true", dest = "nosetscene", default = False)
parser.add_option("", "--remote-server", help = "Connect to the specified server.",
action = "store", dest = "remote_server", default = False)
parser.add_option("-m", "--kill-server", help = "Terminate the remote server.",
action = "store_true", dest = "kill_server", default = False)
parser.add_option("", "--observe-only", help = "Connect to a server as an observing-only client.",
action = "store_true", dest = "observe_only", default = False)
parser.add_option("", "--status-only", help = "Check the status of the remote bitbake server.",
action = "store_true", dest = "status_only", default = False)
options, targets = parser.parse_args(sys.argv)
# some environmental variables set also configuration options
if "BBSERVER" in os.environ:
options.servertype = "xmlrpc"
options.remote_server = os.environ["BBSERVER"]
return options, targets[1:]
def start_server(servermodule, configParams, configuration, features):
server = servermodule.BitBakeServer()
if configParams.bind:
(host, port) = configParams.bind.split(':')
server.initServer((host, int(port)))
configuration.interface = [ server.serverImpl.host, server.serverImpl.port ]
else:
server.initServer()
configuration.interface = []
try:
configuration.setServerRegIdleCallback(server.getServerIdleCB())
cooker = bb.cooker.BBCooker(configuration, features)
server.addcooker(cooker)
server.saveConnectionDetails()
except Exception as e:
exc_info = sys.exc_info()
while hasattr(server, "event_queue"):
try:
import queue
except ImportError:
import Queue as queue
try:
event = server.event_queue.get(block=False)
except (queue.Empty, IOError):
break
if isinstance(event, logging.LogRecord):
logger.handle(event)
raise exc_info[1], None, exc_info[2]
server.detach()
return server
#============================================================================#
# BBOptions
#============================================================================#
class BBConfiguration( object ):
"""
Manages build options and configurations for one run
"""
def __init__( self, options ):
for key, val in options.__dict__.items():
setattr( self, key, val )
#============================================================================#
# main
#============================================================================#
def main():
parser = optparse.OptionParser( version = "BitBake Build Tool Core version %s, %%prog version %s" % ( bb.__version__, __version__ ),
usage = """%prog [options] [package ...]
configParams = BitBakeConfigParameters()
configuration = cookerdata.CookerConfiguration()
configuration.setConfigParameters(configParams)
Executes the specified task (default is 'build') for a given set of BitBake files.
It expects that BBFILES is defined, which is a space separated list of files to
be executed. BBFILES does support wildcards.
Default BBFILES are the .bb files in the current directory.""" )
ui_module = get_ui(configParams)
parser.add_option( "-b", "--buildfile", help = "execute the task against this .bb file, rather than a package from BBFILES.",
action = "store", dest = "buildfile", default = None )
# Server type can be xmlrpc or process currently, if nothing is specified,
# the default server is process
if configParams.servertype:
server_type = configParams.servertype
else:
server_type = 'process'
parser.add_option( "-k", "--continue", help = "continue as much as possible after an error. While the target that failed, and those that depend on it, cannot be remade, the other dependencies of these targets can be processed all the same.",
action = "store_false", dest = "abort", default = True )
try:
module = __import__("bb.server", fromlist = [server_type])
servermodule = getattr(module, server_type)
except AttributeError:
sys.exit("FATAL: Invalid server type '%s' specified.\n"
"Valid interfaces: xmlrpc, process [default]." % server_type)
parser.add_option( "-a", "--tryaltconfigs", help = "continue with builds by trying to use alternative providers where possible.",
action = "store_true", dest = "tryaltconfigs", default = False )
if configParams.server_only:
if configParams.servertype != "xmlrpc":
sys.exit("FATAL: If '--server-only' is defined, we must set the servertype as 'xmlrpc'.\n")
if not configParams.bind:
sys.exit("FATAL: The '--server-only' option requires a name/address to bind to with the -B option.\n")
if configParams.remote_server:
sys.exit("FATAL: The '--server-only' option conflicts with %s.\n" %
("the BBSERVER environment variable" if "BBSERVER" in os.environ else "the '--remote-server' option" ))
parser.add_option( "-f", "--force", help = "force run of specified cmd, regardless of stamp status",
action = "store_true", dest = "force", default = False )
if configParams.bind and configParams.servertype != "xmlrpc":
sys.exit("FATAL: If '-B' or '--bind' is defined, we must set the servertype as 'xmlrpc'.\n")
parser.add_option( "-i", "--interactive", help = "drop into the interactive mode also called the BitBake shell.",
action = "store_true", dest = "interactive", default = False )
if configParams.remote_server and configParams.servertype != "xmlrpc":
sys.exit("FATAL: If '--remote-server' is defined, we must set the servertype as 'xmlrpc'.\n")
parser.add_option( "-c", "--cmd", help = "Specify task to execute. Note that this only executes the specified task for the providee and the packages it depends on, i.e. 'compile' does not implicitly call stage for the dependencies (IOW: use only if you know what you are doing). Depending on the base.bbclass a listtasks tasks is defined and will show available tasks",
action = "store", dest = "cmd" )
if configParams.observe_only and (not configParams.remote_server or configParams.bind):
sys.exit("FATAL: '--observe-only' can only be used by UI clients connecting to a server.\n")
parser.add_option( "-r", "--read", help = "read the specified file before bitbake.conf",
action = "append", dest = "file", default = [] )
if "BBDEBUG" in os.environ:
level = int(os.environ["BBDEBUG"])
if level > configuration.debug:
configuration.debug = level
parser.add_option( "-v", "--verbose", help = "output more chit-chat to the terminal",
action = "store_true", dest = "verbose", default = False )
bb.msg.init_msgconfig(configParams.verbose, configuration.debug,
configuration.debug_domains)
parser.add_option( "-D", "--debug", help = "Increase the debug level. You can specify this more than once.",
action = "count", dest="debug", default = 0)
# Ensure logging messages get sent to the UI as events
handler = bb.event.LogHandler()
if not configParams.status_only:
# In status only mode there are no logs and no UI
logger.addHandler(handler)
parser.add_option( "-n", "--dry-run", help = "don't execute, just go through the motions",
action = "store_true", dest = "dry_run", default = False )
# Clear away any spurious environment variables while we stoke up the cooker
cleanedvars = bb.utils.clean_environment()
parser.add_option( "-p", "--parse-only", help = "quit after parsing the BB files (developers only)",
action = "store_true", dest = "parse_only", default = False )
featureset = []
if not configParams.server_only:
# Collect the feature set for the UI
featureset = getattr(ui_module, "featureSet", [])
parser.add_option( "-d", "--disable-psyco", help = "disable using the psyco just-in-time compiler (not recommended)",
action = "store_true", dest = "disable_psyco", default = False )
if not configParams.remote_server:
# we start a server with a given configuration
server = start_server(servermodule, configParams, configuration, featureset)
bb.event.ui_queue = []
else:
# we start a stub server that is actually a XMLRPClient that connects to a real server
server = servermodule.BitBakeXMLRPCClient(configParams.observe_only)
server.saveConnectionDetails(configParams.remote_server)
server.saveConnectionConfigParams(configParams)
parser.add_option( "-s", "--show-versions", help = "show current and preferred versions of all packages",
action = "store_true", dest = "show_versions", default = False )
if not configParams.server_only:
if configParams.status_only:
try:
server_connection = server.establishConnection(featureset)
except:
sys.exit(1)
if not server_connection:
sys.exit(1)
server_connection.terminate()
sys.exit(0)
parser.add_option( "-e", "--environment", help = "show the global or per-package environment (this is what used to be bbread)",
action = "store_true", dest = "show_environment", default = False )
# Setup a connection to the server (cooker)
server_connection = server.establishConnection(featureset)
if not server_connection:
if configParams.kill_server:
bb.fatal("Server already killed")
configParams.bind = configParams.remote_server
start_server(servermodule, configParams, configuration, featureset)
bb.event.ui_queue = []
server_connection = server.establishConnection(featureset)
parser.add_option( "-g", "--graphviz", help = "emit the dependency trees of the specified packages in the dot syntax",
action = "store_true", dest = "dot_graph", default = False )
# Restore the environment in case the UI needs it
for k in cleanedvars:
os.environ[k] = cleanedvars[k]
parser.add_option( "-I", "--ignore-deps", help = """Stop processing at the given list of dependencies when generating dependency graphs. This can help to make the graph more appealing""",
action = "append", dest = "ignored_dot_deps", default = [] )
logger.removeHandler(handler)
parser.add_option( "-l", "--log-domains", help = """Show debug logging for the specified logging domains""",
action = "append", dest = "debug_domains", default = [] )
parser.add_option( "-P", "--profile", help = "profile the command and print a report",
action = "store_true", dest = "profile", default = False )
parser.add_option( "", "--revisions-changed", help = "Set the exit code depending on whether upstream floating revisions have changed or not",
action = "store_true", dest = "revisions_changed", default = False )
options, args = parser.parse_args(sys.argv)
configuration = BBConfiguration(options)
configuration.pkgs_to_build = []
configuration.pkgs_to_build.extend(args[1:])
cooker = bb.cooker.BBCooker(configuration)
# Clear away any spurious environment variables. But don't wipe the
# environment totally.
bb.utils.clean_environment()
cooker.parseConfiguration()
if configuration.profile:
try:
return ui_module.main(server_connection.connection, server_connection.events, configParams)
finally:
bb.event.ui_queue = []
server_connection.terminate()
else:
print("server address: %s, server port: %s" % (server.serverImpl.host, server.serverImpl.port))
return 0
import cProfile as profile
except:
import profile
return 1
profile.runctx("cooker.cook()", globals(), locals(), "profile.log")
import pstats
p = pstats.Stats('profile.log')
p.sort_stats('time')
p.print_stats()
p.print_callers()
p.sort_stats('cumulative')
p.print_stats()
else:
cooker.cook()
if __name__ == "__main__":
try:
ret = main()
except bb.BBHandledException:
ret = 1
except Exception:
ret = 1
import traceback
traceback.print_exc()
sys.exit(ret)
main()

View File

@@ -1,122 +0,0 @@
#!/usr/bin/env python
# bitbake-diffsigs
# BitBake task signature data comparison utility
#
# Copyright (C) 2012-2013 Intel Corporation
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys
import warnings
import fnmatch
import optparse
import logging
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb.tinfoil
import bb.siggen
def logger_create(name, output=sys.stderr):
logger = logging.getLogger(name)
console = logging.StreamHandler(output)
format = bb.msg.BBLogFormatter("%(levelname)s: %(message)s")
if output.isatty():
format.enable_color()
console.setFormatter(format)
logger.addHandler(console)
logger.setLevel(logging.INFO)
return logger
logger = logger_create('bitbake-diffsigs')
def find_compare_task(bbhandler, pn, taskname):
""" Find the most recent signature files for the specified PN/task and compare them """
if not hasattr(bb.siggen, 'find_siginfo'):
logger.error('Metadata does not support finding signature data files')
sys.exit(1)
if not taskname.startswith('do_'):
taskname = 'do_%s' % taskname
filedates = bb.siggen.find_siginfo(pn, taskname, None, bbhandler.config_data)
latestfiles = sorted(filedates.keys(), key=lambda f: filedates[f])[-2:]
if not latestfiles:
logger.error('No sigdata files found matching %s %s' % (pn, taskname))
sys.exit(1)
elif len(latestfiles) < 2:
logger.error('Only one matching sigdata file found for the specified task (%s %s)' % (pn, taskname))
sys.exit(1)
else:
# Define recursion callback
def recursecb(key, hash1, hash2):
hashes = [hash1, hash2]
hashfiles = bb.siggen.find_siginfo(key, None, hashes, bbhandler.config_data)
recout = []
if len(hashfiles) == 2:
out2 = bb.siggen.compare_sigfiles(hashfiles[hash1], hashfiles[hash2], recursecb)
recout.extend(list(' ' + l for l in out2))
else:
recout.append("Unable to find matching sigdata for %s with hashes %s or %s" % (key, hash1, hash2))
return recout
# Recurse into signature comparison
output = bb.siggen.compare_sigfiles(latestfiles[0], latestfiles[1], recursecb)
if output:
print '\n'.join(output)
sys.exit(0)
parser = optparse.OptionParser(
description = "Compares siginfo/sigdata files written out by BitBake",
usage = """
%prog -t recipename taskname
%prog sigdatafile1 sigdatafile2
%prog sigdatafile1""")
parser.add_option("-t", "--task",
help = "find the signature data files for last two runs of the specified task and compare them",
action="store", dest="taskargs", nargs=2, metavar='recipename taskname')
options, args = parser.parse_args(sys.argv)
if options.taskargs:
tinfoil = bb.tinfoil.Tinfoil()
tinfoil.prepare(config_only = True)
find_compare_task(tinfoil, options.taskargs[0], options.taskargs[1])
else:
if len(args) == 1:
parser.print_help()
else:
import cPickle
try:
if len(args) == 2:
output = bb.siggen.dump_sigfile(sys.argv[1])
else:
output = bb.siggen.compare_sigfiles(sys.argv[1], sys.argv[2])
except IOError as e:
logger.error(str(e))
sys.exit(1)
except cPickle.UnpicklingError, EOFError:
logger.error('Invalid signature data - ensure you are specifying sigdata/siginfo files')
sys.exit(1)
if output:
print '\n'.join(output)

View File

@@ -1,65 +0,0 @@
#!/usr/bin/env python
# bitbake-dumpsig
# BitBake task signature dump utility
#
# Copyright (C) 2013 Intel Corporation
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys
import warnings
import optparse
import logging
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb.siggen
def logger_create(name, output=sys.stderr):
logger = logging.getLogger(name)
console = logging.StreamHandler(output)
format = bb.msg.BBLogFormatter("%(levelname)s: %(message)s")
if output.isatty():
format.enable_color()
console.setFormatter(format)
logger.addHandler(console)
logger.setLevel(logging.INFO)
return logger
logger = logger_create('bitbake-dumpsig')
parser = optparse.OptionParser(
description = "Dumps siginfo/sigdata files written out by BitBake",
usage = """
%prog sigdatafile""")
options, args = parser.parse_args(sys.argv)
if len(args) == 1:
parser.print_help()
else:
import cPickle
try:
output = bb.siggen.dump_sigfile(args[1])
except IOError as e:
logger.error(str(e))
sys.exit(1)
except cPickle.UnpicklingError, EOFError:
logger.error('Invalid signature data - ensure you are specifying a sigdata/siginfo file')
sys.exit(1)
if output:
print '\n'.join(output)

View File

@@ -1,726 +0,0 @@
#!/usr/bin/env python
# This script has subcommands which operate against your bitbake layers, either
# displaying useful information, or acting against them.
# See the help output for details on available commands.
# Copyright (C) 2011 Mentor Graphics Corporation
# Copyright (C) 2012 Intel Corporation
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import cmd
import logging
import os
import sys
import fnmatch
from collections import defaultdict
import re
bindir = os.path.dirname(__file__)
topdir = os.path.dirname(bindir)
sys.path[0:0] = [os.path.join(topdir, 'lib')]
import bb.cache
import bb.cooker
import bb.providers
import bb.utils
import bb.tinfoil
logger = logging.getLogger('BitBake')
def main(args):
cmds = Commands()
if args:
# Allow user to specify e.g. show-layers instead of show_layers
args = [args[0].replace('-', '_')] + args[1:]
cmds.onecmd(' '.join(args))
else:
cmds.do_help('')
return cmds.returncode
class Commands(cmd.Cmd):
def __init__(self):
self.bbhandler = None
self.returncode = 0
self.bblayers = []
cmd.Cmd.__init__(self)
def init_bbhandler(self, config_only = False):
if not self.bbhandler:
self.bbhandler = bb.tinfoil.Tinfoil()
self.bblayers = (self.bbhandler.config_data.getVar('BBLAYERS', True) or "").split()
self.bbhandler.prepare(config_only)
def default(self, line):
"""Handle unrecognised commands"""
sys.stderr.write("Unrecognised command or option\n")
self.do_help('')
def do_help(self, topic):
"""display general help or help on a specified command"""
if topic:
sys.stdout.write('%s: ' % topic)
cmd.Cmd.do_help(self, topic.replace('-', '_'))
else:
sys.stdout.write("usage: bitbake-layers <command> [arguments]\n\n")
sys.stdout.write("Available commands:\n")
procnames = list(set(self.get_names()))
for procname in procnames:
if procname[:3] == 'do_':
sys.stdout.write(" %s\n" % procname[3:].replace('_', '-'))
doc = getattr(self, procname).__doc__
if doc:
sys.stdout.write(" %s\n" % doc.splitlines()[0])
def do_show_layers(self, args):
"""show current configured layers"""
self.init_bbhandler(config_only = True)
logger.plain("%s %s %s" % ("layer".ljust(20), "path".ljust(40), "priority"))
logger.plain('=' * 74)
for layerdir in self.bblayers:
layername = self.get_layer_name(layerdir)
layerpri = 0
for layer, _, regex, pri in self.bbhandler.cooker.recipecache.bbfile_config_priorities:
if regex.match(os.path.join(layerdir, 'test')):
layerpri = pri
break
logger.plain("%s %s %d" % (layername.ljust(20), layerdir.ljust(40), layerpri))
def version_str(self, pe, pv, pr = None):
verstr = "%s" % pv
if pr:
verstr = "%s-%s" % (verstr, pr)
if pe:
verstr = "%s:%s" % (pe, verstr)
return verstr
def do_show_overlayed(self, args):
"""list overlayed recipes (where the same recipe exists in another layer)
usage: show-overlayed [-f] [-s]
Lists the names of overlayed recipes and the available versions in each
layer, with the preferred version first. Note that skipped recipes that
are overlayed will also be listed, with a " (skipped)" suffix.
Options:
-f instead of the default formatting, list filenames of higher priority
recipes with the ones they overlay indented underneath
-s only list overlayed recipes where the version is the same
"""
self.init_bbhandler()
show_filenames = False
show_same_ver_only = False
for arg in args.split():
if arg == '-f':
show_filenames = True
elif arg == '-s':
show_same_ver_only = True
else:
sys.stderr.write("show-overlayed: invalid option %s\n" % arg)
self.do_help('')
return
items_listed = self.list_recipes('Overlayed recipes', None, True, show_same_ver_only, show_filenames, True)
# Check for overlayed .bbclass files
classes = defaultdict(list)
for layerdir in self.bblayers:
classdir = os.path.join(layerdir, 'classes')
if os.path.exists(classdir):
for classfile in os.listdir(classdir):
if os.path.splitext(classfile)[1] == '.bbclass':
classes[classfile].append(classdir)
# Locating classes and other files is a bit more complicated than recipes -
# layer priority is not a factor; instead BitBake uses the first matching
# file in BBPATH, which is manipulated directly by each layer's
# conf/layer.conf in turn, thus the order of layers in bblayers.conf is a
# factor - however, each layer.conf is free to either prepend or append to
# BBPATH (or indeed do crazy stuff with it). Thus the order in BBPATH might
# not be exactly the order present in bblayers.conf either.
bbpath = str(self.bbhandler.config_data.getVar('BBPATH', True))
overlayed_class_found = False
for (classfile, classdirs) in classes.items():
if len(classdirs) > 1:
if not overlayed_class_found:
logger.plain('=== Overlayed classes ===')
overlayed_class_found = True
mainfile = bb.utils.which(bbpath, os.path.join('classes', classfile))
if show_filenames:
logger.plain('%s' % mainfile)
else:
# We effectively have to guess the layer here
logger.plain('%s:' % classfile)
mainlayername = '?'
for layerdir in self.bblayers:
classdir = os.path.join(layerdir, 'classes')
if mainfile.startswith(classdir):
mainlayername = self.get_layer_name(layerdir)
logger.plain(' %s' % mainlayername)
for classdir in classdirs:
fullpath = os.path.join(classdir, classfile)
if fullpath != mainfile:
if show_filenames:
print(' %s' % fullpath)
else:
print(' %s' % self.get_layer_name(os.path.dirname(classdir)))
if overlayed_class_found:
items_listed = True;
if not items_listed:
logger.plain('No overlayed files found.')
def do_show_recipes(self, args):
"""list available recipes, showing the layer they are provided by
usage: show-recipes [-f] [-m] [pnspec]
Lists the names of overlayed recipes and the available versions in each
layer, with the preferred version first. Optionally you may specify
pnspec to match a specified recipe name (supports wildcards). Note that
skipped recipes will also be listed, with a " (skipped)" suffix.
Options:
-f instead of the default formatting, list filenames of higher priority
recipes with other available recipes indented underneath
-m only list where multiple recipes (in the same layer or different
layers) exist for the same recipe name
"""
self.init_bbhandler()
show_filenames = False
show_multi_provider_only = False
pnspec = None
title = 'Available recipes:'
for arg in args.split():
if arg == '-f':
show_filenames = True
elif arg == '-m':
show_multi_provider_only = True
elif not arg.startswith('-'):
pnspec = arg
title = 'Available recipes matching %s:' % pnspec
else:
sys.stderr.write("show-recipes: invalid option %s\n" % arg)
self.do_help('')
return
self.list_recipes(title, pnspec, False, False, show_filenames, show_multi_provider_only)
def list_recipes(self, title, pnspec, show_overlayed_only, show_same_ver_only, show_filenames, show_multi_provider_only):
pkg_pn = self.bbhandler.cooker.recipecache.pkg_pn
(latest_versions, preferred_versions) = bb.providers.findProviders(self.bbhandler.config_data, self.bbhandler.cooker.recipecache, pkg_pn)
allproviders = bb.providers.allProviders(self.bbhandler.cooker.recipecache)
# Ensure we list skipped recipes
# We are largely guessing about PN, PV and the preferred version here,
# but we have no choice since skipped recipes are not fully parsed
skiplist = self.bbhandler.cooker.skiplist.keys()
skiplist.sort( key=lambda fileitem: self.bbhandler.cooker.collection.calc_bbfile_priority(fileitem) )
skiplist.reverse()
for fn in skiplist:
recipe_parts = os.path.splitext(os.path.basename(fn))[0].split('_')
p = recipe_parts[0]
if len(recipe_parts) > 1:
ver = (None, recipe_parts[1], None)
else:
ver = (None, 'unknown', None)
allproviders[p].append((ver, fn))
if not p in pkg_pn:
pkg_pn[p] = 'dummy'
preferred_versions[p] = (ver, fn)
def print_item(f, pn, ver, layer, ispref):
if f in skiplist:
skipped = ' (skipped)'
else:
skipped = ''
if show_filenames:
if ispref:
logger.plain("%s%s", f, skipped)
else:
logger.plain(" %s%s", f, skipped)
else:
if ispref:
logger.plain("%s:", pn)
logger.plain(" %s %s%s", layer.ljust(20), ver, skipped)
preffiles = []
items_listed = False
for p in sorted(pkg_pn):
if pnspec:
if not fnmatch.fnmatch(p, pnspec):
continue
if len(allproviders[p]) > 1 or not show_multi_provider_only:
pref = preferred_versions[p]
preffile = bb.cache.Cache.virtualfn2realfn(pref[1])[0]
if preffile not in preffiles:
preflayer = self.get_file_layer(preffile)
multilayer = False
same_ver = True
provs = []
for prov in allproviders[p]:
provfile = bb.cache.Cache.virtualfn2realfn(prov[1])[0]
provlayer = self.get_file_layer(provfile)
provs.append((provfile, provlayer, prov[0]))
if provlayer != preflayer:
multilayer = True
if prov[0] != pref[0]:
same_ver = False
if (multilayer or not show_overlayed_only) and (same_ver or not show_same_ver_only):
if not items_listed:
logger.plain('=== %s ===' % title)
items_listed = True
print_item(preffile, p, self.version_str(pref[0][0], pref[0][1]), preflayer, True)
for (provfile, provlayer, provver) in provs:
if provfile != preffile:
print_item(provfile, p, self.version_str(provver[0], provver[1]), provlayer, False)
# Ensure we don't show two entries for BBCLASSEXTENDed recipes
preffiles.append(preffile)
return items_listed
def do_flatten(self, args):
"""flattens layer configuration into a separate output directory.
usage: flatten [layer1 layer2 [layer3]...] <outputdir>
Takes the specified layers (or all layers in the current layer
configuration if none are specified) and builds a "flattened" directory
containing the contents of all layers, with any overlayed recipes removed
and bbappends appended to the corresponding recipes. Note that some manual
cleanup may still be necessary afterwards, in particular:
* where non-recipe files (such as patches) are overwritten (the flatten
command will show a warning for these)
* where anything beyond the normal layer setup has been added to
layer.conf (only the lowest priority number layer's layer.conf is used)
* overridden/appended items from bbappends will need to be tidied up
* when the flattened layers do not have the same directory structure (the
flatten command should show a warning when this will cause a problem)
Warning: if you flatten several layers where another layer is intended to
be used "inbetween" them (in layer priority order) such that recipes /
bbappends in the layers interact, and then attempt to use the new output
layer together with that other layer, you may no longer get the same
build results (as the layer priority order has effectively changed).
"""
arglist = args.split()
if len(arglist) < 1:
logger.error('Please specify an output directory')
self.do_help('flatten')
return
if len(arglist) == 2:
logger.error('If you specify layers to flatten you must specify at least two')
self.do_help('flatten')
return
outputdir = arglist[-1]
if os.path.exists(outputdir) and os.listdir(outputdir):
logger.error('Directory %s exists and is non-empty, please clear it out first' % outputdir)
return
self.init_bbhandler()
layers = self.bblayers
if len(arglist) > 2:
layernames = arglist[:-1]
found_layernames = []
found_layerdirs = []
for layerdir in layers:
layername = self.get_layer_name(layerdir)
if layername in layernames:
found_layerdirs.append(layerdir)
found_layernames.append(layername)
for layername in layernames:
if not layername in found_layernames:
logger.error('Unable to find layer %s in current configuration, please run "%s show-layers" to list configured layers' % (layername, os.path.basename(sys.argv[0])))
return
layers = found_layerdirs
else:
layernames = []
# Ensure a specified path matches our list of layers
def layer_path_match(path):
for layerdir in layers:
if path.startswith(os.path.join(layerdir, '')):
return layerdir
return None
appended_recipes = []
for layer in layers:
overlayed = []
for f in self.bbhandler.cooker.collection.overlayed.iterkeys():
for of in self.bbhandler.cooker.collection.overlayed[f]:
if of.startswith(layer):
overlayed.append(of)
logger.plain('Copying files from %s...' % layer )
for root, dirs, files in os.walk(layer):
for f1 in files:
f1full = os.sep.join([root, f1])
if f1full in overlayed:
logger.plain(' Skipping overlayed file %s' % f1full )
else:
ext = os.path.splitext(f1)[1]
if ext != '.bbappend':
fdest = f1full[len(layer):]
fdest = os.path.normpath(os.sep.join([outputdir,fdest]))
bb.utils.mkdirhier(os.path.dirname(fdest))
if os.path.exists(fdest):
if f1 == 'layer.conf' and root.endswith('/conf'):
logger.plain(' Skipping layer config file %s' % f1full )
continue
else:
logger.warn('Overwriting file %s', fdest)
bb.utils.copyfile(f1full, fdest)
if ext == '.bb':
if f1 in self.bbhandler.cooker.collection.appendlist:
appends = self.bbhandler.cooker.collection.appendlist[f1]
if appends:
logger.plain(' Applying appends to %s' % fdest )
for appendname in appends:
if layer_path_match(appendname):
self.apply_append(appendname, fdest)
appended_recipes.append(f1)
# Take care of when some layers are excluded and yet we have included bbappends for those recipes
for recipename in self.bbhandler.cooker.collection.appendlist.iterkeys():
if recipename not in appended_recipes:
appends = self.bbhandler.cooker.collection.appendlist[recipename]
first_append = None
for appendname in appends:
layer = layer_path_match(appendname)
if layer:
if first_append:
self.apply_append(appendname, first_append)
else:
fdest = appendname[len(layer):]
fdest = os.path.normpath(os.sep.join([outputdir,fdest]))
bb.utils.mkdirhier(os.path.dirname(fdest))
bb.utils.copyfile(appendname, fdest)
first_append = fdest
# Get the regex for the first layer in our list (which is where the conf/layer.conf file will
# have come from)
first_regex = None
layerdir = layers[0]
for layername, pattern, regex, _ in self.bbhandler.cooker.recipecache.bbfile_config_priorities:
if regex.match(os.path.join(layerdir, 'test')):
first_regex = regex
break
if first_regex:
# Find the BBFILES entries that match (which will have come from this conf/layer.conf file)
bbfiles = str(self.bbhandler.config_data.getVar('BBFILES', True)).split()
bbfiles_layer = []
for item in bbfiles:
if first_regex.match(item):
newpath = os.path.join(outputdir, item[len(layerdir)+1:])
bbfiles_layer.append(newpath)
if bbfiles_layer:
# Check that all important layer files match BBFILES
for root, dirs, files in os.walk(outputdir):
for f1 in files:
ext = os.path.splitext(f1)[1]
if ext in ['.bb', '.bbappend']:
f1full = os.sep.join([root, f1])
entry_found = False
for item in bbfiles_layer:
if fnmatch.fnmatch(f1full, item):
entry_found = True
break
if not entry_found:
logger.warning("File %s does not match the flattened layer's BBFILES setting, you may need to edit conf/layer.conf or move the file elsewhere" % f1full)
def get_file_layer(self, filename):
for layer, _, regex, _ in self.bbhandler.cooker.recipecache.bbfile_config_priorities:
if regex.match(filename):
for layerdir in self.bblayers:
if regex.match(os.path.join(layerdir, 'test')) and re.match(layerdir, filename):
return self.get_layer_name(layerdir)
return "?"
def get_file_layerdir(self, filename):
for layer, _, regex, _ in self.bbhandler.cooker.recipecache.bbfile_config_priorities:
if regex.match(filename):
for layerdir in self.bblayers:
if regex.match(os.path.join(layerdir, 'test')) and re.match(layerdir, filename):
return layerdir
return "?"
def remove_layer_prefix(self, f):
"""Remove the layer_dir prefix, e.g., f = /path/to/layer_dir/foo/blah, the
return value will be: layer_dir/foo/blah"""
f_layerdir = self.get_file_layerdir(f)
prefix = os.path.join(os.path.dirname(f_layerdir), '')
return f[len(prefix):] if f.startswith(prefix) else f
def get_layer_name(self, layerdir):
return os.path.basename(layerdir.rstrip(os.sep))
def apply_append(self, appendname, recipename):
appendfile = open(appendname, 'r')
recipefile = open(recipename, 'a')
recipefile.write('\n')
recipefile.write('##### bbappended from %s #####\n' % self.get_file_layer(appendname))
recipefile.writelines(appendfile.readlines())
recipefile.close()
appendfile.close()
def do_show_appends(self, args):
"""list bbappend files and recipe files they apply to
usage: show-appends
Recipes are listed with the bbappends that apply to them as subitems.
"""
self.init_bbhandler()
if not self.bbhandler.cooker.collection.appendlist:
logger.plain('No append files found')
return
logger.plain('=== Appended recipes ===')
pnlist = list(self.bbhandler.cooker_data.pkg_pn.keys())
pnlist.sort()
for pn in pnlist:
self.show_appends_for_pn(pn)
self.show_appends_for_skipped()
def show_appends_for_pn(self, pn):
filenames = self.bbhandler.cooker_data.pkg_pn[pn]
best = bb.providers.findBestProvider(pn,
self.bbhandler.config_data,
self.bbhandler.cooker_data,
self.bbhandler.cooker_data.pkg_pn)
best_filename = os.path.basename(best[3])
self.show_appends_output(filenames, best_filename)
def show_appends_for_skipped(self):
filenames = [os.path.basename(f)
for f in self.bbhandler.cooker.skiplist.iterkeys()]
self.show_appends_output(filenames, None, " (skipped)")
def show_appends_output(self, filenames, best_filename, name_suffix = ''):
appended, missing = self.get_appends_for_files(filenames)
if appended:
for basename, appends in appended:
logger.plain('%s%s:', basename, name_suffix)
for append in appends:
logger.plain(' %s', append)
if best_filename:
if best_filename in missing:
logger.warn('%s: missing append for preferred version',
best_filename)
self.returncode |= 1
def get_appends_for_files(self, filenames):
appended, notappended = [], []
for filename in filenames:
_, cls = bb.cache.Cache.virtualfn2realfn(filename)
if cls:
continue
basename = os.path.basename(filename)
appends = self.bbhandler.cooker.collection.appendlist.get(basename)
if appends:
appended.append((basename, list(appends)))
else:
notappended.append(basename)
return appended, notappended
def do_show_cross_depends(self, args):
"""figure out the dependency between recipes that crosses a layer boundary.
usage: show-cross-depends [-f]
Figure out the dependency between recipes that crosses a layer boundary.
Options:
-f show full file path
NOTE:
The .bbappend file can impact the dependency.
"""
self.init_bbhandler()
show_filenames = False
for arg in args.split():
if arg == '-f':
show_filenames = True
else:
sys.stderr.write("show-cross-depends: invalid option %s\n" % arg)
self.do_help('')
return
pkg_fn = self.bbhandler.cooker_data.pkg_fn
bbpath = str(self.bbhandler.config_data.getVar('BBPATH', True))
self.require_re = re.compile(r"require\s+(.+)")
self.include_re = re.compile(r"include\s+(.+)")
self.inherit_re = re.compile(r"inherit\s+(.+)")
# The bb's DEPENDS and RDEPENDS
for f in pkg_fn:
f = bb.cache.Cache.virtualfn2realfn(f)[0]
# Get the layername that the file is in
layername = self.get_file_layer(f)
# The DEPENDS
deps = self.bbhandler.cooker_data.deps[f]
for pn in deps:
if pn in self.bbhandler.cooker_data.pkg_pn:
best = bb.providers.findBestProvider(pn,
self.bbhandler.config_data,
self.bbhandler.cooker_data,
self.bbhandler.cooker_data.pkg_pn)
self.check_cross_depends("DEPENDS", layername, f, best[3], show_filenames)
# The RDPENDS
all_rdeps = self.bbhandler.cooker_data.rundeps[f].values()
# Remove the duplicated or null one.
sorted_rdeps = {}
# The all_rdeps is the list in list, so we need two for loops
for k1 in all_rdeps:
for k2 in k1:
sorted_rdeps[k2] = 1
all_rdeps = sorted_rdeps.keys()
for rdep in all_rdeps:
all_p = bb.providers.getRuntimeProviders(self.bbhandler.cooker_data, rdep)
if all_p:
best = bb.providers.filterProvidersRunTime(all_p, rdep,
self.bbhandler.config_data,
self.bbhandler.cooker_data)[0][0]
self.check_cross_depends("RDEPENDS", layername, f, best, show_filenames)
# The inherit class
cls_re = re.compile('classes/')
if f in self.bbhandler.cooker_data.inherits:
inherits = self.bbhandler.cooker_data.inherits[f]
for cls in inherits:
# The inherits' format is [classes/cls, /path/to/classes/cls]
# ignore the classes/cls.
if not cls_re.match(cls):
inherit_layername = self.get_file_layer(cls)
if inherit_layername != layername:
if not show_filenames:
f_short = self.remove_layer_prefix(f)
cls = self.remove_layer_prefix(cls)
else:
f_short = f
logger.plain("%s inherits %s" % (f_short, cls))
# The 'require/include xxx' in the bb file
pv_re = re.compile(r"\${PV}")
fnfile = open(f, 'r')
line = fnfile.readline()
while line:
m, keyword = self.match_require_include(line)
# Found the 'require/include xxxx'
if m:
needed_file = m.group(1)
# Replace the ${PV} with the real PV
if pv_re.search(needed_file) and f in self.bbhandler.cooker_data.pkg_pepvpr:
pv = self.bbhandler.cooker_data.pkg_pepvpr[f][1]
needed_file = re.sub(r"\${PV}", pv, needed_file)
self.print_cross_files(bbpath, keyword, layername, f, needed_file, show_filenames)
line = fnfile.readline()
fnfile.close()
# The "require/include xxx" in conf/machine/*.conf, .inc and .bbclass
conf_re = re.compile(".*/conf/machine/[^\/]*\.conf$")
inc_re = re.compile(".*\.inc$")
# The "inherit xxx" in .bbclass
bbclass_re = re.compile(".*\.bbclass$")
for layerdir in self.bblayers:
layername = self.get_layer_name(layerdir)
for dirpath, dirnames, filenames in os.walk(layerdir):
for name in filenames:
f = os.path.join(dirpath, name)
s = conf_re.match(f) or inc_re.match(f) or bbclass_re.match(f)
if s:
ffile = open(f, 'r')
line = ffile.readline()
while line:
m, keyword = self.match_require_include(line)
# Only bbclass has the "inherit xxx" here.
bbclass=""
if not m and f.endswith(".bbclass"):
m, keyword = self.match_inherit(line)
bbclass=".bbclass"
# Find a 'require/include xxxx'
if m:
self.print_cross_files(bbpath, keyword, layername, f, m.group(1) + bbclass, show_filenames)
line = ffile.readline()
ffile.close()
def print_cross_files(self, bbpath, keyword, layername, f, needed_filename, show_filenames):
"""Print the depends that crosses a layer boundary"""
needed_file = bb.utils.which(bbpath, needed_filename)
if needed_file:
# Which layer is this file from
needed_layername = self.get_file_layer(needed_file)
if needed_layername != layername:
if not show_filenames:
f = self.remove_layer_prefix(f)
needed_file = self.remove_layer_prefix(needed_file)
logger.plain("%s %s %s" %(f, keyword, needed_file))
def match_inherit(self, line):
"""Match the inherit xxx line"""
return (self.inherit_re.match(line), "inherits")
def match_require_include(self, line):
"""Match the require/include xxx line"""
m = self.require_re.match(line)
keyword = "requires"
if not m:
m = self.include_re.match(line)
keyword = "includes"
return (m, keyword)
def check_cross_depends(self, keyword, layername, f, needed_file, show_filenames):
"""Print the DEPENDS/RDEPENDS file that crosses a layer boundary"""
best_realfn = bb.cache.Cache.virtualfn2realfn(needed_file)[0]
needed_layername = self.get_file_layer(best_realfn)
if needed_layername != layername:
if not show_filenames:
f = self.remove_layer_prefix(f)
best_realfn = self.remove_layer_prefix(best_realfn)
logger.plain("%s %s %s" % (f, keyword, best_realfn))
if __name__ == '__main__':
sys.exit(main(sys.argv[1:]) or 0)

View File

@@ -1,55 +0,0 @@
#!/usr/bin/env python
import os
import sys,logging
import optparse
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(__file__)),'lib'))
import prserv
import prserv.serv
__version__="1.0.0"
PRHOST_DEFAULT='0.0.0.0'
PRPORT_DEFAULT=8585
def main():
parser = optparse.OptionParser(
version="Bitbake PR Service Core version %s, %%prog version %s" % (prserv.__version__, __version__),
usage = "%prog < --start | --stop > [options]")
parser.add_option("-f", "--file", help="database filename(default: prserv.sqlite3)", action="store",
dest="dbfile", type="string", default="prserv.sqlite3")
parser.add_option("-l", "--log", help="log filename(default: prserv.log)", action="store",
dest="logfile", type="string", default="prserv.log")
parser.add_option("--loglevel", help="logging level, i.e. CRITICAL, ERROR, WARNING, INFO, DEBUG",
action = "store", type="string", dest="loglevel", default = "INFO")
parser.add_option("--start", help="start daemon",
action="store_true", dest="start")
parser.add_option("--stop", help="stop daemon",
action="store_true", dest="stop")
parser.add_option("--host", help="ip address to bind", action="store",
dest="host", type="string", default=PRHOST_DEFAULT)
parser.add_option("--port", help="port number(default: 8585)", action="store",
dest="port", type="int", default=PRPORT_DEFAULT)
options, args = parser.parse_args(sys.argv)
prserv.init_logger(os.path.abspath(options.logfile),options.loglevel)
if options.start:
ret=prserv.serv.start_daemon(options.dbfile, options.host, options.port,os.path.abspath(options.logfile))
elif options.stop:
ret=prserv.serv.stop_daemon(options.host, options.port)
else:
ret=parser.print_help()
return ret
if __name__ == "__main__":
try:
ret = main()
except Exception:
ret = 1
import traceback
traceback.print_exc(5)
sys.exit(ret)

View File

@@ -1,49 +0,0 @@
#!/usr/bin/env python
#
# Copyright (C) 2012 Richard Purdie
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
import os
import sys, logging
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(__file__)), 'lib'))
import unittest
try:
import bb
except RuntimeError as exc:
sys.exit(str(exc))
def usage():
print('usage: %s [testname1 [testname2]...]' % os.path.basename(sys.argv[0]))
if len(sys.argv) > 1:
if '--help' in sys.argv[1:]:
usage()
sys.exit(0)
tests = sys.argv[1:]
else:
tests = ["bb.tests.codeparser",
"bb.tests.cow",
"bb.tests.data",
"bb.tests.fetch",
"bb.tests.utils"]
for t in tests:
t = '.'.join(t.split('.')[:3])
__import__(t)
unittest.main(argv=["bitbake-selftest"] + tests)

View File

@@ -1,375 +0,0 @@
#!/usr/bin/env python
import os
import sys
import warnings
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
from bb import fetch2
import logging
import bb
import select
import errno
import signal
# Users shouldn't be running this code directly
if len(sys.argv) != 2 or sys.argv[1] != "decafbad":
print("bitbake-worker is meant for internal execution by bitbake itself, please don't use it standalone.")
sys.exit(1)
logger = logging.getLogger("BitBake")
try:
import cPickle as pickle
except ImportError:
import pickle
bb.msg.note(1, bb.msg.domain.Cache, "Importing cPickle failed. Falling back to a very slow implementation.")
worker_pipe = sys.stdout.fileno()
bb.utils.nonblockingfd(worker_pipe)
handler = bb.event.LogHandler()
logger.addHandler(handler)
if 0:
# Code to write out a log file of all events passing through the worker
logfilename = "/tmp/workerlogfile"
format_str = "%(levelname)s: %(message)s"
conlogformat = bb.msg.BBLogFormatter(format_str)
consolelog = logging.FileHandler(logfilename)
bb.msg.addDefaultlogFilter(consolelog)
consolelog.setFormatter(conlogformat)
logger.addHandler(consolelog)
worker_queue = ""
def worker_fire(event, d):
data = "<event>" + pickle.dumps(event) + "</event>"
worker_fire_prepickled(data)
def worker_fire_prepickled(event):
global worker_queue
worker_queue = worker_queue + event
worker_flush()
def worker_flush():
global worker_queue, worker_pipe
if not worker_queue:
return
try:
written = os.write(worker_pipe, worker_queue)
worker_queue = worker_queue[written:]
except (IOError, OSError) as e:
if e.errno != errno.EAGAIN:
raise
def worker_child_fire(event, d):
global worker_pipe
data = "<event>" + pickle.dumps(event) + "</event>"
worker_pipe.write(data)
bb.event.worker_fire = worker_fire
lf = None
#lf = open("/tmp/workercommandlog", "w+")
def workerlog_write(msg):
if lf:
lf.write(msg)
lf.flush()
def fork_off_task(cfg, data, workerdata, fn, task, taskname, appends, taskdepdata, quieterrors=False):
# We need to setup the environment BEFORE the fork, since
# a fork() or exec*() activates PSEUDO...
envbackup = {}
fakeenv = {}
umask = None
taskdep = workerdata["taskdeps"][fn]
if 'umask' in taskdep and taskname in taskdep['umask']:
# umask might come in as a number or text string..
try:
umask = int(taskdep['umask'][taskname],8)
except TypeError:
umask = taskdep['umask'][taskname]
# We can't use the fakeroot environment in a dry run as it possibly hasn't been built
if 'fakeroot' in taskdep and taskname in taskdep['fakeroot'] and not cfg.dry_run:
envvars = (workerdata["fakerootenv"][fn] or "").split()
for key, value in (var.split('=') for var in envvars):
envbackup[key] = os.environ.get(key)
os.environ[key] = value
fakeenv[key] = value
fakedirs = (workerdata["fakerootdirs"][fn] or "").split()
for p in fakedirs:
bb.utils.mkdirhier(p)
logger.debug(2, 'Running %s:%s under fakeroot, fakedirs: %s' %
(fn, taskname, ', '.join(fakedirs)))
else:
envvars = (workerdata["fakerootnoenv"][fn] or "").split()
for key, value in (var.split('=') for var in envvars):
envbackup[key] = os.environ.get(key)
os.environ[key] = value
fakeenv[key] = value
sys.stdout.flush()
sys.stderr.flush()
try:
pipein, pipeout = os.pipe()
pipein = os.fdopen(pipein, 'rb', 4096)
pipeout = os.fdopen(pipeout, 'wb', 0)
pid = os.fork()
except OSError as e:
bb.msg.fatal("RunQueue", "fork failed: %d (%s)" % (e.errno, e.strerror))
if pid == 0:
global worker_pipe
pipein.close()
signal.signal(signal.SIGTERM, signal.SIG_DFL)
# Save out the PID so that the event can include it the
# events
bb.event.worker_pid = os.getpid()
bb.event.worker_fire = worker_child_fire
worker_pipe = pipeout
# Make the child the process group leader
os.setpgid(0, 0)
# No stdin
newsi = os.open(os.devnull, os.O_RDWR)
os.dup2(newsi, sys.stdin.fileno())
if umask:
os.umask(umask)
data.setVar("BB_WORKERCONTEXT", "1")
data.setVar("BB_TASKDEPDATA", taskdepdata)
data.setVar("BUILDNAME", workerdata["buildname"])
data.setVar("DATE", workerdata["date"])
data.setVar("TIME", workerdata["time"])
bb.parse.siggen.set_taskdata(workerdata["hashes"], workerdata["hash_deps"], workerdata["sigchecksums"])
ret = 0
try:
the_data = bb.cache.Cache.loadDataFull(fn, appends, data)
the_data.setVar('BB_TASKHASH', workerdata["runq_hash"][task])
for h in workerdata["hashes"]:
the_data.setVar("BBHASH_%s" % h, workerdata["hashes"][h])
for h in workerdata["hash_deps"]:
the_data.setVar("BBHASHDEPS_%s" % h, workerdata["hash_deps"][h])
# exported_vars() returns a generator which *cannot* be passed to os.environ.update()
# successfully. We also need to unset anything from the environment which shouldn't be there
exports = bb.data.exported_vars(the_data)
bb.utils.empty_environment()
for e, v in exports:
os.environ[e] = v
for e in fakeenv:
os.environ[e] = fakeenv[e]
the_data.setVar(e, fakeenv[e])
the_data.setVarFlag(e, 'export', "1")
if quieterrors:
the_data.setVarFlag(taskname, "quieterrors", "1")
except Exception as exc:
if not quieterrors:
logger.critical(str(exc))
os._exit(1)
try:
if not cfg.dry_run:
ret = bb.build.exec_task(fn, taskname, the_data, cfg.profile)
os._exit(ret)
except:
os._exit(1)
else:
for key, value in envbackup.iteritems():
if value is None:
del os.environ[key]
else:
os.environ[key] = value
return pid, pipein, pipeout
class runQueueWorkerPipe():
"""
Abstraction for a pipe between a worker thread and the worker server
"""
def __init__(self, pipein, pipeout):
self.input = pipein
if pipeout:
pipeout.close()
bb.utils.nonblockingfd(self.input)
self.queue = ""
def read(self):
start = len(self.queue)
try:
self.queue = self.queue + self.input.read(102400)
except (OSError, IOError) as e:
if e.errno != errno.EAGAIN:
raise
end = len(self.queue)
index = self.queue.find("</event>")
while index != -1:
worker_fire_prepickled(self.queue[:index+8])
self.queue = self.queue[index+8:]
index = self.queue.find("</event>")
return (end > start)
def close(self):
while self.read():
continue
if len(self.queue) > 0:
print("Warning, worker child left partial message: %s" % self.queue)
self.input.close()
normalexit = False
class BitbakeWorker(object):
def __init__(self, din):
self.input = din
bb.utils.nonblockingfd(self.input)
self.queue = ""
self.cookercfg = None
self.databuilder = None
self.data = None
self.build_pids = {}
self.build_pipes = {}
signal.signal(signal.SIGTERM, self.sigterm_exception)
def sigterm_exception(self, signum, stackframe):
bb.warn("Worker recieved SIGTERM, shutting down...")
self.handle_finishnow(None)
signal.signal(signal.SIGTERM, signal.SIG_DFL)
os.kill(os.getpid(), signal.SIGTERM)
def serve(self):
while True:
(ready, _, _) = select.select([self.input] + [i.input for i in self.build_pipes.values()], [] , [], 1)
if self.input in ready or len(self.queue):
start = len(self.queue)
try:
self.queue = self.queue + self.input.read()
except (OSError, IOError):
pass
end = len(self.queue)
self.handle_item("cookerconfig", self.handle_cookercfg)
self.handle_item("workerdata", self.handle_workerdata)
self.handle_item("runtask", self.handle_runtask)
self.handle_item("finishnow", self.handle_finishnow)
self.handle_item("ping", self.handle_ping)
self.handle_item("quit", self.handle_quit)
for pipe in self.build_pipes:
self.build_pipes[pipe].read()
if len(self.build_pids):
self.process_waitpid()
worker_flush()
def handle_item(self, item, func):
if self.queue.startswith("<" + item + ">"):
index = self.queue.find("</" + item + ">")
while index != -1:
func(self.queue[(len(item) + 2):index])
self.queue = self.queue[(index + len(item) + 3):]
index = self.queue.find("</" + item + ">")
def handle_cookercfg(self, data):
self.cookercfg = pickle.loads(data)
self.databuilder = bb.cookerdata.CookerDataBuilder(self.cookercfg, worker=True)
self.databuilder.parseBaseConfiguration()
self.data = self.databuilder.data
def handle_workerdata(self, data):
self.workerdata = pickle.loads(data)
bb.msg.loggerDefaultDebugLevel = self.workerdata["logdefaultdebug"]
bb.msg.loggerDefaultVerbose = self.workerdata["logdefaultverbose"]
bb.msg.loggerVerboseLogs = self.workerdata["logdefaultverboselogs"]
bb.msg.loggerDefaultDomains = self.workerdata["logdefaultdomain"]
self.data.setVar("PRSERV_HOST", self.workerdata["prhost"])
def handle_ping(self, _):
workerlog_write("Handling ping\n")
logger.warn("Pong from bitbake-worker!")
def handle_quit(self, data):
workerlog_write("Handling quit\n")
global normalexit
normalexit = True
sys.exit(0)
def handle_runtask(self, data):
fn, task, taskname, quieterrors, appends, taskdepdata = pickle.loads(data)
workerlog_write("Handling runtask %s %s %s\n" % (task, fn, taskname))
pid, pipein, pipeout = fork_off_task(self.cookercfg, self.data, self.workerdata, fn, task, taskname, appends, taskdepdata, quieterrors)
self.build_pids[pid] = task
self.build_pipes[pid] = runQueueWorkerPipe(pipein, pipeout)
def process_waitpid(self):
"""
Return none is there are no processes awaiting result collection, otherwise
collect the process exit codes and close the information pipe.
"""
try:
pid, status = os.waitpid(-1, os.WNOHANG)
if pid == 0 or os.WIFSTOPPED(status):
return None
except OSError:
return None
workerlog_write("Exit code of %s for pid %s\n" % (status, pid))
if os.WIFEXITED(status):
status = os.WEXITSTATUS(status)
elif os.WIFSIGNALED(status):
# Per shell conventions for $?, when a process exits due to
# a signal, we return an exit code of 128 + SIGNUM
status = 128 + os.WTERMSIG(status)
task = self.build_pids[pid]
del self.build_pids[pid]
self.build_pipes[pid].close()
del self.build_pipes[pid]
worker_fire_prepickled("<exitcode>" + pickle.dumps((task, status)) + "</exitcode>")
def handle_finishnow(self, _):
if self.build_pids:
logger.info("Sending SIGTERM to remaining %s tasks", len(self.build_pids))
for k, v in self.build_pids.iteritems():
try:
os.kill(-k, signal.SIGTERM)
os.waitpid(-1, 0)
except:
pass
for pipe in self.build_pipes:
self.build_pipes[pipe].read()
try:
worker = BitbakeWorker(sys.stdin)
worker.serve()
except BaseException as e:
if not normalexit:
import traceback
sys.stderr.write(traceback.format_exc())
sys.stderr.write(str(e))
while len(worker_queue):
worker_flush()
workerlog_write("exitting")
sys.exit(0)

View File

@@ -20,7 +20,7 @@
import optparse, os, sys
# bitbake
sys.path.append(os.path.join(os.path.dirname(os.path.dirname(__file__), 'lib'))
sys.path.append(os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
import bb
import bb.parse
from string import split, join
@@ -48,7 +48,7 @@ class HTMLFormatter:
From pydoc... almost identical at least
"""
while pairs:
(a, b) = pairs[0]
(a,b) = pairs[0]
text = join(split(text, a), b)
pairs = pairs[1:]
return text
@@ -87,7 +87,7 @@ class HTMLFormatter:
return txt + ",".join(txts)
def groups(self, item):
def groups(self,item):
"""
Create HTML to link to related groups
"""
@@ -99,12 +99,12 @@ class HTMLFormatter:
txt = "<p><b>See also:</b><br>"
txts = []
for group in item.groups():
txts.append( """<a href="group%s.html">%s</a> """ % (group, group) )
txts.append( """<a href="group%s.html">%s</a> """ % (group,group) )
return txt + ",".join(txts)
def createKeySite(self, item):
def createKeySite(self,item):
"""
Create a site for a key. It contains the header/navigator, a heading,
the description, links to related keys and to the groups.
@@ -149,7 +149,8 @@ class HTMLFormatter:
"""
groups = ""
sorted_groups = sorted(doc.groups())
sorted_groups = doc.groups()
sorted_groups.sort()
for group in sorted_groups:
groups += """<a href="group%s.html">%s</a><br>""" % (group, group)
@@ -184,7 +185,8 @@ class HTMLFormatter:
Create Overview of all avilable keys
"""
keys = ""
sorted_keys = sorted(doc.doc_keys())
sorted_keys = doc.doc_keys()
sorted_keys.sort()
for key in sorted_keys:
keys += """<a href="key%s.html">%s</a><br>""" % (key, key)
@@ -212,7 +214,7 @@ class HTMLFormatter:
description += "<h2 Description of Grozp %s</h2>" % gr
description += _description
items.sort(lambda x, y:cmp(x.name(), y.name()))
items.sort(lambda x,y:cmp(x.name(),y.name()))
for group in items:
groups += """<a href="key%s.html">%s</a><br>""" % (group.name(), group.name())
@@ -341,7 +343,7 @@ class DocumentationItem:
def addGroup(self, group):
self._groups.append(group)
def addRelation(self, relation):
def addRelation(self,relation):
self._related.append(relation)
def sort(self):
@@ -394,7 +396,7 @@ class Documentation:
"""
return self.__groups.keys()
def group_content(self, group_name):
def group_content(self,group_name):
"""
Return a list of keys/names that are in a specefic
group or the empty list
@@ -410,7 +412,7 @@ def parse_cmdline(args):
Parse the CMD line and return the result as a n-tuple
"""
parser = optparse.OptionParser( version = "Bitbake Documentation Tool Core version %s, %%prog version %s" % (bb.__version__, __version__))
parser = optparse.OptionParser( version = "Bitbake Documentation Tool Core version %s, %%prog version %s" % (bb.__version__,__version__))
usage = """%prog [options]
Create a set of html pages (documentation) for a bitbake.conf....
@@ -426,12 +428,13 @@ Create a set of html pages (documentation) for a bitbake.conf....
parser.add_option( "-D", "--debug", help = "Increase the debug level",
action = "count", dest = "debug", default = 0 )
parser.add_option( "-v", "--verbose", help = "output more chit-char to the terminal",
parser.add_option( "-v","--verbose", help = "output more chit-char to the terminal",
action = "store_true", dest = "verbose", default = False )
options, args = parser.parse_args( sys.argv )
bb.msg.init_msgconfig(options.verbose, options.debug)
if options.debug:
bb.msg.set_debug_level(options.debug)
return options.config, options.output
@@ -440,7 +443,7 @@ def main():
The main Method
"""
(config_file, output_dir) = parse_cmdline( sys.argv )
(config_file,output_dir) = parse_cmdline( sys.argv )
# right to let us load the file now
try:
@@ -450,8 +453,6 @@ def main():
except bb.parse.ParseError:
bb.fatal( "Unable to parse %s" % config_file )
if isinstance(documentation, dict):
documentation = documentation[""]
# Assuming we've the file loaded now, we will initialize the 'tree'
doc = Documentation()
@@ -462,7 +463,7 @@ def main():
state_group = 2
for key in bb.data.keys(documentation):
data = documentation.getVarFlag(key, "doc")
data = bb.data.getVarFlag(key, "doc", documentation)
if not data:
continue
@@ -496,7 +497,7 @@ def main():
doc.insert_doc_item(doc_ins)
# let us create the HTML now
bb.utils.mkdirhier(output_dir)
bb.mkdirhier(output_dir)
os.chdir(output_dir)
# Let us create the sites now. We do it in the following order

View File

@@ -1,122 +0,0 @@
#!/usr/bin/env python
# Copyright (c) 2012 Wind River Systems, Inc.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# See the GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
import os
import sys
sys.path.insert(0, os.path.join(os.path.dirname(os.path.dirname( \
os.path.abspath(__file__))), 'lib'))
try:
import bb
except RuntimeError as exc:
sys.exit(str(exc))
import gtk
import optparse
import pygtk
from bb.ui.crumbs.hobwidget import HobAltButton, HobButton
from bb.ui.crumbs.hig.crumbsmessagedialog import CrumbsMessageDialog
from bb.ui.crumbs.hig.deployimagedialog import DeployImageDialog
from bb.ui.crumbs.hig.imageselectiondialog import ImageSelectionDialog
# I put all the fs bitbake supported here. Need more test.
DEPLOYABLE_IMAGE_TYPES = ["jffs2", "cramfs", "ext2", "ext3", "btrfs", "squashfs", "ubi", "vmdk"]
Title = "USB Image Writer"
class DeployWindow(gtk.Window):
def __init__(self, image_path=''):
super(DeployWindow, self).__init__()
if len(image_path) > 0:
valid = True
if not os.path.exists(image_path):
valid = False
lbl = "<b>Invalid image file path: %s.</b>\nPress <b>Select Image</b> to select an image." % image_path
else:
image_path = os.path.abspath(image_path)
extend_name = os.path.splitext(image_path)[1][1:]
if extend_name not in DEPLOYABLE_IMAGE_TYPES:
valid = False
lbl = "<b>Undeployable imge type: %s</b>\nPress <b>Select Image</b> to select an image." % extend_name
if not valid:
image_path = ''
crumbs_dialog = CrumbsMessageDialog(self, lbl, gtk.STOCK_DIALOG_INFO)
button = crumbs_dialog.add_button("Close", gtk.RESPONSE_OK)
HobButton.style_button(button)
crumbs_dialog.run()
crumbs_dialog.destroy()
self.deploy_dialog = DeployImageDialog(Title, image_path, self,
gtk.DIALOG_MODAL | gtk.DIALOG_DESTROY_WITH_PARENT
| gtk.DIALOG_NO_SEPARATOR, None, standalone=True)
close_button = self.deploy_dialog.add_button("Close", gtk.RESPONSE_NO)
HobAltButton.style_button(close_button)
close_button.connect('clicked', gtk.main_quit)
write_button = self.deploy_dialog.add_button("Write USB image", gtk.RESPONSE_YES)
HobAltButton.style_button(write_button)
self.deploy_dialog.connect('select_image_clicked', self.select_image_clicked_cb)
self.deploy_dialog.connect('destroy', gtk.main_quit)
response = self.deploy_dialog.show()
def select_image_clicked_cb(self, dialog):
cwd = os.getcwd()
dialog = ImageSelectionDialog(cwd, DEPLOYABLE_IMAGE_TYPES, Title, self, gtk.FILE_CHOOSER_ACTION_SAVE )
button = dialog.add_button("Cancel", gtk.RESPONSE_NO)
HobAltButton.style_button(button)
button = dialog.add_button("Open", gtk.RESPONSE_YES)
HobAltButton.style_button(button)
response = dialog.run()
if response == gtk.RESPONSE_YES:
if not dialog.image_names:
lbl = "<b>No selections made</b>\nClicked the radio button to select a image."
crumbs_dialog = CrumbsMessageDialog(self, lbl, gtk.STOCK_DIALOG_INFO)
button = crumbs_dialog.add_button("Close", gtk.RESPONSE_OK)
HobButton.style_button(button)
crumbs_dialog.run()
crumbs_dialog.destroy()
dialog.destroy()
return
# get the full path of image
image_path = os.path.join(dialog.image_folder, dialog.image_names[0])
self.deploy_dialog.set_image_text_buffer(image_path)
self.deploy_dialog.set_image_path(image_path)
dialog.destroy()
def main():
parser = optparse.OptionParser(
usage = """%prog [-h] [image_file]
%prog writes bootable images to USB devices. You can
provide the image file on the command line or select it using the GUI.""")
options, args = parser.parse_args(sys.argv)
image_file = args[1] if len(args) > 1 else ''
dw = DeployWindow(image_file)
if __name__ == '__main__':
try:
main()
gtk.main()
except Exception:
import traceback
traceback.print_exc(3)

View File

@@ -1,214 +0,0 @@
#!/bin/bash
# (c) 2013 Intel Corp.
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
# This script enables toaster event logging and
# starts bitbake resident server
# use as: source toaster [start|stop]
# Helper function to kill a background toaster development server
function webserverKillAll()
{
local pidfile
for pidfile in ${BUILDDIR}/.toastermain.pid; do
if [ -f ${pidfile} ]; then
while kill -0 $(< ${pidfile}) 2>/dev/null; do
kill -SIGTERM -$(< ${pidfile}) 2>/dev/null
sleep 1;
done;
rm ${pidfile}
fi
done
}
function webserverStartAll()
{
retval=0
python $BBBASEDIR/lib/toaster/manage.py syncdb || retval=1
python $BBBASEDIR/lib/toaster/manage.py migrate orm || retval=2
if [ $retval -eq 1 ]; then
echo "Failed db sync, stopping system start" 1>&2
elif [ $retval -eq 2 ]; then
echo -e "\nError on migration, trying to recover... \n"
python $BBBASEDIR/lib/toaster/manage.py migrate orm 0001_initial --fake
retval=0
python $BBBASEDIR/lib/toaster/manage.py migrate orm || retval=1
fi
if [ $retval -eq 0 ]; then
python $BBBASEDIR/lib/toaster/manage.py runserver 0.0.0.0:8000 </dev/null >${BUILDDIR}/toaster_web.log 2>&1 & echo $! >${BUILDDIR}/.toastermain.pid
sleep 1
if ! cat "${BUILDDIR}/.toastermain.pid" | xargs -I{} kill -0 {} ; then
retval=1
rm "${BUILDDIR}/.toastermain.pid"
fi
fi
return $retval
}
# Helper functions to add a special configuration file
function addtoConfiguration()
{
echo "#Created by toaster start script" > ${BUILDDIR}/conf/$2
echo $1 >> ${BUILDDIR}/conf/$2
}
INSTOPSYSTEM=0
# define the stop command
function stop_system()
{
# prevent reentry
if [ $INSTOPSYSTEM == 1 ]; then return; fi
INSTOPSYSTEM=1
if [ -f ${BUILDDIR}/.toasterui.pid ]; then
kill $(< ${BUILDDIR}/.toasterui.pid ) 2>/dev/null
rm ${BUILDDIR}/.toasterui.pid
fi
BBSERVER=localhost:8200 bitbake -m
unset BBSERVER
webserverKillAll
# force stop any misbehaving bitbake server
lsof bitbake.lock | awk '{print $2}' | grep "[0-9]\+" | xargs -n1 -r kill
trap - SIGHUP
#trap - SIGCHLD
INSTOPSYSTEM=0
}
function check_pidbyfile() {
[ -e $1 ] && kill -0 $(< $1) 2>/dev/null
}
function notify_chldexit() {
if [ $NOTOASTERUI == 0 ]; then
check_pidbyfile ${BUILDDIR}/.toasterui.pid && return
stop_system
fi
}
# We make sure we're running in the current shell and in a good environment
if [ -z "$ZSH_NAME" ] && [ `basename \"$0\"` = `basename \"$BASH_SOURCE\"` ]; then
echo "Error: This script needs to be sourced. Please run as 'source toaster [start|stop]'" 1>&2;
exit 1
fi
if [ -z "$BUILDDIR" ] || [ -z `which bitbake` ]; then
echo "Error: Build environment is not setup or bitbake is not in path." 1>&2;
return 2
fi
BBBASEDIR=`dirname ${BASH_SOURCE}`/..
# Verify prerequisites
if ! echo "import django; print (1,5) == django.VERSION[0:2]" | python 2>/dev/null | grep True >/dev/null; then
echo -e "This program needs Django 1.5. Please install with\n\nsudo pip install django==1.5"
return 2
fi
if ! echo "import south; print [0,8,4] == map(int,south.__version__.split(\".\"))" | python 2>/dev/null | grep True >/dev/null; then
echo -e "This program needs South 0.8.4. Please install with\n\nsudo pip install south==0.8.4"
return 2
fi
# Determine the action. If specified by arguments, fine, if not, toggle it
if [ "x$1" == "xstart" ] || [ "x$1" == "xstop" ]; then
CMD="$1"
else
if [ -z "$BBSERVER" ]; then
CMD="start"
else
CMD="stop"
fi;
fi
NOTOASTERUI=0
for param in $*; do
case $param in
noui )
NOTOASTERUI=1
;;
esac
done
echo "The system will $CMD."
# Make sure it's safe to run by checking bitbake lock
lock=1
if [ -e $BUILDDIR/bitbake.lock ]; then
(flock -n 200 ) 200<$BUILDDIR/bitbake.lock || lock=0
fi
if [ ${CMD} == "start" ] && ( [ $lock -eq 0 ] || [ -e $BUILDDIR/.toastermain.pid ] ); then
echo "Error: bitbake lock state error. File locks show that the system is on." 2>&1
echo "If you see problems, stop and then start the system again." 2>&1
return 3
fi
# Execute the commands
case $CMD in
start )
start_success=1
addtoConfiguration "INHERIT+=\"toaster buildhistory\"" toaster.conf
if ! webserverStartAll; then
echo "Failed ${CMD}."
return 4
fi
unset BBSERVER
bitbake --postread conf/toaster.conf --server-only -t xmlrpc -B localhost:8200
if [ $? -ne 0 ]; then
start_success=0
echo "Bitbake server start failed"
else
export BBSERVER=localhost:8200
if [ $NOTOASTERUI == 0 ]; then # we start the TOASTERUI only if not inhibited
bitbake --observe-only -u toasterui >${BUILDDIR}/toaster_ui.log 2>&1 & echo $! >${BUILDDIR}/.toasterui.pid
fi
fi
if [ $start_success -eq 1 ]; then
# set fail safe stop system on terminal exit
trap stop_system SIGHUP
echo "Successful ${CMD}."
else
# failed start, do stop
stop_system
echo "Failed ${CMD}."
fi
# stop system on terminal exit
set -o monitor
trap stop_system SIGHUP
#trap notify_chldexit SIGCHLD
;;
stop )
stop_system
echo "Successful ${CMD}."
;;
esac

View File

@@ -1,68 +0,0 @@
#!/usr/bin/env python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2012 Wind River Systems, Inc.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
# published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License along
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
# This is used for dumping the bb_cache.dat, the output format is:
# recipe_path PN PV PACKAGES
#
import os
import sys
import warnings
# For importing bb.cache
sys.path.insert(0, os.path.join(os.path.abspath(os.path.dirname(sys.argv[0])), '../lib'))
from bb.cache import CoreRecipeInfo
import cPickle as pickle
def main(argv=None):
"""
Get the mapping for the target recipe.
"""
if len(argv) != 1:
print >>sys.stderr, "Error, need one argument!"
return 2
cachefile = argv[0]
with open(cachefile, "rb") as cachefile:
pickled = pickle.Unpickler(cachefile)
while cachefile:
try:
key = pickled.load()
val = pickled.load()
except Exception:
break
if isinstance(val, CoreRecipeInfo) and (not val.skipped):
pn = val.pn
# Filter out the native recipes.
if key.startswith('virtual:native:') or pn.endswith("-native"):
continue
# 1.0 is the default version for a no PV recipe.
if val.__dict__.has_key("pv"):
pv = val.pv
else:
pv = "1.0"
print("%s %s %s %s" % (key, pn, pv, ' '.join(val.packages)))
if __name__ == "__main__":
sys.exit(main(sys.argv[1:]))

View File

@@ -1,24 +1,4 @@
" Vim filetype detection file
" Language: BitBake
" Author: Ricardo Salveti <rsalveti@rsalveti.net>
" Copyright: Copyright (C) 2008 Ricardo Salveti <rsalveti@rsalveti.net>
" Licence: You may redistribute this under the same terms as Vim itself
"
" This sets up the syntax highlighting for BitBake files, like .bb, .bbclass and .inc
if &compatible || version < 600
finish
endif
" .bb, .bbappend and .bbclass
au BufNewFile,BufRead *.{bb,bbappend,bbclass} set filetype=bitbake
" .inc
au BufNewFile,BufRead *.inc set filetype=bitbake
" .conf
au BufNewFile,BufRead *.conf
\ if (match(expand("%:p:h"), "conf") > 0) |
\ set filetype=bitbake |
\ endif
au BufNewFile,BufRead *.bb setfiletype bitbake
au BufNewFile,BufRead *.bbclass setfiletype bitbake
au BufNewFile,BufRead *.inc setfiletype bitbake
" au BufNewFile,BufRead *.conf setfiletype bitbake

View File

@@ -1,2 +0,0 @@
set sts=4 sw=4 et
set cms=#%s

View File

@@ -1,85 +0,0 @@
" Vim plugin file
" Purpose: Create a template for new bb files
" Author: Ricardo Salveti <rsalveti@gmail.com>
" Copyright: Copyright (C) 2008 Ricardo Salveti <rsalveti@gmail.com>
"
" This file is licensed under the MIT license, see COPYING.MIT in
" this source distribution for the terms.
"
" Based on the gentoo-syntax package
"
" Will try to use git to find the user name and email
if &compatible || v:version < 600
finish
endif
fun! <SID>GetUserName()
let l:user_name = system("git config --get user.name")
if v:shell_error
return "Unknown User"
else
return substitute(l:user_name, "\n", "", "")
endfun
fun! <SID>GetUserEmail()
let l:user_email = system("git config --get user.email")
if v:shell_error
return "unknow@user.org"
else
return substitute(l:user_email, "\n", "", "")
endfun
fun! BBHeader()
let l:current_year = strftime("%Y")
let l:user_name = <SID>GetUserName()
let l:user_email = <SID>GetUserEmail()
0 put ='# Copyright (C) ' . l:current_year .
\ ' ' . l:user_name . ' <' . l:user_email . '>'
put ='# Released under the MIT license (see COPYING.MIT for the terms)'
$
endfun
fun! NewBBTemplate()
let l:paste = &paste
set nopaste
" Get the header
call BBHeader()
" New the bb template
put ='DESCRIPTION = \"\"'
put ='HOMEPAGE = \"\"'
put ='LICENSE = \"\"'
put ='SECTION = \"\"'
put ='DEPENDS = \"\"'
put ='PR = \"r0\"'
put =''
put ='SRC_URI = \"\"'
" Go to the first place to edit
0
/^DESCRIPTION =/
exec "normal 2f\""
if paste == 1
set paste
endif
endfun
if !exists("g:bb_create_on_empty")
let g:bb_create_on_empty = 1
endif
" disable in case of vimdiff
if v:progname =~ "vimdiff"
let g:bb_create_on_empty = 0
endif
augroup NewBB
au BufNewFile *.bb
\ if g:bb_create_on_empty |
\ call NewBBTemplate() |
\ endif
augroup END

View File

@@ -1,126 +1,120 @@
" Vim syntax file
" Language: BitBake bb/bbclasses/inc
" Author: Chris Larson <kergoth@handhelds.org>
" Ricardo Salveti <rsalveti@rsalveti.net>
" Copyright: Copyright (C) 2004 Chris Larson <kergoth@handhelds.org>
" Copyright (C) 2008 Ricardo Salveti <rsalveti@rsalveti.net>
"
" Copyright (C) 2004 Chris Larson <kergoth@handhelds.org>
" This file is licensed under the MIT license, see COPYING.MIT in
" this source distribution for the terms.
"
" Syntax highlighting for bb, bbclasses and inc files.
" Language: BitBake
" Maintainer: Chris Larson <kergoth@handhelds.org>
" Filenames: *.bb, *.bbclass
if version < 600
syntax clear
elseif exists("b:current_syntax")
finish
endif
syn case match
" Catch incorrect syntax (only matches if nothing else does)
"
" It's an entirely new type, just has specific syntax in shell and python code
syn match bbUnmatched "."
if &compatible || v:version < 600
finish
endif
if exists("b:current_syntax")
finish
endif
" Other
syn match bbComment "^#.*$" display contains=bbTodo
syn keyword bbTodo TODO FIXME XXX contained
syn match bbDelimiter "[(){}=]" contained
syn match bbQuote /['"]/ contained
syn match bbArrayBrackets "[\[\]]" contained
" BitBake strings
syn match bbContinue "\\$"
syn region bbString matchgroup=bbQuote start=/"/ skip=/\\$/ excludenl end=/"/ contained keepend contains=bbTodo,bbContinue,bbVarDeref
syn region bbString matchgroup=bbQuote start=/'/ skip=/\\$/ excludenl end=/'/ contained keepend contains=bbTodo,bbContinue,bbVarDeref
" BitBake variable metadata
syn keyword bbExportFlag export contained nextgroup=bbIdentifier skipwhite
syn match bbVarDeref "${[a-zA-Z0-9\-_\.]\+}" contained
syn match bbVarDef "^\(export\s*\)\?\([a-zA-Z0-9\-_\.]\+\(_[${}a-zA-Z0-9\-_\.]\+\)\?\)\s*\(:=\|+=\|=+\|\.=\|=\.\|?=\|=\)\@=" contains=bbExportFlag,bbIdentifier,bbVarDeref nextgroup=bbVarEq
syn match bbIdentifier "[a-zA-Z0-9\-_\.]\+" display contained
"syn keyword bbVarEq = display contained nextgroup=bbVarValue
syn match bbVarEq "\(:=\|+=\|=+\|\.=\|=\.\|?=\|=\)" contained nextgroup=bbVarValue
syn match bbVarValue ".*$" contained contains=bbString,bbVarDeref
" BitBake variable metadata flags
syn match bbVarFlagDef "^\([a-zA-Z0-9\-_\.]\+\)\(\[[a-zA-Z0-9\-_\.]\+\]\)\@=" contains=bbIdentifier nextgroup=bbVarFlagFlag
syn region bbVarFlagFlag matchgroup=bbArrayBrackets start="\[" end="\]\s*\(=\)\@=" keepend excludenl contained contains=bbIdentifier nextgroup=bbVarEq
"syn match bbVarFlagFlag "\[\([a-zA-Z0-9\-_\.]\+\)\]\s*\(=\)\@=" contains=bbIdentifier nextgroup=bbVarEq
" Functions!
syn match bbFunction "\h\w*" display contained
" BitBake python metadata
syn include @python syntax/python.vim
if exists("b:current_syntax")
unlet b:current_syntax
endif
" BitBake syntax
syn keyword bbPythonFlag python contained nextgroup=bbFunction
syn match bbPythonFuncDef "^\(python\s\+\)\(\w\+\)\?\(\s*()\s*\)\({\)\@=" contains=bbPythonFlag,bbFunction,bbDelimiter nextgroup=bbPythonFuncRegion skipwhite
syn region bbPythonFuncRegion matchgroup=bbDelimiter start="{\s*$" end="^}\s*$" keepend contained contains=@python
"hi def link bbPythonFuncRegion Comment
" Matching case
syn case match
" Indicates the error when nothing is matched
syn match bbUnmatched "."
" Comments
syn cluster bbCommentGroup contains=bbTodo,@Spell
syn keyword bbTodo COMBAK FIXME TODO XXX contained
syn match bbComment "#.*$" contains=@bbCommentGroup
" String helpers
syn match bbQuote +['"]+ contained
syn match bbDelimiter "[(){}=]" contained
syn match bbArrayBrackets "[\[\]]" contained
" BitBake strings
syn match bbContinue "\\$"
syn region bbString matchgroup=bbQuote start=+"+ skip=+\\$+ end=+"+ contained contains=bbTodo,bbContinue,bbVarDeref,bbVarPyValue,@Spell
syn region bbString matchgroup=bbQuote start=+'+ skip=+\\$+ end=+'+ contained contains=bbTodo,bbContinue,bbVarDeref,bbVarPyValue,@Spell
" Vars definition
syn match bbExport "^export" nextgroup=bbIdentifier skipwhite
syn keyword bbExportFlag export contained nextgroup=bbIdentifier skipwhite
syn match bbIdentifier "[a-zA-Z0-9\-_\.\/\+]\+" display contained
syn match bbVarDeref "${[a-zA-Z0-9\-_\.\/\+]\+}" contained
syn match bbVarEq "\(:=\|+=\|=+\|\.=\|=\.\|?=\|??=\|=\)" contained nextgroup=bbVarValue
syn match bbVarDef "^\(export\s*\)\?\([a-zA-Z0-9\-_\.\/\+]\+\(_[${}a-zA-Z0-9\-_\.\/\+]\+\)\?\)\s*\(:=\|+=\|=+\|\.=\|=\.\|?=\|??=\|=\)\@=" contains=bbExportFlag,bbIdentifier,bbVarDeref nextgroup=bbVarEq
syn match bbVarValue ".*$" contained contains=bbString,bbVarDeref,bbVarPyValue
syn region bbVarPyValue start=+${@+ skip=+\\$+ end=+}+ contained contains=@python
" Vars metadata flags
syn match bbVarFlagDef "^\([a-zA-Z0-9\-_\.]\+\)\(\[[a-zA-Z0-9\-_\.]\+\]\)\@=" contains=bbIdentifier nextgroup=bbVarFlagFlag
syn region bbVarFlagFlag matchgroup=bbArrayBrackets start="\[" end="\]\s*\(=\|+=\|=+\|?=\)\@=" contained contains=bbIdentifier nextgroup=bbVarEq
" Includes and requires
syn keyword bbInclude inherit include require contained
syn match bbIncludeRest ".*$" contained contains=bbString,bbVarDeref
syn match bbIncludeLine "^\(inherit\|include\|require\)\s\+" contains=bbInclude nextgroup=bbIncludeRest
" Add taks and similar
syn keyword bbStatement addtask addhandler after before EXPORT_FUNCTIONS contained
syn match bbStatementRest ".*$" skipwhite contained contains=bbStatement
syn match bbStatementLine "^\(addtask\|addhandler\|after\|before\|EXPORT_FUNCTIONS\)\s\+" contains=bbStatement nextgroup=bbStatementRest
" OE Important Functions
syn keyword bbOEFunctions do_fetch do_unpack do_patch do_configure do_compile do_stage do_install do_package contained
" Generic Functions
syn match bbFunction "\h[0-9A-Za-z_-]*" display contained contains=bbOEFunctions
" BitBake shell metadata
syn include @shell syntax/sh.vim
if exists("b:current_syntax")
unlet b:current_syntax
endif
syn keyword bbShFakeRootFlag fakeroot contained
syn match bbShFuncDef "^\(fakeroot\s*\)\?\([0-9A-Za-z_${}-]\+\)\(python\)\@<!\(\s*()\s*\)\({\)\@=" contains=bbShFakeRootFlag,bbFunction,bbVarDeref,bbDelimiter nextgroup=bbShFuncRegion skipwhite
syn region bbShFuncRegion matchgroup=bbDelimiter start="{\s*$" end="^}\s*$" contained contains=@shell
" Python value inside shell functions
syn region shDeref start=+${@+ skip=+\\$+ excludenl end=+}+ contained contains=@python
syn keyword bbFakerootFlag fakeroot contained nextgroup=bbFunction
syn match bbShellFuncDef "^\(fakeroot\s*\)\?\(\w\+\)\(python\)\@<!\(\s*()\s*\)\({\)\@=" contains=bbFakerootFlag,bbFunction,bbDelimiter nextgroup=bbShellFuncRegion skipwhite
syn region bbShellFuncRegion matchgroup=bbDelimiter start="{\s*$" end="^}\s*$" keepend contained contains=@shell
"hi def link bbShellFuncRegion Comment
" BitBake python metadata
syn keyword bbPyFlag python contained
syn match bbPyFuncDef "^\(python\s\+\)\([0-9A-Za-z_${}-]\+\)\?\(\s*()\s*\)\({\)\@=" contains=bbPyFlag,bbFunction,bbVarDeref,bbDelimiter nextgroup=bbPyFuncRegion skipwhite
syn region bbPyFuncRegion matchgroup=bbDelimiter start="{\s*$" end="^}\s*$" contained contains=@python
" BitBake 'def'd python functions
syn keyword bbPyDef def contained
syn region bbPyDefRegion start='^\(def\s\+\)\([0-9A-Za-z_-]\+\)\(\s*(.*)\s*\):\s*$' end='^\(\s\|$\)\@!' contains=@python
syn keyword bbDef def contained
syn region bbDefRegion start='^def\s\+\w\+\s*([^)]*)\s*:\s*$' end='^\(\s\|$\)\@!' contains=@python
" Highlighting Definitions
hi def link bbUnmatched Error
hi def link bbInclude Include
hi def link bbTodo Todo
hi def link bbComment Comment
hi def link bbQuote String
hi def link bbString String
hi def link bbDelimiter Keyword
hi def link bbArrayBrackets Statement
hi def link bbContinue Special
hi def link bbExport Type
hi def link bbExportFlag Type
hi def link bbIdentifier Identifier
hi def link bbVarDeref PreProc
hi def link bbVarDef Identifier
hi def link bbVarValue String
hi def link bbShFakeRootFlag Type
hi def link bbFunction Function
hi def link bbPyFlag Type
hi def link bbPyDef Statement
hi def link bbStatement Statement
hi def link bbStatementRest Identifier
hi def link bbOEFunctions Special
hi def link bbVarPyValue PreProc
" BitBake statements
syn keyword bbStatement include inherit require addtask addhandler EXPORT_FUNCTIONS display contained
syn match bbStatementLine "^\(include\|inherit\|require\|addtask\|addhandler\|EXPORT_FUNCTIONS\)\s\+" contains=bbStatement nextgroup=bbStatementRest
syn match bbStatementRest ".*$" contained contains=bbString,bbVarDeref
" Highlight
"
hi def link bbArrayBrackets Statement
hi def link bbUnmatched Error
hi def link bbVarDeref String
hi def link bbContinue Special
hi def link bbDef Statement
hi def link bbPythonFlag Type
hi def link bbExportFlag Type
hi def link bbFakerootFlag Type
hi def link bbStatement Statement
hi def link bbString String
hi def link bbTodo Todo
hi def link bbComment Comment
hi def link bbOperator Operator
hi def link bbError Error
hi def link bbFunction Function
hi def link bbDelimiter Delimiter
hi def link bbIdentifier Identifier
hi def link bbVarEq Operator
hi def link bbQuote String
hi def link bbVarValue String
let b:current_syntax = "bb"

View File

@@ -1,91 +0,0 @@
# This is a single Makefile to handle all generated BitBake documents.
# The Makefile needs to live in the documentation directory and all figures used
# in any manuals must be .PNG files and live in the individual book's figures
# directory.
#
# The Makefile has these targets:
#
# pdf: generates a PDF version of a manual.
# html: generates an HTML version of a manual.
# tarball: creates a tarball for the doc files.
# validate: validates
# clean: removes files
#
# The Makefile generates an HTML and PDF version of every document. The
# variable DOC indicates the folder name for a given manual.
#
# To build a manual, you must invoke 'make' with the DOC argument.
#
# Examples:
#
# make DOC=bitbake-user-manual
# make pdf DOC=bitbake-user-manual
#
# The first example generates the HTML and PDF versions of the User Manual.
# The second example generates the HTML version only of the User Manual.
#
ifeq ($(DOC),bitbake-user-manual)
XSLTOPTS = --stringparam html.stylesheet bitbake-user-manual-style.css \
--stringparam chapter.autolabel 1 \
--stringparam section.autolabel 1 \
--stringparam section.label.includes.component.label 1 \
--xinclude
ALLPREQ = html pdf tarball
TARFILES = bitbake-user-manual-style.css bitbake-user-manual.html bitbake-user-manual.pdf figures/bitbake-title.png
MANUALS = $(DOC)/$(DOC).html $(DOC)/$(DOC).pdf
FIGURES = figures
STYLESHEET = $(DOC)/*.css
endif
##
# These URI should be rewritten by your distribution's xml catalog to
# match your localy installed XSL stylesheets.
XSL_BASE_URI = http://docbook.sourceforge.net/release/xsl/current
XSL_XHTML_URI = $(XSL_BASE_URI)/xhtml/docbook.xsl
all: $(ALLPREQ)
pdf:
ifeq ($(DOC),bitbake-user-manual)
@echo " "
@echo "********** Building."$(DOC)
@echo " "
cd $(DOC); ../tools/docbook-to-pdf $(DOC).xml ../template; cd ..
endif
html:
ifeq ($(DOC),bitbake-user-manual)
# See http://www.sagehill.net/docbookxsl/HtmlOutput.html
@echo " "
@echo "******** Building "$(DOC)
@echo " "
cd $(DOC); xsltproc $(XSLTOPTS) -o $(DOC).html $(DOC)-customization.xsl $(DOC).xml; cd ..
endif
tarball: html
@echo " "
@echo "******** Creating Tarball of document files"
@echo " "
cd $(DOC); tar -cvzf $(DOC).tgz $(TARFILES); cd ..
validate:
cd $(DOC); xmllint --postvalid --xinclude --noout $(DOC).xml; cd ..
publish:
@if test -f $(DOC)/$(DOC).html; \
then \
echo " "; \
echo "******** Publishing "$(DOC)".html"; \
echo " "; \
scp -r $(MANUALS) $(STYLESHEET) docs.yp:/var/www/www.yoctoproject.org-docs/$(VER)/$(DOC); \
cd $(DOC); scp -r $(FIGURES) docs.yp:/var/www/www.yoctoproject.org-docs/$(VER)/$(DOC); \
else \
echo " "; \
echo $(DOC)".html missing. Generate the file first then try again."; \
echo " "; \
fi
clean:
rm -rf $(MANUALS); rm $(DOC)/$(DOC).tgz;

View File

@@ -1,39 +0,0 @@
Documentation
=============
This is the directory that contains the BitBake documentation.
Manual Organization
===================
Folders exist for individual manuals as follows:
* bitbake-user-manual - The BitBake User Manual
Each folder is self-contained regarding content and figures.
If you want to find HTML versions of the BitBake manuals on the web,
go to http://www.openembedded.org/wiki/Documentation.
Makefile
========
The Makefile processes manual directories to create HTML, PDF,
tarballs, etc. Details on how the Makefile work are documented
inside the Makefile. See that file for more information.
To build a manual, you run the make command and pass it the name
of the folder containing the manual's contents.
For example, the following command run from the documentation directory
creates an HTML and a PDF version of the BitBake User Manual.
The DOC variable specifies the manual you are making:
$ make DOC=bitbake-user-manual
template
========
Contains various templates, fonts, and some old PNG files.
tools
=====
Contains a tool to convert the DocBook files to PDF format.

View File

@@ -1,15 +0,0 @@
<?xml version='1.0'?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns="http://www.w3.org/1999/xhtml" xmlns:fo="http://www.w3.org/1999/XSL/Format" version="1.0">
<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/xhtml/docbook.xsl" />
<xsl:param name="html.stylesheet" select="'user-manual-style.css'" />
<xsl:param name="chapter.autolabel" select="1" />
<!-- <xsl:param name="appendix.autolabel" select="A" /> -->
<xsl:param name="section.autolabel" select="1" />
<xsl:param name="section.label.includes.component.label" select="1" />
<xsl:param name="appendix.autolabel">A</xsl:param>
<!-- <xsl:param name="generate.toc" select="'article nop'"></xsl:param> -->
</xsl:stylesheet>

View File

@@ -1,850 +0,0 @@
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<chapter id="bitbake-user-manual-execution">
<title>Execution</title>
<para>
The primary purpose for running BitBake is to produce some kind
of output such as an image, a kernel, or a software development
kit.
Of course, you can execute the <filename>bitbake</filename>
command with options that cause it to execute single tasks,
compile single recipe files, capture or clear data, or simply
return information about the execution environment.
</para>
<para>
This chapter describes BitBake's execution process from start
to finish when you use it to create an image.
The execution process is launched using the following command
form:
<literallayout class='monospaced'>
$ bitbake &lt;target&gt;
</literallayout>
For information on the BitBake command and its options,
see
"<link linkend='bitbake-user-manual-command'>The BitBake Command</link>"
section.
</para>
<note>
Prior to executing BitBake, you should take advantage of parallel
thread execution by setting the
<link linkend='var-BB_NUMBER_THREADS'><filename>BB_NUMBER_THREADS</filename></link>
variable in your <filename>local.conf</filename>
configuration file.
</note>
<section id='parsing-the-base-configuration-metadata'>
<title>Parsing the Base Configuration Metadata</title>
<para>
The first thing BitBake does is parse base configuration
metadata.
Base configuration metadata consists of the
<filename>bblayers.conf</filename> file to determine what
layers BitBake needs to recognize, all necessary
<filename>layer.conf</filename> files (one from each layer),
and <filename>bitbake.conf</filename>.
The data itself is of various types:
<itemizedlist>
<listitem><para><emphasis>Recipes:</emphasis>
Details about particular pieces of software.
</para></listitem>
<listitem><para><emphasis>Class Data:</emphasis>
An abstraction of common build information
(e.g. how to build a Linux kernel).
</para></listitem>
<listitem><para><emphasis>Configuration Data:</emphasis>
Machine-specific settings, policy decisions,
and so forth.
Configuration data acts as the glue to bind everything
together.</para></listitem>
</itemizedlist>
</para>
<para>
The <filename>layer.conf</filename> files are used to
construct key variables such as
<link linkend='var-BBPATH'><filename>BBPATH</filename></link>
and
<link linkend='var-BBFILES'><filename>BBFILES</filename></link>.
<filename>BBPATH</filename> is used to search for
configuration and class files under
<filename>conf/</filename> and <filename>class/</filename>
directories, respectively.
<filename>BBFILES</filename> is used to find recipe files
(<filename>.bb</filename> and <filename>.bbappend</filename>).
If there is no <filename>bblayers.conf</filename> file,
it is assumed the user has set the <filename>BBPATH</filename>
and <filename>BBFILES</filename> directly in the environment.
</para>
<para>
Next, the <filename>bitbake.conf</filename> file is searched
using the <filename>BBPATH</filename> variable that was
just constructed.
The <filename>bitbake.conf</filename> file may also include other
configuration files using the
<filename>include</filename> or
<filename>require</filename> directives.
</para>
<para>
Prior to parsing configuration files, Bitbake looks
at certain variables, including:
<itemizedlist>
<listitem><para><link linkend='var-BB_ENV_WHITELIST'><filename>BB_ENV_WHITELIST</filename></link></para></listitem>
<listitem><para><link linkend='var-BB_PRESERVE_ENV'><filename>BB_PRESERVE_ENV</filename></link></para></listitem>
<listitem><para><link linkend='var-BB_ENV_EXTRAWHITE'><filename>BB_ENV_EXTRAWHITE</filename></link></para></listitem>
<listitem><para>
<link linkend='var-BITBAKE_UI'><filename>BITBAKE_UI</filename></link>
</para></listitem>
</itemizedlist>
You can find information on how to pass environment variables into the BitBake
execution environment in the
"<link linkend='passing-information-into-the-build-task-environment'>Passing Information Into the Build Task Environment</link>" section.
</para>
<para>
The base configuration metadata is global
and therefore affects all recipes and tasks that are executed.
</para>
<para>
BitBake first searches the current working directory for an
optional <filename>conf/bblayers.conf</filename> configuration file.
This file is expected to contain a
<link linkend='var-BBLAYERS'><filename>BBLAYERS</filename></link>
variable that is a space delimited list of 'layer' directories.
Recall that if BitBake cannot find a <filename>bblayers.conf</filename>
file then it is assumed the user has set the <filename>BBPATH</filename>
and <filename>BBFILES</filename> directly in the environment.
</para>
<para>
For each directory (layer) in this list, a <filename>conf/layer.conf</filename>
file is searched for and parsed with the
<link linkend='var-LAYERDIR'><filename>LAYERDIR</filename></link>
variable being set to the directory where the layer was found.
The idea is these files automatically setup
<link linkend='var-BBPATH'><filename>BBPATH</filename></link>
and other variables correctly for a given build directory.
</para>
<para>
BitBake then expects to find the <filename>conf/bitbake.conf</filename>
file somewhere in the user-specified <filename>BBPATH</filename>.
That configuration file generally has include directives to pull
in any other metadata such as files specific to the architecture,
the machine, the local environment, and so forth.
</para>
<para>
Only variable definitions and include directives are allowed
in <filename>.conf</filename> files.
Some variables directly influence BitBake's behavior.
These variables might have been set from the environment
depending on the environment variables previously
mentioned or set in the configuration files.
The
"<link linkend='ref-variables-glos'>Variables Glossary</link>"
chapter presents a full list of variables.
</para>
<para>
After parsing configuration files, BitBake uses its rudimentary
inheritance mechanism, which is through class files, to inherit
some standard classes.
BitBake parses a class when the inherit directive responsible
for getting that class is encountered.
</para>
<para>
The <filename>base.bbclass</filename> file is always included.
Other classes that are specified in the configuration using the
<link linkend='var-INHERIT'><filename>INHERIT</filename></link>
variable are also included.
BitBake searches for class files in a "classes" subdirectory under
the paths in <filename>BBPATH</filename> in the same way as
configuration files.
</para>
<para>
A good way to get an idea of the configuration files and
the class files used in your execution environment is to
run the following BitBake command:
<literallayout class='monospaced'>
$ bitbake -e > mybb.log
</literallayout>
Examining the top of the <filename>mybb.log</filename>
shows you the many configuration files and class files
used in your execution environment.
</para>
<note>
<para>
You need to be aware of how BitBake parses curly braces.
If a recipe uses a closing curly brace within the function and
the character has no leading spaces, BitBake produces a parsing
error.
If you use a pair of curly brace in a shell function, the
closing curly brace must not be located at the start of the line
without leading spaces.
</para>
<para>
Here is an example that causes BitBake to produce a parsing
error:
<literallayout class='monospaced'>
fakeroot create_shar() {
cat &lt;&lt; "EOF" &gt; ${SDK_DEPLOY}/${TOOLCHAIN_OUTPUTNAME}.sh
usage()
{
echo "test"
###### The following "}" at the start of the line causes a parsing error ######
}
EOF
}
</literallayout>
Writing the recipe this way avoids the error:
<literallayout class='monospaced'>
fakeroot create_shar() {
cat &lt;&lt; "EOF" &gt; ${SDK_DEPLOY}/${TOOLCHAIN_OUTPUTNAME}.sh
usage()
{
echo "test"
######The following "}" with a leading space at the start of the line avoids the error ######
}
EOF
}
</literallayout>
</para>
</note>
</section>
<section id='locating-and-parsing-recipes'>
<title>Locating and Parsing Recipes</title>
<para>
During the configuration phase, BitBake will have set
<link linkend='var-BBFILES'><filename>BBFILES</filename></link>.
BitBake now uses it to construct a list of recipes to parse,
along with any append files (<filename>.bbappend</filename>)
to apply.
<filename>BBFILES</filename> is a space-separated list of
available files and supports wildcards.
An example would be:
<literallayout class='monospaced'>
BBFILES = "/path/to/bbfiles/*.bb /path/to/appends/*.bbappend"
</literallayout>
BitBake parses each recipe and append file located
with <filename>BBFILES</filename> and stores the values of
various variables into the datastore.
<note>
Append files are applied in the order they are encountered in
<filename>BBFILES</filename>.
</note>
For each file, a fresh copy of the base configuration is
made, then the recipe is parsed line by line.
Any inherit statements cause BitBake to find and
then parse class files (<filename>.bbclass</filename>)
using
<link linkend='var-BBPATH'><filename>BBPATH</filename></link>
as the search path.
Finally, BitBake parses in order any append files found in
<filename>BBFILES</filename>.
</para>
<para>
One common convention is to use the recipe filename to define
pieces of metadata.
For example, in <filename>bitbake.conf</filename> the recipe
name and version set
<link linkend='var-PN'><filename>PN</filename></link> and
<link linkend='var-PV'><filename>PV</filename></link>:
<literallayout class='monospaced'>
PV = "${@bb.parse.BBHandler.vars_from_file(d.getVar('FILE'),d)[1] or '1.0'}"
PN = "${@bb.parse.BBHandler.vars_from_file(d.getVar('FILE'),d)[0] or 'defaultpkgname'}"
</literallayout>
In this example, a recipe called "something_1.2.3.bb" sets
<filename>PN</filename> to "something" and
<filename>PV</filename> to "1.2.3".
</para>
<para>
By the time parsing is complete for a recipe, BitBake
has a list of tasks that the recipe defines and a set of
data consisting of keys and values as well as
dependency information about the tasks.
</para>
<para>
BitBake does not need all of this information.
It only needs a small subset of the information to make
decisions about the recipe.
Consequently, BitBake caches the values in which it is
interested and does not store the rest of the information.
Experience has shown it is faster to re-parse the metadata than to
try and write it out to the disk and then reload it.
</para>
<para>
Where possible, subsequent BitBake commands reuse this cache of
recipe information.
The validity of this cache is determined by first computing a
checksum of the base configuration data (see
<link linkend='var-BB_HASHCONFIG_WHITELIST'><filename>BB_HASHCONFIG_WHITELIST</filename></link>)
and then checking if the checksum matches.
If that checksum matches what is in the cache and the recipe
and class files have not changed, Bitbake is able to use
the cache.
BitBake then reloads the cached information about the recipe
instead of reparsing it from scratch.
</para>
<para>
Recipe file collections exist to allow the user to
have multiple repositories of
<filename>.bb</filename> files that contain the same
exact package.
For example, one could easily use them to make one's
own local copy of an upstream repository, but with
custom modifications that one does not want upstream.
Here is an example:
<literallayout class='monospaced'>
BBFILES = "/stuff/openembedded/*/*.bb /stuff/openembedded.modified/*/*.bb"
BBFILE_COLLECTIONS = "upstream local"
BBFILE_PATTERN_upstream = "^/stuff/openembedded/"
BBFILE_PATTERN_local = "^/stuff/openembedded.modified/"
BBFILE_PRIORITY_upstream = "5"
BBFILE_PRIORITY_local = "10"
</literallayout>
<note>
The layers mechanism is now the preferred method of collecting
code.
While the collections code remains, its main use is to set layer
priorities and to deal with overlap (conflicts) between layers.
</note>
</para>
</section>
<section id='bb-bitbake-providers'>
<title>Preferences and Providers</title>
<para>
Assuming BitBake has been instructed to execute a target
and that all the recipe files have been parsed, BitBake
starts to figure out how to build the target.
BitBake starts by looking through the
<link linkend='var-PROVIDES'><filename>PROVIDES</filename></link>
set in recipe files.
The default <filename>PROVIDES</filename> for a recipe is its name
(<link linkend='var-PN'><filename>PN</filename></link>),
however, a recipe can provide multiple things.
</para>
<para>
As an example of adding an extra provider, suppose a recipe named
<filename>foo_1.0.bb</filename> contained the following:
<literallayout class='monospaced'>
PROVIDES += "virtual/bar_1.0"
</literallayout>
The recipe now provides both "foo_1.0" and "virtual/bar_1.0".
The "virtual/" namespace is often used to denote cases where
multiple providers are expected with the user choosing between
them.
Kernels and toolchain components are common cases of this in
OpenEmbedded.
</para>
<para>
Sometimes a target might have multiple providers.
A common example is "virtual/kernel", which is provided by each
kernel recipe.
Each machine often selects the best kernel provider by using a
line similar to the following in the machine configuration file:
<literallayout class='monospaced'>
PREFERRED_PROVIDER_virtual/kernel = "linux-yocto"
</literallayout>
The default
<link linkend='var-PREFERRED_PROVIDER'><filename>PREFERRED_PROVIDER</filename></link>
is the provider with the same name as the target.
Bitbake iterates through each target it needs to build and
resolves them and their dependencies using this process.
</para>
<para>
Understanding how providers are chosen is made complicated by the fact
that multiple versions might exist.
BitBake defaults to the highest version of a provider.
Version comparisons are made using the same method as Debian.
You can use the
<link linkend='var-PREFERRED_VERSION'><filename>PREFERRED_VERSION</filename></link>
variable to specify a particular version.
You can influence the order by using the
<link linkend='var-DEFAULT_PREFERENCE'><filename>DEFAULT_PREFERENCE</filename></link>
variable.
By default, files have a preference of "0".
Setting the <filename>DEFAULT_PREFERENCE</filename> to "-1" makes the
recipe unlikely to be used unless it is explicitly referenced.
Setting the <filename>DEFAULT_PREFERENCE</filename> to "1" makes it likely the recipe is used.
<filename>PREFERRED_VERSION</filename> overrides any <filename>DEFAULT_PREFERENCE</filename> setting.
<filename>DEFAULT_PREFERENCE</filename> is often used to mark newer and more experimental recipe
versions until they have undergone sufficient testing to be considered stable.
</para>
<para>
When there are multiple “versions” of a given recipe,
BitBake defaults to selecting the most recent
version, unless otherwise specified.
If the recipe in question has a
<link linkend='var-DEFAULT_PREFERENCE'><filename>DEFAULT_PREFERENCE</filename></link>
set lower than
the other recipes (default is 0), then it will not be
selected.
This allows the person or persons maintaining
the repository of recipe files to specify
their preference for the default selected version.
In addition, the user can specify their preferred version.
</para>
<para>
If the first recipe is named <filename>a_1.1.bb</filename>,
then the
<link linkend='var-PN'><filename>PN</filename></link> variable
will be set to “a”, and the
<link linkend='var-PV'><filename>PV</filename></link>
variable will be set to 1.1.
</para>
<para>
If we then have a recipe named <filename>a_1.2.bb</filename>, BitBake
will choose 1.2 by default.
However, if we define the following variable in a
<filename>.conf</filename> file that BitBake parses, we
can change that.
<literallayout class='monospaced'>
PREFERRED_VERSION_a = "1.1"
</literallayout>
</para>
<para>
In summary, BitBake has created a list of providers, which is prioritized, for each target.
</para>
</section>
<section id='bb-bitbake-dependencies'>
<title>Dependencies</title>
<para>
Each target BitBake builds consists of multiple tasks such as
<filename>fetch</filename>, <filename>unpack</filename>,
<filename>patch</filename>, <filename>configure</filename>,
and <filename>compile</filename>.
For best performance on multi-core systems, BitBake considers each
task as an independent
entity with its own set of dependencies.
</para>
<para>
Dependencies are defined through several variables.
You can find information about variables BitBake uses in
the <link linkend='ref-variables-glos'>Variables Glossary</link>
near the end of this manual.
At a basic level, it is sufficient to know that BitBake uses the
<link linkend='var-DEPENDS'><filename>DEPENDS</filename></link> and
<link linkend='var-RDEPENDS'><filename>RDEPENDS</filename></link> variables when
calculating dependencies.
</para>
<para>
For more information on how BitBake handles dependencies, see the
"<link linkend='dependencies'>Dependencies</link>" section.
</para>
</section>
<section id='ref-bitbake-tasklist'>
<title>The Task List</title>
<para>
Based on the generated list of providers and the dependency information,
BitBake can now calculate exactly what tasks it needs to run and in what
order it needs to run them.
The
"<link linkend='executing-tasks'>Executing Tasks</link>" section has more
information on how BitBake chooses which task to execute next.
</para>
<para>
The build now starts with BitBake forking off threads up to the limit set in the
<link linkend='var-BB_NUMBER_THREADS'><filename>BB_NUMBER_THREADS</filename></link>
variable.
BitBake continues to fork threads as long as there are tasks ready to run,
those tasks have all their dependencies met, and the thread threshold has not been
exceeded.
</para>
<para>
It is worth noting that you can greatly speed up the build time by properly setting
the <filename>BB_NUMBER_THREADS</filename> variable.
</para>
<para>
As each task completes, a timestamp is written to the directory specified by the
<link linkend='var-STAMP'><filename>STAMP</filename></link> variable.
On subsequent runs, BitBake looks in the build directory within
<filename>tmp/stamps</filename>and does not rerun
tasks that are already completed unless a timestamp is found to be invalid.
Currently, invalid timestamps are only considered on a per
recipe file basis.
So, for example, if the configure stamp has a timestamp greater than the
compile timestamp for a given target, then the compile task would rerun.
Running the compile task again, however, has no effect on other providers
that depend on that target.
</para>
<para>
The exact format of the stamps is partly configurable.
In modern versions of BitBake, a hash is appended to the
stamp so that if the configuration changes, the stamp becomes
invalid and the task is automatically rerun.
This hash, or signature used, is governed by the signature policy
that is configured (see the
"<link linkend='checksums'>Checksums (Signatures)</link>"
section for information).
It is also possible to append extra metadata to the stamp using
the "stamp-extra-info" task flag.
For example, OpenEmbedded uses this flag to make some tasks machine-specific.
</para>
<note>
Some tasks are marked as "nostamp" tasks.
No timestamp file is created when these tasks are run.
Consequently, "nostamp" tasks are always rerun.
</note>
<para>
For more information on tasks, see the
"<link linkend='tasks'>Tasks</link>" section.
</para>
</section>
<section id='executing-tasks'>
<title>Executing Tasks</title>
<para>
Tasks can either be a shell task or a Python task.
For shell tasks, BitBake writes a shell script to
<filename>${</filename><link linkend='var-T'><filename>T</filename></link><filename>}/run.do_taskname.pid</filename>
and then executes the script.
The generated shell script contains all the exported variables,
and the shell functions with all variables expanded.
Output from the shell script goes to the file
<filename>${T}/log.do_taskname.pid</filename>.
Looking at the expanded shell functions in the run file and
the output in the log files is a useful debugging technique.
</para>
<para>
For Python tasks, BitBake executes the task internally and logs
information to the controlling terminal.
Future versions of BitBake will write the functions to files
similar to the way shell tasks are handled.
Logging will be handled in a way similar to shell tasks as well.
</para>
<para>
The order in which BitBake runs the tasks is controlled by its
task scheduler.
It is possible to configure the scheduler and define custom
implementations for specific use cases.
For more information, see these variables that control the
behavior:
<itemizedlist>
<listitem><para>
<link linkend='var-BB_SCHEDULER'><filename>BB_SCHEDULER</filename></link>
</para></listitem>
<listitem><para>
<link linkend='var-BB_SCHEDULERS'><filename>BB_SCHEDULERS</filename></link>
</para></listitem>
</itemizedlist>
It is possible to have functions run before and after a task's main
function.
This is done using the "prefuncs" and "postfuncs" flags of the task
that lists the functions to run.
</para>
</section>
<section id='checksums'>
<title>Checksums (Signatures)</title>
<para>
A checksum is a unique signature of a task's inputs.
The signature of a task can be used to determine if a task
needs to be run.
Because it is a change in a task's inputs that triggers running
the task, BitBake needs to detect all the inputs to a given task.
For shell tasks, this turns out to be fairly easy because
BitBake generates a "run" shell script for each task and
it is possible to create a checksum that gives you a good idea of when
the task's data changes.
</para>
<para>
To complicate the problem, some things should not be included in
the checksum.
First, there is the actual specific build path of a given task -
the working directory.
It does not matter if the working directory changes because it should not
affect the output for target packages.
The simplistic approach for excluding the working directory is to set
it to some fixed value and create the checksum for the "run" script.
BitBake goes one step better and uses the
<link linkend='var-BB_HASHBASE_WHITELIST'><filename>BB_HASHBASE_WHITELIST</filename></link>
variable to define a list of variables that should never be included
when generating the signatures.
</para>
<para>
Another problem results from the "run" scripts containing functions that
might or might not get called.
The incremental build solution contains code that figures out dependencies
between shell functions.
This code is used to prune the "run" scripts down to the minimum set,
thereby alleviating this problem and making the "run" scripts much more
readable as a bonus.
</para>
<para>
So far we have solutions for shell scripts.
What about Python tasks?
The same approach applies even though these tasks are more difficult.
The process needs to figure out what variables a Python function accesses
and what functions it calls.
Again, the incremental build solution contains code that first figures out
the variable and function dependencies, and then creates a checksum for the data
used as the input to the task.
</para>
<para>
Like the working directory case, situations exist where dependencies
should be ignored.
For these cases, you can instruct the build process to ignore a dependency
by using a line like the following:
<literallayout class='monospaced'>
PACKAGE_ARCHS[vardepsexclude] = "MACHINE"
</literallayout>
This example ensures that the <filename>PACKAGE_ARCHS</filename> variable does not
depend on the value of <filename>MACHINE</filename>, even if it does reference it.
</para>
<para>
Equally, there are cases where we need to add dependencies BitBake
is not able to find.
You can accomplish this by using a line like the following:
<literallayout class='monospaced'>
PACKAGE_ARCHS[vardeps] = "MACHINE"
</literallayout>
This example explicitly adds the <filename>MACHINE</filename> variable as a
dependency for <filename>PACKAGE_ARCHS</filename>.
</para>
<para>
Consider a case with in-line Python, for example, where BitBake is not
able to figure out dependencies.
When running in debug mode (i.e. using <filename>-DDD</filename>), BitBake
produces output when it discovers something for which it cannot figure out
dependencies.
</para>
<para>
Thus far, this section has limited discussion to the direct inputs into a task.
Information based on direct inputs is referred to as the "basehash" in the
code.
However, there is still the question of a task's indirect inputs - the
things that were already built and present in the build directory.
The checksum (or signature) for a particular task needs to add the hashes
of all the tasks on which the particular task depends.
Choosing which dependencies to add is a policy decision.
However, the effect is to generate a master checksum that combines the basehash
and the hashes of the task's dependencies.
</para>
<para>
At the code level, there are a variety of ways both the basehash and the
dependent task hashes can be influenced.
Within the BitBake configuration file, we can give BitBake some extra information
to help it construct the basehash.
The following statement effectively results in a list of global variable
dependency excludes - variables never included in any checksum.
This example uses variables from OpenEmbedded to help illustrate
the concept:
<literallayout class='monospaced'>
BB_HASHBASE_WHITELIST ?= "TMPDIR FILE PATH PWD BB_TASKHASH BBPATH DL_DIR \
SSTATE_DIR THISDIR FILESEXTRAPATHS FILE_DIRNAME HOME LOGNAME SHELL TERM \
USER FILESPATH STAGING_DIR_HOST STAGING_DIR_TARGET COREBASE PRSERV_HOST \
PRSERV_DUMPDIR PRSERV_DUMPFILE PRSERV_LOCKDOWN PARALLEL_MAKE \
CCACHE_DIR EXTERNAL_TOOLCHAIN CCACHE CCACHE_DISABLE LICENSE_PATH SDKPKGSUFFIX"
</literallayout>
The previous example excludes the work directory, which is part of
<filename>TMPDIR</filename>.
</para>
<para>
The rules for deciding which hashes of dependent tasks to include through
dependency chains are more complex and are generally accomplished with a
Python function.
The code in <filename>meta/lib/oe/sstatesig.py</filename> shows two examples
of this and also illustrates how you can insert your own policy into the system
if so desired.
This file defines the two basic signature generators OpenEmbedded Core
uses: "OEBasic" and "OEBasicHash".
By default, there is a dummy "noop" signature handler enabled in BitBake.
This means that behavior is unchanged from previous versions.
<filename>OE-Core</filename> uses the "OEBasicHash" signature handler by default
through this setting in the <filename>bitbake.conf</filename> file:
<literallayout class='monospaced'>
BB_SIGNATURE_HANDLER ?= "OEBasicHash"
</literallayout>
The "OEBasicHash" <filename>BB_SIGNATURE_HANDLER</filename> is the same as the
"OEBasic" version but adds the task hash to the stamp files.
This results in any metadata change that changes the task hash, automatically
causing the task to be run again.
This removes the need to bump
<link linkend='var-PR'><filename>PR</filename></link>
values, and changes to metadata automatically ripple across the build.
</para>
<para>
It is also worth noting that the end result of these signature generators is to
make some dependency and hash information available to the build.
This information includes:
<itemizedlist>
<listitem><para><filename>BB_BASEHASH_task-&lt;taskname&gt;</filename>:
The base hashes for each task in the recipe.
</para></listitem>
<listitem><para><filename>BB_BASEHASH_&lt;filename:taskname&gt;</filename>:
The base hashes for each dependent task.
</para></listitem>
<listitem><para><filename>BBHASHDEPS_&lt;filename:taskname&gt;</filename>:
The task dependencies for each task.
</para></listitem>
<listitem><para><filename>BB_TASKHASH</filename>:
The hash of the currently running task.
</para></listitem>
</itemizedlist>
</para>
<para>
It is worth noting that BitBake's "-S" option lets you
debug Bitbake's processing of signatures.
The options passed to -S allow different debugging modes
to be used, either using BitBake's own debug functions
or possibly those defined in the metadata/signature handler
itself.
The simplest parameter to pass is "none", which causes a
set of signature information to be written out into
<filename>STAMP_DIR</filename>
corresponding to the targets specified.
The other currently available parameter is "printdiff",
which causes BitBake to try to establish the closest
signature match it can (e.g. in the sstate cache) and then
run <filename>bitbake-diffsigs</filename> over the matches
to determine the stamps and delta where these two
stamp trees diverge.
<note>
It is likely that future versions of BitBake with
provide other signature handlers triggered through
additional "-S" paramters.
</note>
</para>
<para>
You can find more information on checksum metadata in the
"<link linkend='task-checksums-and-setscene'>Task Checksums and Setscene</link>"
section.
</para>
</section>
<section id='setscene'>
<title>Setscene</title>
<para>
The setscene process enables BitBake to handle "pre-built" artifacts.
The ability to handle and reuse these artifacts allows BitBake
the luxury of not having to build something from scratch every time.
Instead, BitBake can use, when possible, existing build artifacts.
</para>
<para>
BitBake needs to have reliable data indicating whether or not an
artifact is compatible.
Signatures, described in the previous section, provide an ideal
way of representing whether an artifact is compatible.
If a signature is the same, an object can be reused.
</para>
<para>
If an object can be reused, the problem then becomes how to
replace a given task or set of tasks with the pre-built artifact.
BitBake solves the problem with the "setscene" process.
</para>
<para>
When BitBake is asked to build a given target, before building anything,
it first asks whether cached information is available for any of the
targets it's building, or any of the intermediate targets.
If cached information is available, BitBake uses this information instead of
running the main tasks.
</para>
<para>
BitBake first calls the function defined by the
<link linkend='var-BB_HASHCHECK_FUNCTION'><filename>BB_HASHCHECK_FUNCTION</filename></link>
variable with a list of tasks and corresponding
hashes it wants to build.
This function is designed to be fast and returns a list
of the tasks for which it believes in can obtain artifacts.
</para>
<para>
Next, for each of the tasks that were returned as possibilities,
BitBake executes a setscene version of the task that the possible
artifact covers.
Setscene versions of a task have the string "_setscene" appended to the
task name.
So, for example, the task with the name <filename>xxx</filename> has
a setscene task named <filename>xxx_setscene</filename>.
The setscene version of the task executes and provides the necessary
artifacts returning either success or failure.
</para>
<para>
As previously mentioned, an artifact can cover more than one task.
For example, it is pointless to obtain a compiler if you
already have the compiled binary.
To handle this, BitBake calls the
<link linkend='var-BB_SETSCENE_DEPVALID'><filename>BB_SETSCENE_DEPVALID</filename></link>
function for each successful setscene task to know whether or not it needs
to obtain the dependencies of that task.
</para>
<para>
Finally, after all the setscene tasks have executed, BitBake calls the
function listed in
<link linkend='var-BB_SETSCENE_VERIFY_FUNCTION'><filename>BB_SETSCENE_VERIFY_FUNCTION</filename></link>
with the list of tasks BitBake thinks has been "covered".
The metadata can then ensure that this list is correct and can
inform BitBake that it wants specific tasks to be run regardless
of the setscene result.
</para>
<para>
You can find more information on setscene metadata in the
"<link linkend='task-checksums-and-setscene'>Task Checksums and Setscene</link>"
section.
</para>
</section>
</chapter>

View File

@@ -1,622 +0,0 @@
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<chapter>
<title>File Download Support</title>
<para>
BitBake's fetch module is a standalone piece of library code
that deals with the intricacies of downloading source code
and files from remote systems.
Fetching source code is one of the corner stones of building software.
As such, this module forms an important part of BitBake.
</para>
<para>
The current fetch module is called "fetch2" and refers to the
fact that it is the second major version of the API.
The original version is obsolete and removed from the codebase.
Thus, in all cases, "fetch" refers to "fetch2" in this
manual.
</para>
<section id='the-download-fetch'>
<title>The Download (Fetch)</title>
<para>
BitBake takes several steps when fetching source code or files.
The fetcher codebase deals with two distinct processes in order:
obtaining the files from somewhere (cached or otherwise)
and then unpacking those files into a specific location and
perhaps in a specific way.
Getting and unpacking the files is often optionally followed
by patching.
Patching, however, is not covered by this module.
</para>
<para>
The code to execute the first part of this process, a fetch,
looks something like the following:
<literallayout class='monospaced'>
src_uri = (d.getVar('SRC_URI', True) or "").split()
fetcher = bb.fetch2.Fetch(src_uri, d)
fetcher.download()
</literallayout>
This code sets up an instance of the fetch class.
The instance uses a space-separated list of URLs from the
<link linkend='var-SRC_URI'><filename>SRC_URI</filename></link>
variable and then calls the <filename>download</filename>
method to download the files.
</para>
<para>
The instantiation of the fetch class is usually followed by:
<literallayout class='monospaced'>
rootdir = l.getVar('WORKDIR', True)
fetcher.unpack(rootdir)
</literallayout>
This code unpacks the downloaded files to the
specified by <filename>WORKDIR</filename>.
<note>
For convenience, the naming in these examples matches
the variables used by OpenEmbedded.
</note>
The <filename>SRC_URI</filename> and <filename>WORKDIR</filename>
variables are not coded into the fetcher.
They variables can (and are) called with different variable names.
In OpenEmbedded for example, the shared state (sstate) code uses
the fetch module to fetch the sstate files.
</para>
<para>
When the <filename>download()</filename> method is called,
BitBake tries to fulfill the URLs by looking for source files
in a specific search order:
<itemizedlist>
<listitem><para><emphasis>Pre-mirror Sites:</emphasis>
BitBake first uses pre-mirrors to try and find source files.
These locations are defined using the
<link linkend='var-PREMIRRORS'><filename>PREMIRRORS</filename></link>
variable.
</para></listitem>
<listitem><para><emphasis>Source URI:</emphasis>
If pre-mirrors fail, BitBake uses the original URL (e.g from
<filename>SRC_URI</filename>).
</para></listitem>
<listitem><para><emphasis>Mirror Sites:</emphasis>
If fetch failures occur, BitBake next uses mirror location as
defined by the
<link linkend='var-MIRRORS'><filename>MIRRORS</filename></link>
variable.
</para></listitem>
</itemizedlist>
</para>
<para>
For each URL passed to the fetcher, the fetcher
calls the submodule that handles that particular URL type.
This behavior can be the source of some confusion when you
are providing URLs for the <filename>SRC_URI</filename>
variable.
Consider the following two URLs:
<literallayout class='monospaced'>
http://git.yoctoproject.org/git/poky;protocol=git
git://git.yoctoproject.org/git/poky;protocol=http
</literallayout>
In the former case, the URL is passed to the
<filename>wget</filename> fetcher, which does not
understand "git".
Therefore, the latter case is the correct form since the
Git fetcher does know how to use HTTP as a transport.
</para>
<para>
Here are some examples that show commonly used mirror
definitions:
<literallayout class='monospaced'>
PREMIRRORS ?= "\
bzr://.*/.* http://somemirror.org/sources/ \n \
cvs://.*/.* http://somemirror.org/sources/ \n \
git://.*/.* http://somemirror.org/sources/ \n \
hg://.*/.* http://somemirror.org/sources/ \n \
osc://.*/.* http://somemirror.org/sources/ \n \
p4://.*/.* http://somemirror.org/sources/ \n \
svn://.*/.* http://somemirror.org/sources/ \n"
MIRRORS =+ "\
ftp://.*/.* http://somemirror.org/sources/ \n \
http://.*/.* http://somemirror.org/sources/ \n \
https://.*/.* http://somemirror.org/sources/ \n"
</literallayout>
It is useful to note that BitBake supports
cross-URLs.
It is possible to mirror a Git repository on an HTTP
server as a tarball.
This is what the <filename>git://</filename> mapping in
the previous example does.
</para>
<para>
Since network accesses are slow, Bitbake maintains a
cache of files downloaded from the network.
Any source files that are not local (i.e.
downloaded from the Internet) are placed into the download
directory, which is specified by the
<link linkend='var-DL_DIR'><filename>DL_DIR</filename></link>
variable.
</para>
<para>
File integrity is of key importance for reproducing builds.
For non-local archive downloads, the fetcher code can verify
sha256 and md5 checksums to ensure the archives have been
downloaded correctly.
You can specify these checksums by using the
<filename>SRC_URI</filename> variable with the appropriate
varflags as follows:
<literallayout class='monospaced'>
SRC_URI[md5sum] = "value"
SRC_URI[sha256sum] = "value"
</literallayout>
You can also specify the checksums as parameters on the
<filename>SRC_URI</filename> as shown below:
<literallayout class='monospaced'>
SRC_URI = "http://example.com/foobar.tar.bz2;md5sum=4a8e0f237e961fd7785d19d07fdb994d"
</literallayout>
If multiple URIs exist, you can specify the checksums either
directly as in the previous example, or you can name the URLs.
The following syntax shows how you name the URIs:
<literallayout class='monospaced'>
SRC_URI = "http://example.com/foobar.tar.bz2;name=foo"
SRC_URI[foo.md5sum] = 4a8e0f237e961fd7785d19d07fdb994d
</literallayout>
After a file has been downloaded and has had its checksum checked,
a ".done" stamp is placed in <filename>DL_DIR</filename>.
BitBake uses this stamp during subsequent builds to avoid
downloading or comparing a checksum for the file again.
<note>
It is assumed that local storage is safe from data corruption.
If this were not the case, there would be bigger issues to worry about.
</note>
</para>
<para>
If
<link linkend='var-BB_STRICT_CHECKSUM'><filename>BB_STRICT_CHECKSUM</filename></link>
is set, any download without a checksum triggers an
error message.
The
<link linkend='var-BB_NO_NETWORK'><filename>BB_NO_NETWORK</filename></link>
variable can be used to make any attempted network access a fatal
error, which is useful for checking that mirrors are complete
as well as other things.
</para>
</section>
<section id='bb-the-unpack'>
<title>The Unpack</title>
<para>
The unpack process usually immediately follows the download.
For all URLs except Git URLs, BitBake uses the common
<filename>unpack</filename> method.
</para>
<para>
A number of parameters exist that you can specify within the
URL to govern the behavior of the unpack stage:
<itemizedlist>
<listitem><para><emphasis>unpack:</emphasis>
Controls whether the URL components are unpacked.
If set to "1", which is the default, the components
are unpacked.
If set to "0", the unpack stage leaves the file alone.
This parameter is useful when you want an archive to be
copied in and not be unpacked.
</para></listitem>
<listitem><para><emphasis>dos:</emphasis>
Applies to <filename>.zip</filename> and
<filename>.jar</filename> files and specifies whether to
use DOS line ending conversion on text files.
</para></listitem>
<listitem><para><emphasis>basepath:</emphasis>
Instructs the unpack stage to strip the specified
directories from the source path when unpacking.
</para></listitem>
<listitem><para><emphasis>subdir:</emphasis>
Unpacks the specific URL to the specified subdirectory
within the root directory.
</para></listitem>
</itemizedlist>
The unpack call automatically decompresses and extracts files
with ".Z", ".z", ".gz", ".xz", ".zip", ".jar", ".ipk", ".rpm".
".srpm", ".deb" and ".bz2" extensions as well as various combinations
of tarball extensions.
</para>
<para>
As mentioned, the Git fetcher has its own unpack method that
is optimized to work with Git trees.
Basically, this method works by cloning the tree into the final
directory.
The process is completed using references so that there is
only one central copy of the Git metadata needed.
</para>
</section>
<section id='bb-fetchers'>
<title>Fetchers</title>
<para>
As mentioned earlier, the URL prefix determines which
fetcher submodule BitBake uses.
Each submodule can support different URL parameters,
which are described in the following sections.
</para>
<section id='local-file-fetcher'>
<title>Local file fetcher (<filename>file://</filename>)</title>
<para>
This submodule handles URLs that begin with
<filename>file://</filename>.
The filename you specify with in the URL can
either be an absolute or relative path to a file.
If the filename is relative, the contents of the
<link linkend='var-FILESPATH'><filename>FILESPATH</filename></link>
variable is used in the same way
<filename>PATH</filename> is used to find executables.
Failing that,
<link linkend='var-FILESDIR'><filename>FILESDIR</filename></link>
is used to find the appropriate relative file.
<note>
<filename>FILESDIR</filename> is deprecated and can
be replaced with <filename>FILESPATH</filename>.
Because <filename>FILESDIR</filename> is likely to be
removed, you should not use this variable in any new code.
</note>
If the file cannot be found, it is assumed that it is available in
<link linkend='var-DL_DIR'><filename>DL_DIR</filename></link>
by the time the <filename>download()</filename> method is called.
</para>
<para>
If you specify a directory, the entire directory is
unpacked.
</para>
<para>
Here are some example URLs:
<literallayout class='monospaced'>
SRC_URI = "file://relativefile.patch"
SRC_URI = "file://relativefile.patch;this=ignored"
SRC_URI = "file:///Users/ich/very_important_software"
</literallayout>
</para>
</section>
<section id='cvs-fetcher'>
<title>CVS fetcher (<filename>(cvs://</filename>)</title>
<para>
This submodule handles checking out files from the
CVS version control system.
You can configure it using a number of different variables:
<itemizedlist>
<listitem><para><emphasis><filename>FETCHCMD_cvs</filename>:</emphasis>
The name of the executable to use when running
the <filename>cvs</filename> command.
This name is usually "cvs".
</para></listitem>
<listitem><para><emphasis><filename>SRCDATE</filename>:</emphasis>
The date to use when fetching the CVS source code.
A special value of "now" causes the checkout to
be updated on every build.
</para></listitem>
<listitem><para><emphasis><filename>CVSDIR</filename>:</emphasis>
Specifies where a temporary checkout is saved.
The location is often <filename>DL_DIR/cvs</filename>.
</para></listitem>
<listitem><para><emphasis><filename>CVS_PROXY_HOST</filename>:</emphasis>
The name to use as a "proxy=" parameter to the
<filename>cvs</filename> command.
</para></listitem>
<listitem><para><emphasis><filename>CVS_PROXY_PORT</filename>:</emphasis>
The port number to use as a "proxyport=" parameter to
the <filename>cvs</filename> command.
</para></listitem>
</itemizedlist>
As well as the standard username and password URL syntax,
you can also configure the fetcher with various URL parameters:
</para>
<para>
The supported parameters are as follows:
<itemizedlist>
<listitem><para><emphasis>"method":</emphasis>
The protocol over which to communicate with the cvs server.
By default, this protocol is "pserver".
If "method" is set to "ext", BitBake examines the
"rsh" parameter and sets <filename>CVS_RSH</filename>.
You can use "dir" for local directories.
</para></listitem>
<listitem><para><emphasis>"module":</emphasis>
Specifies the module to check out.
You must supply this parameter.
</para></listitem>
<listitem><para><emphasis>"tag":</emphasis>
Describes which CVS TAG should be used for
the checkout.
By default, the TAG is empty.
</para></listitem>
<listitem><para><emphasis>"date":</emphasis>
Specifies a date.
If no "date" is specified, the
<link linkend='var-SRCDATE'><filename>SRCDATE</filename></link>
of the configuration is used to checkout a specific date.
The special value of "now" causes the checkout to be
updated on every build.
</para></listitem>
<listitem><para><emphasis>"localdir":</emphasis>
Used to rename the module.
Effectively, you are renaming the output directory
to which the module is unpacked.
You are forcing the module into a special
directory relative to <filename>CVSDIR</filename>.
</para></listitem>
<listitem><para><emphasis>"rsh"</emphasis>
Used in conjunction with the "method" parameter.
</para></listitem>
<listitem><para><emphasis>"scmdata":</emphasis>
Causes the CVS metadata to be maintained in the tarball
the fetcher creates when set to "keep".
The tarball is expanded into the work directory.
By default, the CVS metadata is removed.
</para></listitem>
<listitem><para><emphasis>"fullpath":</emphasis>
Controls whether the resulting checkout is at the
module level, which is the default, or is at deeper
paths.
</para></listitem>
<listitem><para><emphasis>"norecurse":</emphasis>
Causes the fetcher to only checkout the specified
directory with no recurse into any subdirectories.
</para></listitem>
<listitem><para><emphasis>"port":</emphasis>
The port to which the CVS server connects.
</para></listitem>
</itemizedlist>
Some example URLs are as follows:
<literallayout class='monospaced'>
SRC_URI = "cvs://CVSROOT;module=mymodule;tag=some-version;method=ext"
SRC_URI = "cvs://CVSROOT;module=mymodule;date=20060126;localdir=usethat"
</literallayout>
</para>
</section>
<section id='http-ftp-fetcher'>
<title>HTTP/FTP wget fetcher (<filename>http://</filename>, <filename>ftp://</filename>, <filename>https://</filename>)</title>
<para>
This fetcher obtains files from web and FTP servers.
Internally, the fetcher uses the wget utility.
</para>
<para>
The executable and parameters used are specified by the
<filename>FETCHCMD_wget</filename> variable, which defaults
to a sensible values.
The fetcher supports a parameter "downloadfilename" that
allows the name of the downloaded file to be specified.
Specifying the name of the downloaded file is useful
for avoiding collisions in
<link linkend='var-DL_DIR'><filename>DL_DIR</filename></link>
when dealing with multiple files that have the same name.
</para>
<para>
Some example URLs are as follows:
<literallayout class='monospaced'>
SRC_URI = "http://oe.handhelds.org/not_there.aac"
SRC_URI = "ftp://oe.handhelds.org/not_there_as_well.aac"
SRC_URI = "ftp://you@oe.handheld.sorg/home/you/secret.plan"
</literallayout>
</para>
</section>
<section id='svn-fetcher'>
<title>Subversion (SVN) Fetcher (<filename>svn://</filename>)</title>
<para>
This fetcher submodule fetches code from the
Subversion source control system.
The executable used is specified by
<filename>FETCHCMD_svn</filename>, which defaults
to "svn".
The fetcher's temporary working directory is set
by <filename>SVNDIR</filename>, which is usually
<filename>DL_DIR/svn</filename>.
</para>
<para>
The supported parameters are as follows:
<itemizedlist>
<listitem><para><emphasis>"module":</emphasis>
The name of the svn module to checkout.
You must provide this parameter.
You can think of this parameter as the top-level
directory of the repository data you want.
</para></listitem>
<listitem><para><emphasis>"protocol":</emphasis>
The protocol to use, which defaults to "svn".
Other options are "svn+ssh" and "rsh".
For "rsh", the "rsh" parameter is also used.
</para></listitem>
<listitem><para><emphasis>"rev":</emphasis>
The revision of the source code to checkout.
</para></listitem>
<listitem><para><emphasis>"date":</emphasis>
The date of the source code to checkout.
Specific revisions are generally much safer to checkout
rather than by date as they do not involve timezones
(e.g. they are much more deterministic).
</para></listitem>
<listitem><para><emphasis>"scmdata":</emphasis>
Causes the “.svn” directories to be available during
compile-time when set to "keep".
By default, these directories are removed.
</para></listitem>
</itemizedlist>
Following are two examples using svn:
<literallayout class='monospaced'>
SRC_URI = "svn://svn.oe.handhelds.org/svn;module=vip;proto=http;rev=667"
SRC_URI = "svn://svn.oe.handhelds.org/svn/;module=opie;proto=svn+ssh;date=20060126"
</literallayout>
</para>
</section>
<section id='git-fetcher'>
<title>GIT Fetcher (<filename>git://</filename>)</title>
<para>
This fetcher submodule fetches code from the Git
source control system.
The fetcher works by creating a bare clone of the
remote into <filename>GITDIR</filename>, which is
usually <filename>DL_DIR/git</filename>.
This bare clone is then cloned into the work directory during the
unpack stage when a specific tree is checked out.
This is done using alternates and by reference to
minimize the amount of duplicate data on the disk and
make the unpack process fast.
The executable used can be set with
<filename>FETCHCMD_git</filename>.
</para>
<para>
This fetcher supports the following parameters:
<itemizedlist>
<listitem><para><emphasis>"protocol":</emphasis>
The protocol used to fetch the files.
The default is "git" when a hostname is set.
If a hostname is not set, the Git protocol is "file".
You can also use "http", "https", "ssh" and "rsync".
</para></listitem>
<listitem><para><emphasis>"nocheckout":</emphasis>
Tells the fetcher to not checkout source code when
unpacking when set to "1".
Set this option for the URL where there is a custom
routine to checkout code.
The default is "0".
</para></listitem>
<listitem><para><emphasis>"rebaseable":</emphasis>
Indicates that the upstream Git repository can be rebased.
You should set this parameter to "1" if
revisions can become detached from branches.
In this case, the source mirror tarball is done per
revision, which has a loss of efficiency.
Rebasing the upstream Git repository could cause the
current revision to disappear from the upstream repository.
This option reminds the fetcher to preserve the local cache
carefully for future use.
The default value for this parameter is "0".
</para></listitem>
<listitem><para><emphasis>"nobranch":</emphasis>
Tells the fetcher to not check the SHA validation
for the branch when set to "1".
The default is "0".
Set this option for the recipe that refers to
the commit that is valid for a tag instead of
the branch.
</para></listitem>
<listitem><para><emphasis>"bareclone":</emphasis>
Tells the fetcher to clone a bare clone into the
destination directory without checking out a working tree.
Only the raw Git metadata is provided.
This parameter implies the "nocheckout" parameter as well.
</para></listitem>
<listitem><para><emphasis>"branch":</emphasis>
The branch(es) of the Git tree to clone.
If unset, this is assumed to be "master".
The number of branch parameters much match the number of
name parameters.
</para></listitem>
<listitem><para><emphasis>"rev":</emphasis>
The revision to use for the checkout.
The default is "master".
</para></listitem>
<listitem><para><emphasis>"tag":</emphasis>
Specifies a tag to use for the checkout.
To correctly resolve tags, BitBake must access the
network.
For that reason, tags are often not used.
As far as Git is concerned, the "tag" parameter behaves
effectively the same as the "revision" parameter.
</para></listitem>
<listitem><para><emphasis>"subpath":</emphasis>
Limits the checkout to a specific subpath of the tree.
By default, the whole tree is checked out.
</para></listitem>
<listitem><para><emphasis>"destsuffix":</emphasis>
The name of the path in which to place the checkout.
By default, the path is <filename>git/</filename>.
</para></listitem>
</itemizedlist>
Here are some example URLs:
<literallayout class='monospaced'>
SRC_URI = "git://git.oe.handhelds.org/git/vip.git;tag=version-1"
SRC_URI = "git://git.oe.handhelds.org/git/vip.git;protocol=http"
</literallayout>
</para>
</section>
<section id='other-fetchers'>
<title>Other Fetchers</title>
<para>
Fetch submodules also exist for the following:
<itemizedlist>
<listitem><para>
Bazaar (<filename>bzr://</filename>)
</para></listitem>
<listitem><para>
Perforce (<filename>p4://</filename>)
</para></listitem>
<listitem><para>
Git Submodules (<filename>gitsm://</filename>)
</para></listitem>
<listitem><para>
Trees using Git Annex (<filename>gitannex://</filename>)
</para></listitem>
<listitem><para>
Secure FTP (<filename>sftp://</filename>)
</para></listitem>
<listitem><para>
Secure Shell (<filename>ssh://</filename>)
</para></listitem>
<listitem><para>
Repo (<filename>repo://</filename>)
</para></listitem>
<listitem><para>
OSC (<filename>osc://</filename>)
</para></listitem>
<listitem><para>
Mercurial (<filename>hg://</filename>)
</para></listitem>
</itemizedlist>
No documentation currently exists for these lesser used
fetcher submodules.
However, you might find the code helpful and readable.
</para>
</section>
</section>
<section id='auto-revisions'>
<title>Auto Revisions</title>
<para>
We need to document <filename>AUTOREV</filename> and
<filename>SRCREV_FORMAT</filename> here.
</para>
</section>
</chapter>

View File

@@ -1,506 +0,0 @@
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<appendix id='hello-world-example'>
<title>Hello World Example</title>
<section id='bitbake-hello-world'>
<title>BitBake Hello World</title>
<para>
The simplest example commonly used to demonstrate any new
programming language or tool is the
"<ulink url="http://en.wikipedia.org/wiki/Hello_world_program">Hello World</ulink>"
example.
This appendix demonstrates, in tutorial form, Hello
World within the context of BitBake.
The tutorial describes how to create a new project
and the applicable metadata files necessary to allow
BitBake to build it.
</para>
</section>
<section id='example-obtaining-bitbake'>
<title>Obtaining BitBake</title>
<para>
See the
"<link linkend='obtaining-bitbake'>Obtaining BitBake</link>"
section for information on how to obtain BitBake.
Once you have the source code on your machine, the BitBake directory
appears as follows:
<literallayout class='monospaced'>
$ ls -al
total 100
drwxrwxr-x. 9 wmat wmat 4096 Jan 31 13:44 .
drwxrwxr-x. 3 wmat wmat 4096 Feb 4 10:45 ..
-rw-rw-r--. 1 wmat wmat 365 Nov 26 04:55 AUTHORS
drwxrwxr-x. 2 wmat wmat 4096 Nov 26 04:55 bin
drwxrwxr-x. 4 wmat wmat 4096 Jan 31 13:44 build
-rw-rw-r--. 1 wmat wmat 16501 Nov 26 04:55 ChangeLog
drwxrwxr-x. 2 wmat wmat 4096 Nov 26 04:55 classes
drwxrwxr-x. 2 wmat wmat 4096 Nov 26 04:55 conf
drwxrwxr-x. 3 wmat wmat 4096 Nov 26 04:55 contrib
-rw-rw-r--. 1 wmat wmat 17987 Nov 26 04:55 COPYING
drwxrwxr-x. 3 wmat wmat 4096 Nov 26 04:55 doc
-rw-rw-r--. 1 wmat wmat 69 Nov 26 04:55 .gitignore
-rw-rw-r--. 1 wmat wmat 849 Nov 26 04:55 HEADER
drwxrwxr-x. 5 wmat wmat 4096 Jan 31 13:44 lib
-rw-rw-r--. 1 wmat wmat 195 Nov 26 04:55 MANIFEST.in
-rwxrwxr-x. 1 wmat wmat 3195 Jan 31 11:57 setup.py
-rw-rw-r--. 1 wmat wmat 2887 Nov 26 04:55 TODO
</literallayout>
</para>
<para>
At this point, you should have BitBake cloned to
a directory that matches the previous listing except for
dates and user names.
</para>
</section>
<section id='setting-up-the-bitbake-environment'>
<title>Setting Up the BitBake Environment</title>
<para>
First, you need to be sure that you can run BitBake.
Set your working directory to where your local BitBake
files are and run the following command:
<literallayout class='monospaced'>
$ ./bin/bitbake --version
BitBake Build Tool Core version 1.23.0, bitbake version 1.23.0
</literallayout>
The console output tells you what version you are running.
</para>
<para>
The recommended method to run BitBake is from a directory of your
choice.
To be able to run BitBake from any directory, you need to add the
executable binary to your binary to your shell's environment
<filename>PATH</filename> variable.
First, look at your current <filename>PATH</filename> variable
by entering the following:
<literallayout class='monospaced'>
$ echo $PATH
</literallayout>
Next, add the directory location for the BitBake binary to the
<filename>PATH</filename>.
Here is an example that adds the
<filename>/home/scott-lenovo/bitbake/bin</filename> directory
to the front of the <filename>PATH</filename> variable:
<literallayout class='monospaced'>
$ export PATH=/home/scott-lenovo/bitbake/bin:$PATH
</literallayout>
You should now be able to enter the <filename>bitbake</filename>
command from the command line while working from any directory.
</para>
</section>
<section id='the-hello-world-example'>
<title>The Hello World Example</title>
<para>
The overall goal of this exercise is to build a
complete "Hello World" example utilizing task and layer
concepts.
Because this is how modern projects such as OpenEmbedded and
the Yocto Project utilize BitBake, the example
provides an excellent starting point for understanding
BitBake.
</para>
<para>
To help you understand how to use BitBake to build targets,
the example starts with nothing but the <filename>bitbake</filename>
command, which causes BitBake to fail and report problems.
The example progresses by adding pieces to the build to
eventually conclude with a working, minimal "Hello World"
example.
</para>
<para>
While every attempt is made to explain what is happening during
the example, the descriptions cannot cover everything.
You can find further information throughout this manual.
Also, you can actively participate in the
<ulink url='http://lists.openembedded.org/mailman/listinfo/bitbake-devel'></ulink>
discussion mailing list about the BitBake build tool.
</para>
<note>
This example was inspired by and drew heavily from these sources:
<itemizedlist>
<listitem><para>
<ulink url="http://www.mail-archive.com/yocto@yoctoproject.org/msg09379.html">Mailing List post - The BitBake equivalent of "Hello, World!"</ulink>
</para></listitem>
<listitem><para>
<ulink url="http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/">Hambedded Linux blog post - From Bitbake Hello World to an Image</ulink>
</para></listitem>
</itemizedlist>
</note>
<para>
As stated earlier, the goal of this example
is to eventually compile "Hello World".
However, it is unknown what BitBake needs and what you have
to provide in order to achieve that goal.
Recall that BitBake utilizes three types of metadata files:
<link linkend='configuration-files'>Configuration Files</link>,
<link linkend='classes'>Classes</link>, and
<link linkend='recipes'>Recipes</link>.
But where do they go?
How does BitBake find them?
BitBake's error messaging helps you answer these types of questions
and helps you better understand exactly what is going on.
</para>
<para>
Following is the complete "Hello World" example.
</para>
<orderedlist>
<listitem><para><emphasis>Create a Project Directory:</emphasis>
First, set up a directory for the "Hello World" project.
Here is how you can do so in your home directory:
<literallayout class='monospaced'>
$ mkdir ~/hello
$ cd ~/hello
</literallayout>
This is the directory that BitBake will use to do all of
its work.
You can use this directory to keep all the metafiles needed
by BitBake.
Having a project directory is a good way to isolate your
project.
</para></listitem>
<listitem><para><emphasis>Run Bitbake:</emphasis>
At this point, you have nothing but a project directory.
Run the <filename>bitbake</filename> command and see what
it does:
<literallayout class='monospaced'>
$ bitbake
The BBPATH variable is not set and bitbake did not
find a conf/bblayers.conf file in the expected location.
Maybe you accidentally invoked bitbake from the wrong directory?
DEBUG: Removed the following variables from the environment:
GNOME_DESKTOP_SESSION_ID, XDG_CURRENT_DESKTOP,
GNOME_KEYRING_CONTROL, DISPLAY, SSH_AGENT_PID, LANG, no_proxy,
XDG_SESSION_PATH, XAUTHORITY, SESSION_MANAGER, SHLVL,
MANDATORY_PATH, COMPIZ_CONFIG_PROFILE, WINDOWID, EDITOR,
GPG_AGENT_INFO, SSH_AUTH_SOCK, GDMSESSION, GNOME_KEYRING_PID,
XDG_SEAT_PATH, XDG_CONFIG_DIRS, LESSOPEN, DBUS_SESSION_BUS_ADDRESS,
_, XDG_SESSION_COOKIE, DESKTOP_SESSION, LESSCLOSE, DEFAULTS_PATH,
UBUNTU_MENUPROXY, OLDPWD, XDG_DATA_DIRS, COLORTERM, LS_COLORS
</literallayout>
The majority of this output is specific to environment variables
that are not directly relevant to BitBake.
However, the very first message regarding the
<filename>BBPATH</filename> variable and the
<filename>conf/bblayers.conf</filename> file
is relevant.</para>
<para>
When you run BitBake, it begins looking for metadata files.
The
<link linkend='var-BBPATH'><filename>BBPATH</filename></link>
variable is what tells BitBake where to look for those files.
<filename>BBPATH</filename> is not set and you need to set it.
Without <filename>BBPATH</filename>, Bitbake cannot
find any configuration files (<filename>.conf</filename>)
or recipe files (<filename>.bb</filename>) at all.
BitBake also cannot find the <filename>bitbake.conf</filename>
file.
</para></listitem>
<listitem><para><emphasis>Setting <filename>BBPATH</filename>:</emphasis>
For this example, you can set <filename>BBPATH</filename>
in the same manner that you set <filename>PATH</filename>
earlier in the appendix.
You should realize, though, that it is much more flexible to set the
<filename>BBPATH</filename> variable up in a configuration
file for each project.</para>
<para>From your shell, enter the following commands to set and
export the <filename>BBPATH</filename> variable:
<literallayout class='monospaced'>
$ BBPATH="&lt;projectdirectory&gt;"
$ export BBPATH
</literallayout>
Use your actual project directory in the command.
BitBake uses that directory to find the metadata it needs for
your project.
<note>
When specifying your project directory, do not use the
tilde ("~") character as BitBake does not expand that character
as the shell would.
</note>
</para></listitem>
<listitem><para><emphasis>Run Bitbake:</emphasis>
Now that you have <filename>BBPATH</filename> defined, run
the <filename>bitbake</filename> command again:
<literallayout class='monospaced'>
$ bitbake
ERROR: Traceback (most recent call last):
File "/home/scott-lenovo/bitbake/lib/bb/cookerdata.py", line 163, in wrapped
return func(fn, *args)
File "/home/scott-lenovo/bitbake/lib/bb/cookerdata.py", line 173, in parse_config_file
return bb.parse.handle(fn, data, include)
File "/home/scott-lenovo/bitbake/lib/bb/parse/__init__.py", line 99, in handle
return h['handle'](fn, data, include)
File "/home/scott-lenovo/bitbake/lib/bb/parse/parse_py/ConfHandler.py", line 120, in handle
abs_fn = resolve_file(fn, data)
File "/home/scott-lenovo/bitbake/lib/bb/parse/__init__.py", line 117, in resolve_file
raise IOError("file %s not found in %s" % (fn, bbpath))
IOError: file conf/bitbake.conf not found in /home/scott-lenovo/hello
ERROR: Unable to parse conf/bitbake.conf: file conf/bitbake.conf not found in /home/scott-lenovo/hello
</literallayout>
This sample output shows that BitBake could not find the
<filename>conf/bitbake.conf</filename> file in the project
directory.
This file is the first thing BitBake must find in order
to build a target.
And, since the project directory for this example is
empty, you need to provide a <filename>conf/bitbake.conf</filename>
file.
</para></listitem>
<listitem><para><emphasis>Creating <filename>conf/bitbake.conf</filename>:</emphasis>
The <filename>conf/bitbake.conf</filename> includes a number of
configuration variables BitBake uses for metadata and recipe
files.
For this example, you need to create the file in your project directory
and define some key BitBake variables.
For more information on the <filename>bitbake.conf</filename>,
see
<ulink url='http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/#an-overview-of-bitbakeconf'></ulink>
</para>
<para>Use the following commands to create the <filename>conf</filename>
directory in the project directory:
<literallayout class='monospaced'>
$ mkdir conf
</literallayout>
From within the <filename>conf</filename> directory, use
some editor to create the <filename>bitbake.conf</filename>
so that it contains the following:
<literallayout class='monospaced'>
TMPDIR = "${<link linkend='var-TOPDIR'>TOPDIR</link>}/tmp"
<link linkend='var-CACHE'>CACHE</link> = "${TMPDIR}/cache"
<link linkend='var-STAMP'>STAMP</link> = "${TMPDIR}/stamps"
<link linkend='var-T'>T</link> = "${TMPDIR}/work"
<link linkend='var-B'>B</link> = "${TMPDIR}"
</literallayout>
The <filename>TMPDIR</filename> variable establishes a directory
that BitBake uses for build output and intermediate files (other
than the cached information used by the
<link linkend='setscene'>Setscene</link> process.
Here, the <filename>TMPDIR</filename> directory is set to
<filename>hello/tmp</filename>.
<note><title>Tip</title>
You can always safely delete the <filename>tmp</filename>
directory in order to rebuild a BitBake target.
The build process creates the directory for you
when you run BitBake.
</note></para>
<para>For information about each of the other variables defined in this
example, click on the links to take you to the definitions in
the glossary.
</para></listitem>
<listitem><para><emphasis>Run Bitbake:</emphasis>
After making sure that the <filename>conf/bitbake.conf</filename>
file exists, you can run the <filename>bitbake</filename>
command again:
<literallayout class='monospaced'>
$ bitbake
ERROR: Traceback (most recent call last):
File "/home/scott-lenovo/bitbake/lib/bb/cookerdata.py", line 163, in wrapped
return func(fn, *args)
File "/home/scott-lenovo/bitbake/lib/bb/cookerdata.py", line 177, in _inherit
bb.parse.BBHandler.inherit(bbclass, "configuration INHERITs", 0, data)
File "/home/scott-lenovo/bitbake/lib/bb/parse/parse_py/BBHandler.py", line 92, in inherit
include(fn, file, lineno, d, "inherit")
File "/home/scott-lenovo/bitbake/lib/bb/parse/parse_py/ConfHandler.py", line 100, in include
raise ParseError("Could not %(error_out)s file %(fn)s" % vars(), oldfn, lineno)
ParseError: ParseError in configuration INHERITs: Could not inherit file classes/base.bbclass
ERROR: Unable to parse base: ParseError in configuration INHERITs: Could not inherit file classes/base.bbclass
</literallayout>
In the sample output, BitBake could not find the
<filename>classes/base.bbclass</filename> file.
You need to create that file next.
</para></listitem>
<listitem><para><emphasis>Creating <filename>classes/base.bbclass</filename>:</emphasis>
BitBake uses class files to provide common code and functionality.
The minimally required class for BitBake is the
<filename>classes/base.bbclass</filename> file.
The <filename>base</filename> class is implicitly inherited by
every recipe.
BitBake looks for the class in the <filename>classes</filename>
directory of the project (i.e <filename>hello/classes</filename>
in this example).
</para>
<para>Create the <filename>classes</filename> directory as follows:
<literallayout class='monospaced'>
$ cd $HOME/hello
$ mkdir classes
</literallayout>
Move to the <filename>classes</filename> directory and then
create the <filename>base.bbclass</filename> file by inserting
this single line:
<literallayout class='monospaced'>
addtask build
</literallayout>
The minimal task that BitBake runs is the
<filename>do_build</filename> task.
This is all the example needs in order to build the project.
Of course, the <filename>base.bbclass</filename> can have much
more depending on which build environments BitBake is
supporting.
For more information on the <filename>base.bbclass</filename> file,
you can look at
<ulink url='http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/#tasks'></ulink>.
</para></listitem>
<listitem><para><emphasis>Run Bitbake:</emphasis>
After making sure that the <filename>classes/base.bbclass</filename>
file exists, you can run the <filename>bitbake</filename>
command again:
<literallayout class='monospaced'>
$ bitbake
Nothing to do. Use 'bitbake world' to build everything, or run 'bitbake --help' for usage information.
</literallayout>
BitBake is finally reporting no errors.
However, you can see that it really does not have anything
to do.
You need to create a recipe that gives BitBake something to do.
</para></listitem>
<listitem><para><emphasis>Creating a Layer:</emphasis>
While it is not really necessary for such a small example,
it is good practice to create a layer in which to keep your
code separate from the general metadata used by BitBake.
Thus, this example creates and uses a layer called "mylayer".
<note>
You can find additional information on adding a layer at
<ulink url='http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/#adding-an-example-layer'></ulink>.
</note>
</para>
<para>Minimally, you need a recipe file and a layer configuration
file in your layer.
The configuration file needs to be in the <filename>conf</filename>
directory inside the layer.
Use these commands to set up the layer and the <filename>conf</filename>
directory:
<literallayout class='monospaced'>
$ cd $HOME
$ mkdir mylayer
$ cd mylayer
$ mkdir conf
</literallayout>
Move to the <filename>conf</filename> directory and create a
<filename>layer.conf</filename> file that has the following:
<literallayout class='monospaced'>
BBPATH .= ":${<link linkend='var-LAYERDIR'>LAYERDIR</link>}"
<link linkend='var-BBFILES'>BBFILES</link> += "${LAYERDIR}/*.bb"
<link linkend='var-BBFILE_COLLECTIONS'>BBFILE_COLLECTIONS</link> += "mylayer"
<link linkend='var-BBFILE_PATTERN'>BBFILE_PATTERN_mylayer</link> := "^${LAYERDIR}/"
</literallayout>
For information on these variables, click the links
to go to the definitions in the glossary.</para>
<para>You need to create the recipe file next.
Inside your layer at the top-level, use an editor and create
a recipe file named <filename>printhello.bb</filename> that
has the following:
<literallayout class='monospaced'>
<link linkend='var-DESCRIPTION'>DESCRIPTION</link> = "Prints Hello World"
<link linkend='var-PN'>PN</link> = 'printhello'
<link linkend='var-PV'>PV</link> = '1'
python do_build() {
bb.plain("********************");
bb.plain("* *");
bb.plain("* Hello, World! *");
bb.plain("* *");
bb.plain("********************");
}
</literallayout>
The recipe file simply provides a description of the
recipe, the name, version, and the <filename>do_build</filename>
task, which prints out "Hello World" to the console.
For more information on these variables, follow the links
to the glossary.
</para></listitem>
<listitem><para><emphasis>Run Bitbake With a Target:</emphasis>
Now that a BitBake target exists, run the command and provide
that target:
<literallayout class='monospaced'>
$ cd $HOME/hello
$ bitbake printhello
ERROR: no recipe files to build, check your BBPATH and BBFILES?
Summary: There was 1 ERROR message shown, returning a non-zero exit code.
</literallayout>
We have created the layer with the recipe and the layer
configuration file but it still seems that BitBake cannot
find the recipe.
BitBake needs a <filename>conf/bblayers.conf</filename> that
lists the layers for the project.
Without this file, BitBake cannot find the recipe.
</para></listitem>
<listitem><para><emphasis>Creating <filename>conf/bblayers.conf</filename>:</emphasis>
BitBake uses the <filename>conf/bblayers.conf</filename> file
to locate layers needed for the project.
This file must reside in the <filename>conf</filename> directory
of the project (i.e. <filename>hello/conf</filename> for this
example).</para>
<para>Set your working directory to the <filename>hello/conf</filename>
directory and then create the <filename>bblayers.conf</filename>
file so that it contains the following:
<literallayout class='monospaced'>
BBLAYERS ?= " \
/home/&lt;you&gt;/mylayer \
"
</literallayout>
You need to provide your own information for
<filename>you</filename> in the file.
</para></listitem>
<listitem><para><emphasis>Run Bitbake With a Target:</emphasis>
Now that you have supplied the <filename>bblayers.conf</filename>
file, run the <filename>bitbake</filename> command and provide
the target:
<literallayout class='monospaced'>
$ bitbake printhello
Parsing recipes: 100% |##################################################################################|
Time: 00:00:00
Parsing of 1 .bb files complete (0 cached, 1 parsed). 1 targets, 0 skipped, 0 masked, 0 errors.
NOTE: Resolving any missing task queue dependencies
NOTE: Preparing runqueue
NOTE: Executing RunQueue Tasks
********************
* *
* Hello, World! *
* *
********************
NOTE: Tasks Summary: Attempted 1 tasks of which 0 didn't need to be rerun and all succeeded.
</literallayout>
BitBake finds the <filename>printhello</filename> recipe and
successfully runs the task.
<note>
After the first execution, re-running
<filename>bitbake printhello</filename> again will not
result in a BitBake run that prints the same console
output.
The reason for this is that the first time the
<filename>printhello.bb</filename> recipe's
<filename>do_build</filename> task executes
successfully, BitBake writes a stamp file for the task.
Thus, the next time you attempt to run the task
using that same <filename>bitbake</filename> command,
BitBake notices the stamp and therefore determines
that the task does not need to be re-run.
If you delete the <filename>tmp</filename> directory
or run <filename>bitbake -c clean printhello</filename>
and then re-run the build, the "Hello, World!" message will
be printed again.
</note>
</para></listitem>
</orderedlist>
</section>
</appendix>

View File

@@ -1,644 +0,0 @@
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<chapter id="bitbake-user-manual-intro">
<title>Overview</title>
<para>
Welcome to the BitBake User Manual.
This manual provides information on the BitBake tool.
The information attempts to be as independent as possible regarding
systems that use BitBake, such as the Yocto Project and
OpenEmbedded.
In some cases, scenarios or examples that within the context of
a build system are used in the manual to help with understanding.
For these cases, the manual clearly states the context.
</para>
<section id="intro">
<title>Introduction</title>
<para>
Fundamentally, BitBake is a generic task execution
engine that allows shell and Python tasks to be run
efficiently and in parallel while working within
complex inter-task dependency constraints.
One of BitBake's main users, OpenEmbedded, takes this core
and builds embedded Linux software stacks using
a task-oriented approach.
</para>
<para>
Conceptually, BitBake is similar to GNU Make in
some regards but has significant differences:
<itemizedlist>
<listitem><para>
BitBake executes tasks according to provided
metadata that builds up the tasks.
Metadata is stored in recipe (<filename>.bb</filename>),
configuration (<filename>.conf</filename>), and class
(<filename>.bbclass</filename>) files and provides
BitBake with instructions on what tasks to run and
the dependencies between those tasks.
</para></listitem>
<listitem><para>
BitBake includes a fetcher library for obtaining source
code from various places such as source control
systems or websites.
</para></listitem>
<listitem><para>
The instructions for each unit to be built (e.g. a piece
of software) are known as recipe files and
contain all the information about the unit
(dependencies, source file locations, checksums, description
and so on).
</para></listitem>
<listitem><para>
BitBake includes a client/server abstraction and can
be used from a command line or used as a service over XMLRPC and
has several different user interfaces.
</para></listitem>
</itemizedlist>
</para>
</section>
<section id="history-and-goals">
<title>History and Goals</title>
<para>
BitBake was originally a part of the OpenEmbedded project.
It was inspired by the Portage package management system
used by the Gentoo Linux distribution.
On December 7, 2004, OpenEmbedded project team member,
Chris Larson split the project into two distinct pieces:
<itemizedlist>
<listitem><para>BitBake, a generic task executor</para></listitem>
<listitem><para>OpenEmbedded, a metadata set utilized by
BitBake</para></listitem>
</itemizedlist>
Today, BitBake is the primary basis of the
<ulink url="http://www.openembedded.org/">OpenEmbedded</ulink>
project, which is being used to build and maintain Linux
distributions such as the Angstrom Distribution and which is used
as the build tool for Linux projects such as the Yocto Project.
</para>
<para>
Prior to BitBake, no other build tool adequately met the needs of
an aspiring embedded Linux distribution.
All of the build systems used by traditional desktop Linux
distributions lacked important functionality, and none of the
ad-hoc Buildroot-based systems, prevalent in the
embedded space, were scalable or maintainable.
</para>
<para>
Some important original goals for BitBake were:
<itemizedlist>
<listitem><para>
Handle cross-compilation.
</para></listitem>
<listitem><para>
Handle inter-package dependencies (build time on
target architecture, build time on native
architecture, and runtime).
</para></listitem>
<listitem><para>
Support running any number of tasks within a given
package, including, but not limited to, fetching
upstream sources, unpacking them, patching them,
configuring them, and so forth.
</para></listitem>
<listitem><para>
Be Linux distribution agnostic for both build and
target systems.
</para></listitem>
<listitem><para>
Be architecture agnostic.
</para></listitem>
<listitem><para>
Support multiple build and target operating systems
(e.g. Cygwin, the BSDs, and so forth).
</para></listitem>
<listitem><para>
Be self contained, rather than tightly
integrated into the build machine's root
filesystem.
</para></listitem>
<listitem><para>
Handle conditional metadata on the target architecture,
operating system, distribution, and machine.
</para></listitem>
<listitem><para>
Be easy to use the tools to supply local metadata and packages
against which to operate.
</para></listitem>
<listitem><para>
Be easy to use BitBake to collaborate between multiple
projects for their builds.
</para></listitem>
<listitem><para>
Provide an inheritance mechanism that share
common metadata between many packages.
</para></listitem>
</itemizedlist>
Over time it became apparent that some further requirements
were necessary:
<itemizedlist>
<listitem><para>
Handle variants of a base recipe (e.g. native, sdk,
and multilib).
</para></listitem>
<listitem><para>
Split metadata into layers and allow layers
to override each other.
</para></listitem>
<listitem><para>
Allow representation of a given set of input variables
to a task as a checksum.
Based on that checksum, allow acceleration of builds
with prebuilt components.
</para></listitem>
</itemizedlist>
BitBake satisfies all the original requirements and many more
with extensions being made to the basic functionality to
reflect the additional requirements.
Flexibility and power have always been the priorities.
BitBake is highly extensible and supports embedded Python code and
execution of any arbitrary tasks.
</para>
</section>
<section id="Concepts">
<title>Concepts</title>
<para>
BitBake is a program written in the Python language.
At the highest level, BitBake interprets metadata, decides
what tasks are required to run, and executes those tasks.
Similar to GNU Make, BitBake controls how software is
built.
GNU Make achieves its control through "makefiles".
BitBake uses "recipes".
</para>
<para>
BitBake extends the capabilities of a simple
tool like GNU Make by allowing for much more complex tasks
to be completed, such as assembling entire embedded Linux
distributions.
</para>
<para>
The remainder of this section introduces several concepts
that should be understood in order to better leverage
the power of BitBake.
</para>
<section id='recipes'>
<title>Recipes</title>
<para>
BitBake Recipes, which are denoted by the file extension
<filename>.bb</filename>, are the most basic metadata files.
These recipe files provide BitBake with the following:
<itemizedlist>
<listitem><para>Descriptive information about the package</para></listitem>
<listitem><para>The version of the recipe</para></listitem>
<listitem><para>Existing Dependencies</para></listitem>
<listitem><para>Where the source code resides</para></listitem>
<listitem><para>Whether the source code requires any patches</para></listitem>
<listitem><para>How to compile the source code</para></listitem>
<listitem><para>Where on the target machine to install the
package being compiled</para></listitem>
</itemizedlist>
</para>
<para>
Within the context of BitBake, or any project utilizing BitBake
as its build system, files with the <filename>.bb</filename>
extension are referred to as recipes.
<note>
The term "package" is also commonly used to describe recipes.
However, since the same word is used to describe packaged
output from a project, it is best to maintain a single
descriptive term, "recipes".
</note>
</para>
</section>
<section id='configuration-files'>
<title>Configuration Files</title>
<para>
Configuration files, which are denoted by the
<filename>.conf</filename> extension, define
various configuration variables that govern the project's build
process.
These files fall into several areas that define
machine configuration options, distribution configuration
options, compiler tuning options, general common
configuration options, and user configuration options.
The main configuration file is the sample
<filename>bitbake.conf</filename> file, which is
located within the BitBake source tree
<filename>conf</filename> directory.
</para>
</section>
<section id='classes'>
<title>Classes</title>
<para>
Class files, which are denoted by the
<filename>.bbclass</filename> extension, contain
information that is useful to share between metadata files.
The BitBake source tree currently comes with one class metadata file
called <filename>base.bbclass</filename>.
You can find this file in the
<filename>classes</filename> directory.
The <filename>base.bbclass</filename> is special since it
is always included automatically for all recipes
and classes.
This class contains definitions for standard basic tasks such
as fetching, unpacking, configuring (empty by default),
compiling (runs any Makefile present), installing (empty by
default) and packaging (empty by default).
These tasks are often overridden or extended by other classes
added during the project development process.
</para>
</section>
<section id='layers'>
<title>Layers</title>
<para>
Layers allow you to isolate different types of
customizations from each other.
While you might find it tempting to keep everything in one layer
when working on a single project, the more modular you organize
your metadata, the easier it is to cope with future changes.
</para>
<para>
To illustrate how you can use layers to keep things modular,
consider customizations you might make to support a specific target machine.
These types of customizations typically reside in a special layer,
rather than a general layer, called a Board Specific Package (BSP) Layer.
Furthermore, the machine customizations should be isolated from
recipes and metadata that support a new GUI environment, for
example.
This situation gives you a couple of layers: one for the machine
configurations and one for the GUI environment.
It is important to understand, however, that the BSP layer can still
make machine-specific additions to recipes within
the GUI environment layer without polluting the GUI layer itself
with those machine-specific changes.
You can accomplish this through a recipe that is a BitBake append
(<filename>.bbappend</filename>) file.
</para>
</section>
<section id='append-bbappend-files'>
<title>Append Files</title>
<para>
Append files, which are files that have the
<filename>.bbappend</filename> file extension, add or
extend build information to an existing
recipe file.
</para>
<para>
BitBake expects every append file to have a corresponding recipe file.
Furthermore, the append file and corresponding recipe file
must use the same root filename.
The filenames can differ only in the file type suffix used
(e.g. <filename>formfactor_0.0.bb</filename> and
<filename>formfactor_0.0.bbappend</filename>).
</para>
<para>
Information in append files overrides the information in the
similarly-named recipe file.
</para>
<para>
When you name an append file, you can use the
wildcard character (%) to allow for matching recipe names.
For example, suppose you have an append file named
as follows:
<literallayout class='monospaced'>
busybox_1.21.%.bbappend
</literallayout>
That append file would match any <filename>busybox_1.21.x.bb</filename>
version of the recipe.
So, the append file would match the following recipe names:
<literallayout class='monospaced'>
busybox_1.21.1.bb
busybox_1.21.2.bb
busybox_1.21.3.bb
</literallayout>
If the <filename>busybox</filename> recipe was updated to
<filename>busybox_1.3.0.bb</filename>, the append name would not
match.
However, if you named the append file
<filename>busybox_1.%.bbappend</filename>, then you would have a match.
</para>
</section>
</section>
<section id='obtaining-bitbake'>
<title>Obtaining BitBake</title>
<para>
You can obtain BitBake several different ways:
<itemizedlist>
<listitem><para><emphasis>Cloning BitBake:</emphasis>
Using Git to clone the BitBake source code repository
is the recommended method for obtaining BitBake.
Cloning the repository makes it easy to get bug fixes
and have access to stable branches and the master
branch.
Once you have cloned BitBake, you should use
the latest stable
branch for development since the master branch is for
BitBake development and might contain less stable changes.
</para>
<para>You usually need a version of BitBake
that matches the metadata you are using.
The metadata is generally backwards compatible but
not forward compatible.</para>
<para>Here is an example that clones the BitBake repository:
<literallayout class='monospaced'>
$ git clone git://git.openembedded.org/bitbake
</literallayout>
This command clones the BitBake Git repository into a
directory called <filename>bitbake</filename>.
Alternatively, you can
designate a directory after the
<filename>git clone</filename> command
if you want to call the new directory something
other than <filename>bitbake</filename>.
Here is an example that names the directory
<filename>bbdev</filename>:
<literallayout class='monospaced'>
$ git clone git://git.openembedded.org/bitbake bbdev
</literallayout></para></listitem>
<listitem><para><emphasis>Installation using your Distribution
Package Management System:</emphasis>
This method is not
recommended because the BitBake version that is
provided by your distribution, in most cases,
is several
releases behind a snapshot of the BitBake repository.
</para></listitem>
<listitem><para><emphasis>Taking a snapshot of BitBake:</emphasis>
Downloading a snapshot of BitBake from the
source code repository gives you access to a known
branch or release of BitBake.
<note>
Cloning the Git repository, as described earlier,
is the preferred method for getting BitBake.
Cloning the repository makes it easier to update as
patches are added to the stable branches.
</note></para>
<para>The following example downloads a snapshot of
BitBake version 1.17.0:
<literallayout class='monospaced'>
$ wget http://git.openembedded.org/bitbake/snapshot/bitbake-1.17.0.tar.gz
$ tar zxpvf bitbake-1.17.0.tar.gz
</literallayout>
After extraction of the tarball using the tar utility,
you have a directory entitled
<filename>bitbake-1.17.0</filename>.
</para></listitem>
</itemizedlist>
</para>
</section>
<section id="bitbake-user-manual-command">
<title>The BitBake Command</title>
<para>
The <filename>bitbake</filename> command is the primary interface
to the BitBake tool.
This section presents the BitBake command syntax and provides
several execution examples.
</para>
<section id='usage-and-syntax'>
<title>Usage and syntax</title>
<para>
Following is the usage and syntax for BitBake:
<literallayout class='monospaced'>
$ bitbake -h
Usage: bitbake [options] [recipename/target ...]
Executes the specified task (default is 'build') for a given set of target recipes (.bb files).
It is assumed there is a conf/bblayers.conf available in cwd or in BBPATH which
will provide the layer, BBFILES and other configuration information.
Options:
--version show program's version number and exit
-h, --help show this help message and exit
-b BUILDFILE, --buildfile=BUILDFILE
Execute tasks from a specific .bb recipe directly.
WARNING: Does not handle any dependencies from other
recipes.
-k, --continue Continue as much as possible after an error. While the
target that failed and anything depending on it cannot
be built, as much as possible will be built before
stopping.
-a, --tryaltconfigs Continue with builds by trying to use alternative
providers where possible.
-f, --force Force the specified targets/task to run (invalidating
any existing stamp file).
-c CMD, --cmd=CMD Specify the task to execute. The exact options
available depend on the metadata. Some examples might
be 'compile' or 'populate_sysroot' or 'listtasks' may
give a list of the tasks available.
-C INVALIDATE_STAMP, --clear-stamp=INVALIDATE_STAMP
Invalidate the stamp for the specified task such as
'compile' and then run the default task for the
specified target(s).
-r PREFILE, --read=PREFILE
Read the specified file before bitbake.conf.
-R POSTFILE, --postread=POSTFILE
Read the specified file after bitbake.conf.
-v, --verbose Output more log message data to the terminal.
-D, --debug Increase the debug level. You can specify this more
than once.
-n, --dry-run Don't execute, just go through the motions.
-S DUMP_SIGNATURES, --dump-signatures=DUMP_SIGNATURES
Dump out the signature construction information, with
no task execution. Parameters are passed to the
signature handling code, use 'none' if no specific
handler is required.
-p, --parse-only Quit after parsing the BB recipes.
-s, --show-versions Show current and preferred versions of all recipes.
-e, --environment Show the global or per-package environment complete
with information about where variables were
set/changed.
-g, --graphviz Save dependency tree information for the specified
targets in the dot syntax.
-I EXTRA_ASSUME_PROVIDED, --ignore-deps=EXTRA_ASSUME_PROVIDED
Assume these dependencies don't exist and are already
provided (equivalent to ASSUME_PROVIDED). Useful to
make dependency graphs more appealing
-l DEBUG_DOMAINS, --log-domains=DEBUG_DOMAINS
Show debug logging for the specified logging domains
-P, --profile Profile the command and save reports.
-u UI, --ui=UI The user interface to use (e.g. knotty, hob, depexp).
-t SERVERTYPE, --servertype=SERVERTYPE
Choose which server to use, process or xmlrpc.
--revisions-changed Set the exit code depending on whether upstream
floating revisions have changed or not.
--server-only Run bitbake without a UI, only starting a server
(cooker) process.
-B BIND, --bind=BIND The name/address for the bitbake server to bind to.
--no-setscene Do not run any setscene tasks. sstate will be ignored
and everything needed, built.
--remote-server=REMOTE_SERVER
Connect to the specified server.
-m, --kill-server Terminate the remote server.
--observe-only Connect to a server as an observing-only client.
--status-only Check the status of the remote bitbake server.
</literallayout>
</para>
</section>
<section id='bitbake-examples'>
<title>Examples</title>
<para>
This section presents some examples showing how to use BitBake.
</para>
<section id='example-executing-a-task-against-a-single-recipe'>
<title>Executing a Task Against a Single Recipe</title>
<para>
Executing tasks for a single recipe file is relatively simple.
You specify the file in question, and BitBake parses
it and executes the specified task.
If you do not specify a task, BitBake executes the default
task, which is "build”.
BitBake obeys inter-task dependencies when doing
so.
</para>
<para>
The following command runs the build task, which is
the default task, on the <filename>foo_1.0.bb</filename>
recipe file:
<literallayout class='monospaced'>
$ bitbake -b foo_1.0.bb
</literallayout>
The following command runs the clean task on the
<filename>foo.bb</filename> recipe file:
<literallayout class='monospaced'>
$ bitbake -b foo.bb -c clean
</literallayout>
<note>
The "-b" option explicitly does not handle recipe
dependencies.
Other than for debugging purposes, it is instead
recommended that you use the syntax presented in the
next section.
</note>
</para>
</section>
<section id='executing-tasks-against-a-set-of-recipe-files'>
<title>Executing Tasks Against a Set of Recipe Files</title>
<para>
There are a number of additional complexities introduced
when one wants to manage multiple <filename>.bb</filename>
files.
Clearly there needs to be a way to tell BitBake what
files are available, and of those, which you
want to execute.
There also needs to be a way for each recipe
to express its dependencies, both for build-time and
runtime.
There must be a way for you to express recipe preferences
when multiple recipes provide the same functionality, or when
there are multiple versions of a recipe.
</para>
<para>
The <filename>bitbake</filename> command, when not using
"--buildfile" or "-b" only accepts a "PROVIDES".
You cannot provide anything else.
By default, a recipe file generally "PROVIDES" its
"packagename" as shown in the following example:
<literallayout class='monospaced'>
$ bitbake foo
</literallayout>
This next example "PROVIDES" the package name and also uses
the "-c" option to tell BitBake to just execute the
<filename>do_clean</filename> task:
<literallayout class='monospaced'>
$ bitbake -c clean foo
</literallayout>
</para>
</section>
<section id='generating-dependency-graphs'>
<title>Generating Dependency Graphs</title>
<para>
BitBake is able to generate dependency graphs using
the <filename>dot</filename> syntax.
You can convert these graphs into images using the
<filename>dot</filename> tool from
<ulink url='http://www.graphviz.org'>Graphviz</ulink>.
</para>
<para>
When you generate a dependency graph, BitBake writes four files
to the current working directory:
<itemizedlist>
<listitem><para><emphasis><filename>package-depends.dot</filename>:</emphasis>
Shows BitBake's knowledge of dependencies between
runtime targets.
</para></listitem>
<listitem><para><emphasis><filename>pn-depends.dot</filename>:</emphasis>
Shows dependencies between build-time targets
(i.e. recipes).
</para></listitem>
<listitem><para><emphasis><filename>task-depends.dot</filename>:</emphasis>
Shows dependencies between tasks.
</para></listitem>
<listitem><para><emphasis><filename>pn-buildlist</filename>:</emphasis>
Shows a simple list of targets that are to be built.
</para></listitem>
</itemizedlist>
</para>
<para>
To stop depending on common depends, use the "-I" depend
option and BitBake omits them from the graph.
Leaving this information out can produce more readable graphs.
This way, you can remove from the graph
<filename>DEPENDS</filename> from inherited classes
such as <filename>base.bbclass</filename>.
</para>
<para>
Here are two examples that create dependency graphs.
The second example omits depends common in OpenEmbedded from
the graph:
<literallayout class='monospaced'>
$ bitbake -g foo
$ bitbake -g -I virtual/kernel -I eglibc foo
</literallayout>
</para>
</section>
</section>
</section>
</chapter>

View File

@@ -1,978 +0,0 @@
/*
Generic XHTML / DocBook XHTML CSS Stylesheet.
Browser wrangling and typographic design by
Oyvind Kolas / pippin@gimp.org
Customised for Poky by
Matthew Allum / mallum@o-hand.com
Thanks to:
Liam R. E. Quin
William Skaggs
Jakub Steiner
Structure
---------
The stylesheet is divided into the following sections:
Positioning
Margins, paddings, width, font-size, clearing.
Decorations
Borders, style
Colors
Colors
Graphics
Graphical backgrounds
Nasty IE tweaks
Workarounds needed to make it work in internet explorer,
currently makes the stylesheet non validating, but up until
this point it is validating.
Mozilla extensions
Transparency for footer
Rounded corners on boxes
*/
/*************** /
/ Positioning /
/ ***************/
body {
font-family: Verdana, Sans, sans-serif;
min-width: 640px;
width: 80%;
margin: 0em auto;
padding: 2em 5em 5em 5em;
color: #333;
}
h1,h2,h3,h4,h5,h6,h7 {
font-family: Arial, Sans;
color: #00557D;
clear: both;
}
h1 {
font-size: 2em;
text-align: left;
padding: 0em 0em 0em 0em;
margin: 2em 0em 0em 0em;
}
h2.subtitle {
margin: 0.10em 0em 3.0em 0em;
padding: 0em 0em 0em 0em;
font-size: 1.8em;
padding-left: 20%;
font-weight: normal;
font-style: italic;
}
h2 {
margin: 2em 0em 0.66em 0em;
padding: 0.5em 0em 0em 0em;
font-size: 1.5em;
font-weight: bold;
}
h3.subtitle {
margin: 0em 0em 1em 0em;
padding: 0em 0em 0em 0em;
font-size: 142.14%;
text-align: right;
}
h3 {
margin: 1em 0em 0.5em 0em;
padding: 1em 0em 0em 0em;
font-size: 140%;
font-weight: bold;
}
h4 {
margin: 1em 0em 0.5em 0em;
padding: 1em 0em 0em 0em;
font-size: 120%;
font-weight: bold;
}
h5 {
margin: 1em 0em 0.5em 0em;
padding: 1em 0em 0em 0em;
font-size: 110%;
font-weight: bold;
}
h6 {
margin: 1em 0em 0em 0em;
padding: 1em 0em 0em 0em;
font-size: 110%;
font-weight: bold;
}
.authorgroup {
background-color: transparent;
background-repeat: no-repeat;
padding-top: 256px;
background-image: url("figures/bitbake-title.png");
background-position: left top;
margin-top: -256px;
padding-right: 50px;
margin-left: 0px;
text-align: right;
width: 740px;
}
h3.author {
margin: 0em 0me 0em 0em;
padding: 0em 0em 0em 0em;
font-weight: normal;
font-size: 100%;
color: #333;
clear: both;
}
.author tt.email {
font-size: 66%;
}
.titlepage hr {
width: 0em;
clear: both;
}
.revhistory {
padding-top: 2em;
clear: both;
}
.toc,
.list-of-tables,
.list-of-examples,
.list-of-figures {
padding: 1.33em 0em 2.5em 0em;
color: #00557D;
}
.toc p,
.list-of-tables p,
.list-of-figures p,
.list-of-examples p {
padding: 0em 0em 0em 0em;
padding: 0em 0em 0.3em;
margin: 1.5em 0em 0em 0em;
}
.toc p b,
.list-of-tables p b,
.list-of-figures p b,
.list-of-examples p b{
font-size: 100.0%;
font-weight: bold;
}
.toc dl,
.list-of-tables dl,
.list-of-figures dl,
.list-of-examples dl {
margin: 0em 0em 0.5em 0em;
padding: 0em 0em 0em 0em;
}
.toc dt {
margin: 0em 0em 0em 0em;
padding: 0em 0em 0em 0em;
}
.toc dd {
margin: 0em 0em 0em 2.6em;
padding: 0em 0em 0em 0em;
}
div.glossary dl,
div.variablelist dl {
}
.glossary dl dt,
.variablelist dl dt,
.variablelist dl dt span.term {
font-weight: normal;
width: 20em;
text-align: right;
}
.variablelist dl dt {
margin-top: 0.5em;
}
.glossary dl dd,
.variablelist dl dd {
margin-top: -1em;
margin-left: 25.5em;
}
.glossary dd p,
.variablelist dd p {
margin-top: 0em;
margin-bottom: 1em;
}
div.calloutlist table td {
padding: 0em 0em 0em 0em;
margin: 0em 0em 0em 0em;
}
div.calloutlist table td p {
margin-top: 0em;
margin-bottom: 1em;
}
div p.copyright {
text-align: left;
}
div.legalnotice p.legalnotice-title {
margin-bottom: 0em;
}
p {
line-height: 1.5em;
margin-top: 0em;
}
dl {
padding-top: 0em;
}
hr {
border: solid 1px;
}
.mediaobject,
.mediaobjectco {
text-align: center;
}
img {
border: none;
}
ul {
padding: 0em 0em 0em 1.5em;
}
ul li {
padding: 0em 0em 0em 0em;
}
ul li p {
text-align: left;
}
table {
width :100%;
}
th {
padding: 0.25em;
text-align: left;
font-weight: normal;
vertical-align: top;
}
td {
padding: 0.25em;
vertical-align: top;
}
p a[id] {
margin: 0px;
padding: 0px;
display: inline;
background-image: none;
}
a {
text-decoration: underline;
color: #444;
}
pre {
overflow: auto;
}
a:hover {
text-decoration: underline;
/*font-weight: bold;*/
}
div.informalfigure,
div.informalexample,
div.informaltable,
div.figure,
div.table,
div.example {
margin: 1em 0em;
padding: 1em;
page-break-inside: avoid;
}
div.informalfigure p.title b,
div.informalexample p.title b,
div.informaltable p.title b,
div.figure p.title b,
div.example p.title b,
div.table p.title b{
padding-top: 0em;
margin-top: 0em;
font-size: 100%;
font-weight: normal;
}
.mediaobject .caption,
.mediaobject .caption p {
text-align: center;
font-size: 80%;
padding-top: 0.5em;
padding-bottom: 0.5em;
}
.epigraph {
padding-left: 55%;
margin-bottom: 1em;
}
.epigraph p {
text-align: left;
}
.epigraph .quote {
font-style: italic;
}
.epigraph .attribution {
font-style: normal;
text-align: right;
}
span.application {
font-style: italic;
}
.programlisting {
font-family: monospace;
font-size: 80%;
white-space: pre;
margin: 1.33em 0em;
padding: 1.33em;
}
.tip,
.warning,
.caution,
.note {
margin-top: 1em;
margin-bottom: 1em;
}
/* force full width of table within div */
.tip table,
.warning table,
.caution table,
.note table {
border: none;
width: 100%;
}
.tip table th,
.warning table th,
.caution table th,
.note table th {
padding: 0.8em 0.0em 0.0em 0.0em;
margin : 0em 0em 0em 0em;
}
.tip p,
.warning p,
.caution p,
.note p {
margin-top: 0.5em;
margin-bottom: 0.5em;
padding-right: 1em;
text-align: left;
}
.acronym {
text-transform: uppercase;
}
b.keycap,
.keycap {
padding: 0.09em 0.3em;
margin: 0em;
}
.itemizedlist li {
clear: none;
}
.filename {
font-size: medium;
font-family: Courier, monospace;
}
div.navheader, div.heading{
position: absolute;
left: 0em;
top: 0em;
width: 100%;
background-color: #cdf;
width: 100%;
}
div.navfooter, div.footing{
position: fixed;
left: 0em;
bottom: 0em;
background-color: #eee;
width: 100%;
}
div.navheader td,
div.navfooter td {
font-size: 66%;
}
div.navheader table th {
/*font-family: Georgia, Times, serif;*/
/*font-size: x-large;*/
font-size: 80%;
}
div.navheader table {
border-left: 0em;
border-right: 0em;
border-top: 0em;
width: 100%;
}
div.navfooter table {
border-left: 0em;
border-right: 0em;
border-bottom: 0em;
width: 100%;
}
div.navheader table td a,
div.navfooter table td a {
color: #777;
text-decoration: none;
}
/* normal text in the footer */
div.navfooter table td {
color: black;
}
div.navheader table td a:visited,
div.navfooter table td a:visited {
color: #444;
}
/* links in header and footer */
div.navheader table td a:hover,
div.navfooter table td a:hover {
text-decoration: underline;
background-color: transparent;
color: #33a;
}
div.navheader hr,
div.navfooter hr {
display: none;
}
.qandaset tr.question td p {
margin: 0em 0em 1em 0em;
padding: 0em 0em 0em 0em;
}
.qandaset tr.answer td p {
margin: 0em 0em 1em 0em;
padding: 0em 0em 0em 0em;
}
.answer td {
padding-bottom: 1.5em;
}
.emphasis {
font-weight: bold;
}
/************* /
/ decorations /
/ *************/
.titlepage {
}
.part .title {
}
.subtitle {
border: none;
}
/*
h1 {
border: none;
}
h2 {
border-top: solid 0.2em;
border-bottom: solid 0.06em;
}
h3 {
border-top: 0em;
border-bottom: solid 0.06em;
}
h4 {
border: 0em;
border-bottom: solid 0.06em;
}
h5 {
border: 0em;
}
*/
.programlisting {
border: solid 1px;
}
div.figure,
div.table,
div.informalfigure,
div.informaltable,
div.informalexample,
div.example {
border: 1px solid;
}
.tip,
.warning,
.caution,
.note {
border: 1px solid;
}
.tip table th,
.warning table th,
.caution table th,
.note table th {
border-bottom: 1px solid;
}
.question td {
border-top: 1px solid black;
}
.answer {
}
b.keycap,
.keycap {
border: 1px solid;
}
div.navheader, div.heading{
border-bottom: 1px solid;
}
div.navfooter, div.footing{
border-top: 1px solid;
}
/********* /
/ colors /
/ *********/
body {
color: #333;
background: white;
}
a {
background: transparent;
}
a:hover {
background-color: #dedede;
}
h1,
h2,
h3,
h4,
h5,
h6,
h7,
h8 {
background-color: transparent;
}
hr {
border-color: #aaa;
}
.tip, .warning, .caution, .note {
border-color: #fff;
}
.tip table th,
.warning table th,
.caution table th,
.note table th {
border-bottom-color: #fff;
}
.warning {
background-color: #f0f0f2;
}
.caution {
background-color: #f0f0f2;
}
.tip {
background-color: #f0f0f2;
}
.note {
background-color: #f0f0f2;
}
.glossary dl dt,
.variablelist dl dt,
.variablelist dl dt span.term {
color: #044;
}
div.figure,
div.table,
div.example,
div.informalfigure,
div.informaltable,
div.informalexample {
border-color: #aaa;
}
pre.programlisting {
color: black;
background-color: #fff;
border-color: #aaa;
border-width: 2px;
}
.guimenu,
.guilabel,
.guimenuitem {
background-color: #eee;
}
b.keycap,
.keycap {
background-color: #eee;
border-color: #999;
}
div.navheader {
border-color: black;
}
div.navfooter {
border-color: black;
}
/*********** /
/ graphics /
/ ***********/
/*
body {
background-image: url("images/body_bg.jpg");
background-attachment: fixed;
}
.navheader,
.note,
.tip {
background-image: url("images/note_bg.jpg");
background-attachment: fixed;
}
.warning,
.caution {
background-image: url("images/warning_bg.jpg");
background-attachment: fixed;
}
.figure,
.informalfigure,
.example,
.informalexample,
.table,
.informaltable {
background-image: url("images/figure_bg.jpg");
background-attachment: fixed;
}
*/
h1,
h2,
h3,
h4,
h5,
h6,
h7{
}
/*
Example of how to stick an image as part of the title.
div.article .titlepage .title
{
background-image: url("figures/white-on-black.png");
background-position: center;
background-repeat: repeat-x;
}
*/
div.preface .titlepage .title,
div.colophon .title,
div.chapter .titlepage .title,
div.article .titlepage .title
{
}
div.section div.section .titlepage .title,
div.sect2 .titlepage .title {
background: none;
}
h1.title {
background-color: transparent;
background-image: url("figures/yocto-project-bw.png");
background-repeat: no-repeat;
height: 256px;
text-indent: -9000px;
overflow:hidden;
}
h2.subtitle {
background-color: transparent;
text-indent: -9000px;
overflow:hidden;
width: 0px;
display: none;
}
/*************************************** /
/ pippin.gimp.org specific alterations /
/ ***************************************/
/*
div.heading, div.navheader {
color: #777;
font-size: 80%;
padding: 0;
margin: 0;
text-align: left;
position: absolute;
top: 0px;
left: 0px;
width: 100%;
height: 50px;
background: url('/gfx/heading_bg.png') transparent;
background-repeat: repeat-x;
background-attachment: fixed;
border: none;
}
div.heading a {
color: #444;
}
div.footing, div.navfooter {
border: none;
color: #ddd;
font-size: 80%;
text-align:right;
width: 100%;
padding-top: 10px;
position: absolute;
bottom: 0px;
left: 0px;
background: url('/gfx/footing_bg.png') transparent;
}
*/
/****************** /
/ nasty ie tweaks /
/ ******************/
/*
div.heading, div.navheader {
width:expression(document.body.clientWidth + "px");
}
div.footing, div.navfooter {
width:expression(document.body.clientWidth + "px");
margin-left:expression("-5em");
}
body {
padding:expression("4em 5em 0em 5em");
}
*/
/**************************************** /
/ mozilla vendor specific css extensions /
/ ****************************************/
/*
div.navfooter, div.footing{
-moz-opacity: 0.8em;
}
div.figure,
div.table,
div.informalfigure,
div.informaltable,
div.informalexample,
div.example,
.tip,
.warning,
.caution,
.note {
-moz-border-radius: 0.5em;
}
b.keycap,
.keycap {
-moz-border-radius: 0.3em;
}
*/
table tr td table tr td {
display: none;
}
hr {
display: none;
}
table {
border: 0em;
}
.photo {
float: right;
margin-left: 1.5em;
margin-bottom: 1.5em;
margin-top: 0em;
max-width: 17em;
border: 1px solid gray;
padding: 3px;
background: white;
}
.seperator {
padding-top: 2em;
clear: both;
}
#validators {
margin-top: 5em;
text-align: right;
color: #777;
}
@media print {
body {
font-size: 8pt;
}
.noprint {
display: none;
}
}
.tip,
.note {
background: #f0f0f2;
color: #333;
padding: 20px;
margin: 20px;
}
.tip h3,
.note h3 {
padding: 0em;
margin: 0em;
font-size: 2em;
font-weight: bold;
color: #333;
}
.tip a,
.note a {
color: #333;
text-decoration: underline;
}
.footnote {
font-size: small;
color: #333;
}
/* Changes the announcement text */
.tip h3,
.warning h3,
.caution h3,
.note h3 {
font-size:large;
color: #00557D;
}

View File

@@ -1,88 +0,0 @@
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<book id='bitbake-user-manual' lang='en'
xmlns:xi="http://www.w3.org/2003/XInclude"
xmlns="http://docbook.org/ns/docbook"
>
<bookinfo>
<mediaobject>
<imageobject>
<imagedata fileref='figures/bitbake-title.png'
format='SVG'
align='left' scalefit='1' width='100%'/>
</imageobject>
</mediaobject>
<title>
BitBake User Manual
</title>
<authorgroup>
<author>
<firstname>Richard Purdie, Chris Larson, and </firstname> <surname>Phil Blundell</surname>
<affiliation>
<orgname>BitBake Community</orgname>
</affiliation>
<email>bitbake-devel@lists.openembedded.org</email>
</author>
</authorgroup>
<!--
# Add in some revision history if we want it here.
<revhistory>
<revision>
<revnumber>x.x</revnumber>
<date>dd month year</date>
<revremark>Some relevent comment</revremark>
</revision>
<revision>
<revnumber>x.x</revnumber>
<date>dd month year</date>
<revremark>Some relevent comment</revremark>
</revision>
<revision>
<revnumber>x.x</revnumber>
<date>dd month year</date>
<revremark>Some relevent comment</revremark>
</revision>
<revision>
<revnumber>x.x</revnumber>
<date>dd month year</date>
<revremark>Some relevent comment</revremark>
</revision>
</revhistory>
-->
<copyright>
<year>2004-2014</year>
<holder>Richard Purdie</holder>
<holder>Chris Larson</holder>
<holder>and Phil Blundell</holder>
</copyright>
<legalnotice>
<para>
This work is licensed under the Creative Commons Attribution License.
To view a copy of this license, visit
<ulink url="http://creativecommons.org/licenses/by/2.5/">http://creativecommons.org/licenses/by/2.5/</ulink>
or send a letter to Creative Commons, 444 Castro Street,
Suite 900, Mountain View, California 94041, USA.
</para>
</legalnotice>
</bookinfo>
<xi:include href="bitbake-user-manual-intro.xml"/>
<xi:include href="bitbake-user-manual-execution.xml"/>
<xi:include href="bitbake-user-manual-metadata.xml"/>
<xi:include href="bitbake-user-manual-fetching.xml"/>
<xi:include href="bitbake-user-manual-ref-variables.xml"/>
<xi:include href="bitbake-user-manual-hello.xml"/>
</book>

Binary file not shown.

Before

Width:  |  Height:  |  Size: 5.0 KiB

View File

@@ -32,7 +32,7 @@ command.
\fBbitbake\fP is a program that executes the specified task (default is 'build')
for a given set of BitBake files.
.br
It expects that BBFILES is defined, which is a space separated list of files to
It expects that BBFILES is defined, which is a space seperated list of files to
be executed. BBFILES does support wildcards.
.br
Default BBFILES are the .bb files in the current directory.
@@ -67,7 +67,7 @@ drop into the interactive mode also called the BitBake shell.
Specify task to execute. Note that this only executes the specified task for
the providee and the packages it depends on, i.e. 'compile' does not implicitly
call stage for the dependencies (IOW: use only if you know what you are doing).
Depending on the base.bbclass a listtasks task is defined and will show
Depending on the base.bbclass a listtaks tasks is defined and will show
available tasks.
.TP
.B \-rFILE, \-\-read=FILE
@@ -85,6 +85,9 @@ don't execute, just go through the motions
.B \-p, \-\-parse-only
quit after parsing the BB files (developers only)
.TP
.B \-d, \-\-disable-psyco
disable using the psyco just-in-time compiler (not recommended)
.TP
.B \-s, \-\-show-versions
show current and preferred versions of all packages
.TP
@@ -104,30 +107,6 @@ Show debug logging for the specified logging domains
.B \-P, \-\-profile
profile the command and print a report
.TP
.B \-uUI, \-\-ui=UI
User interface to use. Currently, hob, depexp, goggle or ncurses can be specified as UI.
.TP
.B \-tSERVERTYPE, \-\-servertype=SERVERTYPE
Choose which server to use, none, process or xmlrpc.
.TP
.B \-\-revisions-changed
Set the exit code depending on whether upstream floating revisions have changed or not.
.TP
.B \-\-server-only
Run bitbake without UI, the frontend can connect with bitbake server itself.
.TP
.B \-BBIND, \-\-bind=BIND
The name/address for the bitbake server to bind to.
.TP
.B \-\-no\-setscene
Do not run any setscene tasks, forces builds.
.SH ENVIRONMENT VARIABLES
bitbake uses the following environment variables to control its
operation:
.TP
.B BITBAKE_UI
The bitbake user interface; overridden by the \fB-u\fP commandline option.
.SH AUTHORS
BitBake was written by

View File

@@ -0,0 +1,56 @@
topdir = .
manual = $(topdir)/usermanual.xml
# types = pdf txt rtf ps xhtml html man tex texi dvi
# types = pdf txt
types = $(xmltotypes) $(htmltypes)
xmltotypes = pdf txt
htmltypes = html xhtml
htmlxsl = $(if $(filter $@,$(foreach type,$(htmltypes),$(type)-nochunks)),http://docbook.sourceforge.net/release/xsl/current/xhtml/docbook.xsl,http://docbook.sourceforge.net/release/xsl/current/$@/chunk.xsl)
htmlcssfile = docbook.css
htmlcss = $(topdir)/html.css
# htmlcssfile =
# htmlcss =
cleanfiles = $(foreach i,$(types),$(topdir)/$(i))
ifdef DEBUG
define command
$(1)
endef
else
define command
@echo $(2) $(3) $(4)
@$(1) >/dev/null
endef
endif
all: $(types)
lint: $(manual) FORCE
$(call command,xmllint --xinclude --postvalid --noout $(manual),XMLLINT $(manual))
$(types) $(foreach type,$(htmltypes),$(type)-nochunks): lint FORCE
$(foreach type,$(htmltypes),$(type)-nochunks): $(if $(htmlcss),$(htmlcss)) $(manual)
@mkdir -p $@
ifdef htmlcss
$(call command,install -m 0644 $(htmlcss) $@/$(htmlcssfile),CP $(htmlcss) $@/$(htmlcssfile))
endif
$(call command,xsltproc --stringparam base.dir $@/ $(if $(htmlcssfile),--stringparam html.stylesheet $(htmlcssfile)) $(htmlxsl) $(manual) > $@/index.$(patsubst %-nochunks,%,$@),XSLTPROC $@ $(manual))
$(htmltypes): $(if $(htmlcss),$(htmlcss)) $(manual)
@mkdir -p $@
ifdef htmlcss
$(call command,install -m 0644 $(htmlcss) $@/$(htmlcssfile),CP $(htmlcss) $@/$(htmlcssfile))
endif
$(call command,xsltproc --stringparam base.dir $@/ $(if $(htmlcssfile),--stringparam html.stylesheet $(htmlcssfile)) $(htmlxsl) $(manual),XSLTPROC $@ $(manual))
$(xmltotypes): $(manual)
$(call command,xmlto --extensions -o $(topdir)/$@ $@ $(manual),XMLTO $@ $(manual))
clean:
rm -rf $(cleanfiles)
$(foreach i,$(types) $(foreach type,$(htmltypes),$(type)-nochunks),clean-$(i)):
rm -rf $(patsubst clean-%,%,$@)
FORCE:

View File

@@ -0,0 +1,519 @@
<?xml version="1.0"?>
<!--
ex:ts=4:sw=4:sts=4:et
-*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
-->
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
<book>
<bookinfo>
<title>BitBake User Manual</title>
<authorgroup>
<corpauthor>BitBake Team</corpauthor>
</authorgroup>
<copyright>
<year>2004, 2005, 2006</year>
<holder>Chris Larson</holder>
<holder>Phil Blundell</holder>
</copyright>
<legalnotice>
<para>This work is licensed under the Creative Commons Attribution License. To view a copy of this license, visit <ulink url="http://creativecommons.org/licenses/by/2.5/">http://creativecommons.org/licenses/by/2.5/</ulink> or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.</para>
</legalnotice>
</bookinfo>
<chapter>
<title>Introduction</title>
<section>
<title>Overview</title>
<para>BitBake is, at its simplest, a tool for executing
tasks and managing metadata. As such, its similarities to GNU make and other
build tools are readily apparent. It was inspired by Portage, the package management system used by the Gentoo Linux distribution. BitBake is the basis of the <ulink url="http://www.openembedded.org/">OpenEmbedded</ulink> project, which is being used to build and maintain a number of embedded Linux distributions, including OpenZaurus and Familiar.</para>
</section>
<section>
<title>Background and Goals</title>
<para>Prior to BitBake, no other build tool adequately met
the needs of an aspiring embedded Linux distribution. All of the
buildsystems used by traditional desktop Linux distributions lacked
important functionality, and none of the ad-hoc
<emphasis>buildroot</emphasis> systems, prevalent in the
embedded space, were scalable or maintainable.</para>
<para>Some important goals for BitBake were:
<itemizedlist>
<listitem><para>Handle crosscompilation.</para></listitem>
<listitem><para>Handle interpackage dependencies (build time on target architecture, build time on native architecture, and runtime).</para></listitem>
<listitem><para>Support running any number of tasks within a given package, including, but not limited to, fetching upstream sources, unpacking them, patching them, configuring them, et cetera.</para></listitem>
<listitem><para>Must be linux distribution agnostic (both build and target).</para></listitem>
<listitem><para>Must be architecture agnostic</para></listitem>
<listitem><para>Must support multiple build and target operating systems (including cygwin, the BSDs, etc).</para></listitem>
<listitem><para>Must be able to be self contained, rather than tightly integrated into the build machine's root filesystem.</para></listitem>
<listitem><para>There must be a way to handle conditional metadata (on target architecture, operating system, distribution, machine).</para></listitem>
<listitem><para>It must be easy for the person using the tools to supply their own local metadata and packages to operate against.</para></listitem>
<listitem><para>Must make it easy to collaborate
between multiple projects using BitBake for their
builds.</para></listitem>
<listitem><para>Should provide an inheritance mechanism to
share common metadata between many packages.</para></listitem>
<listitem><para>Et cetera...</para></listitem>
</itemizedlist>
</para>
<para>BitBake satisfies all these and many more. Flexibility and power have always been the priorities. It is highly extensible, supporting embedded Python code and execution of any arbitrary tasks.</para>
</section>
</chapter>
<chapter>
<title>Metadata</title>
<section>
<title>Description</title>
<itemizedlist>
<para>BitBake metadata can be classified into 3 major areas:</para>
<listitem>
<para>Configuration Files</para>
</listitem>
<listitem>
<para>.bb Files</para>
</listitem>
<listitem>
<para>Classes</para>
</listitem>
</itemizedlist>
<para>What follows are a large number of examples of BitBake metadata. Any syntax which isn't supported in any of the aforementioned areas will be documented as such.</para>
<section>
<title>Basic variable setting</title>
<para><screen><varname>VARIABLE</varname> = "value"</screen></para>
<para>In this example, <varname>VARIABLE</varname> is <literal>value</literal>.</para>
</section>
<section>
<title>Variable expansion</title>
<para>BitBake supports variables referencing one another's contents using a syntax which is similar to shell scripting</para>
<para><screen><varname>A</varname> = "aval"
<varname>B</varname> = "pre${A}post"</screen></para>
<para>This results in <varname>A</varname> containing <literal>aval</literal> and <varname>B</varname> containing <literal>preavalpost</literal>.</para>
</section>
<section>
<title>Immediate variable expansion (:=)</title>
<para>:= results in a variable's contents being expanded immediately, rather than when the variable is actually used.</para>
<para><screen><varname>T</varname> = "123"
<varname>A</varname> := "${B} ${A} test ${T}"
<varname>T</varname> = "456"
<varname>B</varname> = "${T} bval"
<varname>C</varname> = "cval"
<varname>C</varname> := "${C}append"</screen></para>
<para>In that example, <varname>A</varname> would contain <literal> test 123</literal>, <varname>B</varname> would contain <literal>456 bval</literal>, and <varname>C</varname> would be <literal>cvalappend</literal>.</para>
</section>
<section>
<title>Appending (+=) and prepending (=+)</title>
<para><screen><varname>B</varname> = "bval"
<varname>B</varname> += "additionaldata"
<varname>C</varname> = "cval"
<varname>C</varname> =+ "test"</screen></para>
<para>In this example, <varname>B</varname> is now <literal>bval additionaldata</literal> and <varname>C</varname> is <literal>test cval</literal>.</para>
</section>
<section>
<title>Appending (.=) and prepending (=.) without spaces</title>
<para><screen><varname>B</varname> = "bval"
<varname>B</varname> .= "additionaldata"
<varname>C</varname> = "cval"
<varname>C</varname> =. "test"</screen></para>
<para>In this example, <varname>B</varname> is now <literal>bvaladditionaldata</literal> and <varname>C</varname> is <literal>testcval</literal>. In contrast to the above Appending and Prepending operators no additional space
will be introduced.</para>
</section>
<section>
<title>Conditional metadata set</title>
<para>OVERRIDES is a <quote>:</quote> seperated variable containing each item you want to satisfy conditions. So, if you have a variable which is conditional on <quote>arm</quote>, and <quote>arm</quote> is in OVERRIDES, then the <quote>arm</quote> specific version of the variable is used rather than the non-conditional version. Example:</para>
<para><screen><varname>OVERRIDES</varname> = "architecture:os:machine"
<varname>TEST</varname> = "defaultvalue"
<varname>TEST_os</varname> = "osspecificvalue"
<varname>TEST_condnotinoverrides</varname> = "othercondvalue"</screen></para>
<para>In this example, <varname>TEST</varname> would be <literal>osspecificvalue</literal>, due to the condition <quote>os</quote> being in <varname>OVERRIDES</varname>.</para>
</section>
<section>
<title>Conditional appending</title>
<para>BitBake also supports appending and prepending to variables based on whether something is in OVERRIDES. Example:</para>
<para><screen><varname>DEPENDS</varname> = "glibc ncurses"
<varname>OVERRIDES</varname> = "machine:local"
<varname>DEPENDS_append_machine</varname> = " libmad"</screen></para>
<para>In this example, <varname>DEPENDS</varname> is set to <literal>glibc ncurses libmad</literal>.</para>
</section>
<section>
<title>Inclusion</title>
<para>Next, there is the <literal>include</literal> directive, which causes BitBake to parse in whatever file you specify, and insert it at that location, which is not unlike <command>make</command>. However, if the path specified on the <literal>include</literal> line is a relative path, BitBake will locate the first one it can find within <envar>BBPATH</envar>.</para>
</section>
<section>
<title>Requiring Inclusion</title>
<para>In contrast to the <literal>include</literal> directive, <literal>require</literal> will
raise an ParseError if the to be included file can not be found. Otherwise it will behave just like the <literal>
include</literal> directive.</para>
</section>
<section>
<title>Python variable expansion</title>
<para><screen><varname>DATE</varname> = "${@time.strftime('%Y%m%d',time.gmtime())}"</screen></para>
<para>This would result in the <varname>DATE</varname> variable containing today's date.</para>
</section>
<section>
<title>Defining executable metadata</title>
<para><emphasis>NOTE:</emphasis> This is only supported in .bb and .bbclass files.</para>
<para><screen>do_mytask () {
echo "Hello, world!"
}</screen></para>
<para>This is essentially identical to setting a variable, except that this variable happens to be executable shell code.</para>
<para><screen>python do_printdate () {
import time
print time.strftime('%Y%m%d', time.gmtime())
}</screen></para>
<para>This is the similar to the previous, but flags it as python so that BitBake knows it is python code.</para>
</section>
<section>
<title>Defining python functions into the global python namespace</title>
<para><emphasis>NOTE:</emphasis> This is only supported in .bb and .bbclass files.</para>
<para><screen>def get_depends(bb, d):
if bb.data.getVar('SOMECONDITION', d, True):
return "dependencywithcond"
else:
return "dependency"
<varname>SOMECONDITION</varname> = "1"
<varname>DEPENDS</varname> = "${@get_depends(bb, d)}"</screen></para>
<para>This would result in <varname>DEPENDS</varname> containing <literal>dependencywithcond</literal>.</para>
</section>
<section>
<title>Variable Flags</title>
<para>Variables can have associated flags which provide a way of tagging extra information onto a variable. Several flags are used internally by bitbake but they can be used externally too if needed. The standard operations mentioned above also work on flags.</para>
<para><screen><varname>VARIABLE</varname>[<varname>SOMEFLAG</varname>] = "value"</screen></para>
<para>In this example, <varname>VARIABLE</varname> has a flag, <varname>SOMEFLAG</varname> which is set to <literal>value</literal>.</para>
</section>
<section>
<title>Inheritance</title>
<para><emphasis>NOTE:</emphasis> This is only supported in .bb and .bbclass files.</para>
<para>The <literal>inherit</literal> directive is a means of specifying what classes of functionality your .bb requires. It is a rudamentary form of inheritence. For example, you can easily abstract out the tasks involved in building a package that uses autoconf and automake, and put that into a bbclass for your packages to make use of. A given bbclass is located by searching for classes/filename.oeclass in <envar>BBPATH</envar>, where filename is what you inherited.</para>
</section>
<section>
<title>Tasks</title>
<para><emphasis>NOTE:</emphasis> This is only supported in .bb and .bbclass files.</para>
<para>In BitBake, each step that needs to be run for a given .bb is known as a task. There is a command <literal>addtask</literal> to add new tasks (must be a defined python executable metadata and must start with <quote>do_</quote>) and describe intertask dependencies.</para>
<para><screen>python do_printdate () {
import time
print time.strftime('%Y%m%d', time.gmtime())
}
addtask printdate before do_build</screen></para>
<para>This defines the necessary python function and adds it as a task which is now a dependency of do_build (the default task). If anyone executes the do_build task, that will result in do_printdate being run first.</para>
</section>
<section>
<title>Events</title>
<para><emphasis>NOTE:</emphasis> This is only supported in .bb and .bbclass files.</para>
<para>BitBake allows to install event handlers. Events are triggered at certain points during operation, such as, the beginning of operation against a given .bb, the start of a given task, task failure, task success, et cetera. The intent was to make it easy to do things like email notifications on build failure.</para>
<para><screen>addhandler myclass_eventhandler
python myclass_eventhandler() {
from bb.event import NotHandled, getName
from bb import data
print "The name of the Event is %s" % getName(e)
print "The file we run for is %s" % data.getVar('FILE', e.data, True)
return NotHandled
}
</screen></para><para>
This event handler gets called every time an event is triggered. A global variable <varname>e</varname> is defined. <varname>e</varname>.data contains an instance of bb.data. With the getName(<varname>e</varname>)
method one can get the name of the triggered event.</para><para>The above event handler prints the name
of the event and the content of the <varname>FILE</varname> variable.</para>
</section>
</section>
<section>
<title>Dependency Handling</title>
<para>Bitbake 1.7.x onwards works with the metadata at the task level since this is optimal when dealing with multiple threads of execution. A robust method of specifing task dependencies is therefore needed. </para>
<section>
<title>Dependencies internal to the .bb file</title>
<para>Where the dependencies are internal to a given .bb file, the dependencies are handled by the previously detailed addtask directive.</para>
</section>
<section>
<title>DEPENDS</title>
<para>DEPENDS is taken to specify build time dependencies. The 'deptask' flag for tasks is used to signify the task of each DEPENDS which must have completed before that task can be executed.</para>
<para><screen>do_configure[deptask] = "do_populate_staging"</screen></para>
<para>means the do_populate_staging task of each item in DEPENDS must have completed before do_configure can execute.</para>
</section>
<section>
<title>RDEPENDS</title>
<para>RDEPENDS is taken to specify runtime dependencies. The 'rdeptask' flag for tasks is used to signify the task of each RDEPENDS which must have completed before that task can be executed.</para>
<para><screen>do_package_write[rdeptask] = "do_package"</screen></para>
<para>means the do_package task of each item in RDEPENDS must have completed before do_package_write can execute.</para>
</section>
<section>
<title>Recursive DEPENDS</title>
<para>These are specified with the 'recdeptask' flag and is used signify the task(s) of each DEPENDS which must have completed before that task can be executed. It applies recursively so also, the DEPENDS of each item in the original DEPENDS must be met and so on.</para>
</section>
<section>
<title>Recursive RDEPENDS</title>
<para>These are specified with the 'recrdeptask' flag and is used signify the task(s) of each RDEPENDS which must have completed before that task can be executed. It applies recursively so also, the RDEPENDS of each item in the original RDEPENDS must be met and so on. It also runs all DEPENDS first too.</para>
</section>
<section>
<title>Inter Task</title>
<para>The 'depends' flag for tasks is a more generic form of which allows an interdependency on specific tasks rather than specifying the data in DEPENDS or RDEPENDS.</para>
<para><screen>do_patch[depends] = "quilt-native:do_populate_staging"</screen></para>
<para>means the do_populate_staging task of the target quilt-native must have completed before the do_patch can execute.</para>
</section>
</section>
<section>
<title>Parsing</title>
<section>
<title>Configuration Files</title>
<para>The first of the classifications of metadata in BitBake is configuration metadata. This metadata is global, and therefore affects <emphasis>all</emphasis> packages and tasks which are executed. Currently, BitBake has hardcoded knowledge of a single configuration file. It expects to find 'conf/bitbake.conf' somewhere in the user specified <envar>BBPATH</envar>. That configuration file generally has include directives to pull in any other metadata (generally files specific to architecture, machine, <emphasis>local</emphasis> and so on.</para>
<para>Only variable definitions and include directives are allowed in .conf files.</para>
</section>
<section>
<title>Classes</title>
<para>BitBake classes are our rudamentary inheritence mechanism. As briefly mentioned in the metadata introduction, they're parsed when an <literal>inherit</literal> directive is encountered, and they are located in classes/ relative to the dirs in <envar>BBPATH</envar>.</para>
</section>
<section>
<title>.bb Files</title>
<para>A BitBake (.bb) file is a logical unit of tasks to be executed. Normally this is a package to be built. Inter-.bb dependencies are obeyed. The files themselves are located via the <varname>BBFILES</varname> variable, which is set to a space seperated list of .bb files, and does handle wildcards.</para>
</section>
</section>
</chapter>
<chapter>
<title>File Download support</title>
<section>
<title>Overview</title>
<para>BitBake provides support to download files this procedure is called fetching. The SRC_URI is normally used to indicate BitBake which files to fetch. The next sections will describe th available fetchers and the options they have. Each Fetcher honors a set of Variables and
a per URI parameters separated by a <quote>;</quote> consisting of a key and a value. The semantic of the Variables and Parameters are defined by the Fetcher. BitBakes tries to have a consistent semantic between the different Fetchers.
</para>
</section>
<section>
<title>Local File Fetcher</title>
<para>The URN for the Local File Fetcher is <emphasis>file</emphasis>. The filename can be either absolute or relative. If the filename is relative <varname>FILESPATH</varname> and <varname>FILESDIR</varname> will be used to find the appropriate relative file depending on the <varname>OVERRIDES</varname>. Single files and complete directories can be specified.
<screen><varname>SRC_URI</varname>= "file://relativefile.patch"
<varname>SRC_URI</varname>= "file://relativefile.patch;this=ignored"
<varname>SRC_URI</varname>= "file:///Users/ich/very_important_software"
</screen>
</para>
</section>
<section>
<title>CVS File Fetcher</title>
<para>The URN for the CVS Fetcher is <emphasis>cvs</emphasis>. This Fetcher honors the variables <varname>DL_DIR</varname>, <varname>SRCDATE</varname>, <varname>FETCHCOMMAND_cvs</varname>, <varname>UPDATECOMMAND_cvs</varname>. <varname>DL_DIRS</varname> specifies where a temporary checkout is saved, <varname>SRCDATE</varname> specifies which date to use when doing the fetching (the special value of "now" will cause the checkout to be updated on every build), <varname>FETCHCOMMAND</varname> and <varname>UPDATECOMMAND</varname> specify which executables should be used when doing the CVS checkout or update.
</para>
<para>The supported Parameters are <varname>module</varname>, <varname>tag</varname>, <varname>date</varname>, <varname>method</varname>, <varname>localdir</varname>, <varname>rsh</varname>. The <varname>module</varname> specifies which module to check out, the <varname>tag</varname> describes which CVS TAG should be used for the checkout by default the TAG is empty. A <varname>date</varname> can be specified to override the SRCDATE of the configuration to checkout a specific date. The special value of "now" will cause the checkout to be updated on every build.<varname>method</varname> is by default <emphasis>pserver</emphasis>, if <emphasis>ext</emphasis> is used the <varname>rsh</varname> parameter will be evaluated and <varname>CVS_RSH</varname> will be set. Finally <varname>localdir</varname> is used to checkout into a special directory relative to <varname>CVSDIR></varname>.
<screen><varname>SRC_URI</varname> = "cvs://CVSROOT;module=mymodule;tag=some-version;method=ext"
<varname>SRC_URI</varname> = "cvs://CVSROOT;module=mymodule;date=20060126;localdir=usethat"
</screen>
</para>
</section>
<section>
<title>HTTP/FTP Fetcher</title>
<para>The URNs for the HTTP/FTP are <emphasis>http</emphasis>, <emphasis>https</emphasis> and <emphasis>ftp</emphasis>. This Fetcher honors the variables <varname>DL_DIR</varname>, <varname>FETCHCOMMAND_wget</varname>, <varname>PREMIRRORS</varname>, <varname>MIRRORS</varname>. The <varname>DL_DIR</varname> defines where to store the fetched file, <varname>FETCHCOMMAND</varname> contains the command used for fetching. <quote>${URI}</quote> and <quote>${FILES}</quote> will be replaced by the uri and basename of the to be fetched file. <varname>PREMIRRORS</varname>
will be tried first when fetching a file if that fails the actual file will be tried and finally all <varname>MIRRORS</varname> will be tried.
</para>
<para>The only supported Parameter is <varname>md5sum</varname>. After a fetch the <varname>md5sum</varname> of the file will be calculated and the two sums will be compared.
</para>
<para><screen><varname>SRC_URI</varname> = "http://oe.handhelds.org/not_there.aac;md5sum=12343"
<varname>SRC_URI</varname> = "ftp://oe.handhelds.org/not_there_as_well.aac;md5sum=1234"
<varname>SRC_URI</varname> = "ftp://you@oe.handheld.sorg/home/you/secret.plan;md5sum=1234"
</screen></para>
</section>
<section>
<title>SVK Fetcher</title>
<para>
<emphasis>Currently NOT supported</emphasis>
</para>
</section>
<section>
<title>SVN Fetcher</title>
<para>The URN for the SVN Fetcher is <emphasis>svn</emphasis>.
</para>
<para>This Fetcher honors the variables <varname>FETCHCOMMAND_svn</varname>, <varname>DL_DIR</varname>, <varname>SRCDATE</varname>. <varname>FETCHCOMMAND</varname> contains the subversion command, <varname>DL_DIR</varname> is the directory where tarballs will be saved, <varname>SRCDATE</varname> specifies which date to use when doing the fetching (the special value of "now" will cause the checkout to be updated on every build).
</para>
<para>The supported Parameters are <varname>proto</varname>, <varname>rev</varname>. <varname>proto</varname> is the subversion prototype, <varname>rev</varname> is the subversions revision.
</para>
<para><screen><varname>SRC_URI</varname> = "svn://svn.oe.handhelds.org/svn;module=vip;proto=http;rev=667"
<varname>SRC_URI</varname> = "svn://svn.oe.handhelds.org/svn/;module=opie;proto=svn+ssh;date=20060126"
</screen></para>
</section>
<section>
<title>GIT Fetcher</title>
<para>The URN for the GIT Fetcher is <emphasis>git</emphasis>.
</para>
<para>The Variables <varname>DL_DIR</varname>, <varname>GITDIR</varname> are used. <varname>DL_DIR</varname> will be used to store the checkedout version. <varname>GITDIR</varname> will be used as the base directory where the git tree is cloned to.
</para>
<para>The Parameters are <emphasis>tag</emphasis>, <emphasis>protocol</emphasis>. <emphasis>tag</emphasis> is a git tag, the default is <quote>master</quote>. <emphasis>protocol</emphasis> is the git protocol to use and defaults to <quote>rsync</quote>.
</para>
<para><screen><varname>SRC_URI</varname> = "git://git.oe.handhelds.org/git/vip.git;tag=version-1"
<varname>SRC_URI</varname> = "git://git.oe.handhelds.org/git/vip.git;protocol=http"
</screen></para>
</section>
</chapter>
<chapter>
<title>Commands</title>
<section>
<title>bbread</title>
<para>bbread is a command for displaying BitBake metadata. When run with no arguments, it has the core parse 'conf/bitbake.conf', as located in BBPATH, and displays that. If you supply a file on the commandline, such as a .bb, then it parses that afterwards, using the aforementioned configuration metadata.</para>
<para><emphasis>NOTE: the stand a lone bbread command was removed. Instead of bbread use bitbake -e.
</emphasis></para>
</section>
<section>
<title>bitbake</title>
<section>
<title>Introduction</title>
<para>bitbake is the primary command in the system. It facilitates executing tasks in a single .bb file, or executing a given task on a set of multiple .bb files, accounting for interdependencies amongst them.</para>
</section>
<section>
<title>Usage and Syntax</title>
<para>
<screen><prompt>$ </prompt>bitbake --help
usage: bitbake [options] [package ...]
Executes the specified task (default is 'build') for a given set of BitBake files.
It expects that BBFILES is defined, which is a space seperated list of files to
be executed. BBFILES does support wildcards.
Default BBFILES are the .bb files in the current directory.
options:
--version show program's version number and exit
-h, --help show this help message and exit
-b BUILDFILE, --buildfile=BUILDFILE
execute the task against this .bb file, rather than a
package from BBFILES.
-k, --continue continue as much as possible after an error. While the
target that failed, and those that depend on it,
cannot be remade, the other dependencies of these
targets can be processed all the same.
-f, --force force run of specified cmd, regardless of stamp status
-i, --interactive drop into the interactive mode also called the BitBake
shell.
-c CMD, --cmd=CMD Specify task to execute. Note that this only executes
the specified task for the providee and the packages
it depends on, i.e. 'compile' does not implicitly call
stage for the dependencies (IOW: use only if you know
what you are doing). Depending on the base.bbclass a
listtasks tasks is defined and will show available
tasks
-r FILE, --read=FILE read the specified file before bitbake.conf
-v, --verbose output more chit-chat to the terminal
-D, --debug Increase the debug level. You can specify this more
than once.
-n, --dry-run don't execute, just go through the motions
-p, --parse-only quit after parsing the BB files (developers only)
-d, --disable-psyco disable using the psyco just-in-time compiler (not
recommended)
-s, --show-versions show current and preferred versions of all packages
-e, --environment show the global or per-package environment (this is
what used to be bbread)
-g, --graphviz emit the dependency trees of the specified packages in
the dot syntax
-I IGNORED_DOT_DEPS, --ignore-deps=IGNORED_DOT_DEPS
Stop processing at the given list of dependencies when
generating dependency graphs. This can help to make
the graph more appealing
-l DEBUG_DOMAINS, --log-domains=DEBUG_DOMAINS
Show debug logging for the specified logging domains
-P, --profile profile the command and print a report
</screen>
</para>
<para>
<example>
<title>Executing a task against a single .bb</title>
<para>Executing tasks for a single file is relatively simple. You specify the file in question, and bitbake parses it and executes the specified task (or <quote>build</quote> by default). It obeys intertask dependencies when doing so.</para>
<para><quote>clean</quote> task:</para>
<para><screen><prompt>$ </prompt>bitbake -b blah_1.0.bb -c clean</screen></para>
<para><quote>build</quote> task:</para>
<para><screen><prompt>$ </prompt>bitbake -b blah_1.0.bb</screen></para>
</example>
</para>
<para>
<example>
<title>Executing tasks against a set of .bb files</title>
<para>There are a number of additional complexities introduced when one wants to manage multiple .bb files. Clearly there needs to be a way to tell bitbake what files are available, and of those, which we want to execute at this time. There also needs to be a way for each .bb to express its dependencies, both for build time and runtime. There must be a way for the user to express their preferences when multiple .bb's provide the same functionality, or when there are multiple versions of a .bb.</para>
<para>The next section, Metadata, outlines how one goes about specifying such things.</para>
<para>Note that the bitbake command, when not using --buildfile, accepts a <varname>PROVIDER</varname>, not a filename or anything else. By default, a .bb generally PROVIDES its packagename, packagename-version, and packagename-version-revision.</para>
<screen><prompt>$ </prompt>bitbake blah</screen>
<screen><prompt>$ </prompt>bitbake blah-1.0</screen>
<screen><prompt>$ </prompt>bitbake blah-1.0-r0</screen>
<screen><prompt>$ </prompt>bitbake -c clean blah</screen>
<screen><prompt>$ </prompt>bitbake virtual/whatever</screen>
<screen><prompt>$ </prompt>bitbake -c clean virtual/whatever</screen>
</example>
<example>
<title>Generating dependency graphs</title>
<para>BitBake is able to generate dependency graphs using the dot syntax. These graphs can be converted
to images using the <application>dot</application> application from <ulink url="http://www.graphviz.org">graphviz</ulink>.
Two files will be written into the current working directory, <emphasis>depends.dot</emphasis> containing dependency information at the package level and <emphasis>task-depends.dot</emphasis> containing a breakdown of the dependencies at the task level. To stop depending on common depends one can use the <prompt>-I depend</prompt> to omit these from the graph. This can lead to more readable graphs. E.g. this way <varname>DEPENDS</varname> from inherited classes, e.g. base.bbclass, can be removed from the graph.</para>
<screen><prompt>$ </prompt>bitbake -g blah</screen>
<screen><prompt>$ </prompt>bitbake -g -I virtual/whatever -I bloom blah</screen>
</example>
</para>
</section>
<section>
<title>Special variables</title>
<para>Certain variables affect bitbake operation:</para>
<section>
<title><varname>BB_NUMBER_THREADS</varname></title>
<para> The number of threads bitbake should run at once (default: 1).</para>
</section>
</section>
<section>
<title>Metadata</title>
<para>As you may have seen in the usage information, or in the information about .bb files, the BBFILES variable is how the bitbake tool locates its files. This variable is a space seperated list of files that are available, and supports wildcards.
<example>
<title>Setting BBFILES</title>
<programlisting><varname>BBFILES</varname> = "/path/to/bbfiles/*.bb"</programlisting>
</example></para>
<para>With regard to dependencies, it expects the .bb to define a <varname>DEPENDS</varname> variable, which contains a space seperated list of <quote>package names</quote>, which themselves are the <varname>PN</varname> variable. The <varname>PN</varname> variable is, in general, by default, set to a component of the .bb filename.</para>
<example>
<title>Depending on another .bb</title>
<para>a.bb:
<screen>PN = "package-a"
DEPENDS += "package-b"</screen>
</para>
<para>b.bb:
<screen>PN = "package-b"</screen>
</para>
</example>
<example>
<title>Using PROVIDES</title>
<para>This example shows the usage of the PROVIDES variable, which allows a given .bb to specify what functionality it provides.</para>
<para>package1.bb:
<screen>PROVIDES += "virtual/package"</screen>
</para>
<para>package2.bb:
<screen>DEPENDS += "virtual/package"</screen>
</para>
<para>package3.bb:
<screen>PROVIDES += "virtual/package"</screen>
</para>
<para>As you can see, here there are two different .bb's that provide the same functionality (virtual/package). Clearly, there needs to be a way for the person running bitbake to control which of those providers gets used. There is, indeed, such a way.</para>
<para>The following would go into a .conf file, to select package1:
<screen>PREFERRED_PROVIDER_virtual/package = "package1"</screen>
</para>
</example>
<example>
<title>Specifying version preference</title>
<para>When there are multiple <quote>versions</quote> of a given package, bitbake defaults to selecting the most recent version, unless otherwise specified. If the .bb in question has a <varname>DEFAULT_PREFERENCE</varname> set lower than the other .bb's (default is 0), then it will not be selected. This allows the person or persons maintaining the repository of .bb files to specify their preferences for the default selected version. In addition, the user can specify their preferences with regard to version.</para>
<para>If the first .bb is named <filename>a_1.1.bb</filename>, then the <varname>PN</varname> variable will be set to <quote>a</quote>, and the <varname>PV</varname> variable will be set to 1.1.</para>
<para>If we then have an <filename>a_1.2.bb</filename>, bitbake will choose 1.2 by default. However, if we define the following variable in a .conf that bitbake parses, we can change that.
<screen>PREFERRED_VERSION_a = "1.1"</screen>
</para>
</example>
<example>
<title>Using <quote>bbfile collections</quote></title>
<para>bbfile collections exist to allow the user to have multiple repositories of bbfiles that contain the same exact package. For example, one could easily use them to make one's own local copy of an upstream repository, but with custom modifications that one does not want upstream. Usage:</para>
<screen>BBFILES = "/stuff/openembedded/*/*.bb /stuff/openembedded.modified/*/*.bb"
BBFILE_COLLECTIONS = "upstream local"
BBFILE_PATTERN_upstream = "^/stuff/openembedded/"
BBFILE_PATTERN_local = "^/stuff/openembedded.modified/"
BBFILE_PRIORITY_upstream = "5"
BBFILE_PRIORITY_local = "10"</screen>
</example>
</section>
</section>
</chapter>
</book>

View File

@@ -1,59 +0,0 @@
<!ENTITY DISTRO "1.4">
<!ENTITY DISTRO_NAME "tbd">
<!ENTITY YOCTO_DOC_VERSION "1.4">
<!ENTITY POKYVERSION "8.0">
<!ENTITY YOCTO_POKY "poky-&DISTRO_NAME;-&POKYVERSION;">
<!ENTITY COPYRIGHT_YEAR "2010-2013">
<!ENTITY YOCTO_DL_URL "http://downloads.yoctoproject.org">
<!ENTITY YOCTO_HOME_URL "http://www.yoctoproject.org">
<!ENTITY YOCTO_LISTS_URL "http://lists.yoctoproject.org">
<!ENTITY YOCTO_BUGZILLA_URL "http://bugzilla.yoctoproject.org">
<!ENTITY YOCTO_WIKI_URL "https://wiki.yoctoproject.org">
<!ENTITY YOCTO_AB_URL "http://autobuilder.yoctoproject.org">
<!ENTITY YOCTO_GIT_URL "http://git.yoctoproject.org">
<!ENTITY YOCTO_ADTREPO_URL "http://adtrepo.yoctoproject.org">
<!ENTITY OE_HOME_URL "http://www.openembedded.org">
<!ENTITY OE_LISTS_URL "http://lists.linuxtogo.org/cgi-bin/mailman">
<!ENTITY OE_DOCS_URL "http://docs.openembedded.org">
<!ENTITY OH_HOME_URL "http://o-hand.com">
<!ENTITY BITBAKE_HOME_URL "http://developer.berlios.de/projects/bitbake/">
<!ENTITY ECLIPSE_MAIN_URL "http://www.eclipse.org/downloads">
<!ENTITY ECLIPSE_DL_URL "http://download.eclipse.org">
<!ENTITY ECLIPSE_DL_PLUGIN_URL "&YOCTO_DL_URL;/releases/eclipse-plugin/&DISTRO;">
<!ENTITY ECLIPSE_UPDATES_URL "&ECLIPSE_DL_URL;/tm/updates/3.3">
<!ENTITY ECLIPSE_INDIGO_URL "&ECLIPSE_DL_URL;/releases/indigo">
<!ENTITY ECLIPSE_JUNO_URL "&ECLIPSE_DL_URL;/releases/juno">
<!ENTITY ECLIPSE_INDIGO_CDT_URL "&ECLIPSE_DL_URL;tools/cdt/releases/indigo">
<!ENTITY YOCTO_DOCS_URL "&YOCTO_HOME_URL;/docs">
<!ENTITY YOCTO_SOURCES_URL "&YOCTO_HOME_URL;/sources/">
<!ENTITY YOCTO_AB_PORT_URL "&YOCTO_AB_URL;:8010">
<!ENTITY YOCTO_AB_NIGHTLY_URL "&YOCTO_AB_URL;/nightly/">
<!ENTITY YOCTO_POKY_URL "&YOCTO_DL_URL;/releases/poky/">
<!ENTITY YOCTO_RELEASE_DL_URL "&YOCTO_DL_URL;/releases/yocto/yocto-&DISTRO;">
<!ENTITY YOCTO_TOOLCHAIN_DL_URL "&YOCTO_RELEASE_DL_URL;/toolchain/">
<!ENTITY YOCTO_ECLIPSE_DL_URL "&YOCTO_RELEASE_DL_URL;/eclipse-plugin/indigo;">
<!ENTITY YOCTO_ADTINSTALLER_DL_URL "&YOCTO_RELEASE_DL_URL;/adt_installer">
<!ENTITY YOCTO_POKY_DL_URL "&YOCTO_RELEASE_DL_URL;/&YOCTO_POKY;.tar.bz2">
<!ENTITY YOCTO_MACHINES_DL_URL "&YOCTO_RELEASE_DL_URL;/machines">
<!ENTITY YOCTO_QEMU_DL_URL "&YOCTO_MACHINES_DL_URL;/qemu">
<!ENTITY YOCTO_PYTHON-i686_DL_URL "&YOCTO_DL_URL;/releases/miscsupport/python-nativesdk-standalone-i686.tar.bz2">
<!ENTITY YOCTO_PYTHON-x86_64_DL_URL "&YOCTO_DL_URL;/releases/miscsupport/python-nativesdk-standalone-x86_64.tar.bz2">
<!ENTITY YOCTO_DOCS_QS_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/yocto-project-qs/yocto-project-qs.html">
<!ENTITY YOCTO_DOCS_ADT_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/adt-manual/adt-manual.html">
<!ENTITY YOCTO_DOCS_REF_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/ref-manual/ref-manual.html">
<!ENTITY YOCTO_DOCS_BSP_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/bsp-guide/bsp-guide.html">
<!ENTITY YOCTO_DOCS_DEV_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/dev-manual/dev-manual.html">
<!ENTITY YOCTO_DOCS_KERNEL_URL "&YOCTO_DOCS_URL;/&YOCTO_DOC_VERSION;/kernel-manual/kernel-manual.html">
<!ENTITY YOCTO_ADTPATH_DIR "/opt/poky/&DISTRO;">
<!ENTITY YOCTO_POKY_TARBALL "&YOCTO_POKY;.tar.bz2">
<!ENTITY OE_INIT_PATH "&YOCTO_POKY;/oe-init-build-env">
<!ENTITY OE_INIT_FILE "oe-init-build-env">
<!ENTITY UBUNTU_HOST_PACKAGES_ESSENTIAL "gawk wget git-core diffstat unzip texinfo \
build-essential chrpath">
<!ENTITY FEDORA_HOST_PACKAGES_ESSENTIAL "gawk make wget tar bzip2 gzip python unzip perl patch \
diffutils diffstat git cpp gcc gcc-c++ eglibc-devel texinfo chrpath \
ccache">
<!ENTITY OPENSUSE_HOST_PACKAGES_ESSENTIAL "python gcc gcc-c++ git chrpath make wget python-xml \
diffstat texinfo python-curses">
<!ENTITY CENTOS_HOST_PACKAGES_ESSENTIAL "gawk make wget tar bzip2 gzip python unzip perl patch \
diffutils diffstat git cpp gcc gcc-c++ glibc-devel texinfo chrpath">

View File

@@ -1,64 +0,0 @@
<?xml version='1.0'?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" xmlns="http://www.w3.org/1999/xhtml" xmlns:fo="http://www.w3.org/1999/XSL/Format" version="1.0">
<xsl:import href="http://docbook.sourceforge.net/release/xsl/current/fo/docbook.xsl" />
<!-- check project-plan.sh for how this is generated, needed to tweak
the cover page
-->
<xsl:include href="/tmp/titlepage.xsl"/>
<!-- To force a page break in document, i.e per section add a
<?hard-pagebreak?> tag.
-->
<xsl:template match="processing-instruction('hard-pagebreak')">
<fo:block break-before='page' />
</xsl:template>
<!--Fix for defualt indent getting TOC all wierd..
See http://sources.redhat.com/ml/docbook-apps/2005-q1/msg00455.html
FIXME: must be a better fix
-->
<xsl:param name="body.start.indent" select="'0'"/>
<!--<xsl:param name="title.margin.left" select="'0'"/>-->
<!-- stop long-ish header titles getting wrapped -->
<xsl:param name="header.column.widths">1 10 1</xsl:param>
<!-- customise headers and footers a little -->
<xsl:template name="head.sep.rule">
<xsl:if test="$header.rule != 0">
<xsl:attribute name="border-bottom-width">0.5pt</xsl:attribute>
<xsl:attribute name="border-bottom-style">solid</xsl:attribute>
<xsl:attribute name="border-bottom-color">#cccccc</xsl:attribute>
</xsl:if>
</xsl:template>
<xsl:template name="foot.sep.rule">
<xsl:if test="$footer.rule != 0">
<xsl:attribute name="border-top-width">0.5pt</xsl:attribute>
<xsl:attribute name="border-top-style">solid</xsl:attribute>
<xsl:attribute name="border-top-color">#cccccc</xsl:attribute>
</xsl:if>
</xsl:template>
<xsl:attribute-set name="header.content.properties">
<xsl:attribute name="color">#cccccc</xsl:attribute>
</xsl:attribute-set>
<xsl:attribute-set name="footer.content.properties">
<xsl:attribute name="color">#cccccc</xsl:attribute>
</xsl:attribute-set>
<!-- general settings -->
<xsl:param name="fop1.extensions" select="1"></xsl:param>
<xsl:param name="paper.type" select="'A4'"></xsl:param>
<xsl:param name="section.autolabel" select="1"></xsl:param>
<xsl:param name="body.font.family" select="'verasans'"></xsl:param>
<xsl:param name="title.font.family" select="'verasans'"></xsl:param>
<xsl:param name="monospace.font.family" select="'veramono'"></xsl:param>
</xsl:stylesheet>

View File

@@ -1,58 +0,0 @@
<fop version="1.0">
<!-- Strict user configuration -->
<strict-configuration>true</strict-configuration>
<!-- Strict FO validation -->
<strict-validation>true</strict-validation>
<!--
Set the baseDir so common/openedhand.svg references in plans still
work ok. Note, relative file references to current dir should still work.
-->
<base>../template</base>
<font-base>../template</font-base>
<!-- Source resolution in dpi (dots/pixels per inch) for determining the
size of pixels in SVG and bitmap images, default: 72dpi -->
<!-- <source-resolution>72</source-resolution> -->
<!-- Target resolution in dpi (dots/pixels per inch) for specifying the
target resolution for generated bitmaps, default: 72dpi -->
<!-- <target-resolution>72</target-resolution> -->
<!-- default page-height and page-width, in case
value is specified as auto -->
<default-page-settings height="11in" width="8.26in"/>
<!-- <use-cache>false</use-cache> -->
<renderers>
<renderer mime="application/pdf">
<fonts>
<font metrics-file="VeraMono.xml"
kerning="yes"
embed-url="VeraMono.ttf">
<font-triplet name="veramono" style="normal" weight="normal"/>
</font>
<font metrics-file="VeraMoBd.xml"
kerning="yes"
embed-url="VeraMoBd.ttf">
<font-triplet name="veramono" style="normal" weight="bold"/>
</font>
<font metrics-file="Vera.xml"
kerning="yes"
embed-url="Vera.ttf">
<font-triplet name="verasans" style="normal" weight="normal"/>
<font-triplet name="verasans" style="normal" weight="bold"/>
<font-triplet name="verasans" style="italic" weight="normal"/>
<font-triplet name="verasans" style="italic" weight="bold"/>
</font>
<auto-detect/>
</fonts>
</renderer>
</renderers>
</fop>

File diff suppressed because it is too large Load Diff

View File

@@ -1,51 +0,0 @@
#!/bin/sh
if [ -z "$1" -o -z "$2" ]; then
echo "usage: [-v] $0 <docbook file> <templatedir>"
echo
echo "*NOTE* you need xsltproc, fop and nwalsh docbook stylesheets"
echo " installed for this to work!"
echo
exit 0
fi
FO=`echo $1 | sed s/.xml/.fo/` || exit 1
PDF=`echo $1 | sed s/.xml/.pdf/` || exit 1
TEMPLATEDIR=$2
##
# These URI should be rewritten by your distribution's xml catalog to
# match your localy installed XSL stylesheets.
XSL_BASE_URI="http://docbook.sourceforge.net/release/xsl/current"
# Creates a temporary XSL stylesheet based on titlepage.xsl
xsltproc -o /tmp/titlepage.xsl \
--xinclude \
$XSL_BASE_URI/template/titlepage.xsl \
$TEMPLATEDIR/titlepage.templates.xml || exit 1
# Creates the file needed for FOP
xsltproc --xinclude \
--stringparam hyphenate false \
--stringparam formal.title.placement "figure after" \
--stringparam ulink.show 1 \
--stringparam body.font.master 9 \
--stringparam title.font.master 11 \
--stringparam draft.watermark.image "$TEMPLATEDIR/draft.png" \
--stringparam chapter.autolabel 1 \
--stringparam appendix.autolabel A \
--stringparam section.autolabel 1 \
--stringparam section.label.includes.component.label 1 \
--output $FO \
$TEMPLATEDIR/db-pdf.xsl \
$1 || exit 1
# Invokes the Java version of FOP. Uses the additional configuration file common/fop-config.xml
fop -c $TEMPLATEDIR/fop-config.xml -fo $FO -pdf $PDF || exit 1
rm -f $FO
rm -f /tmp/titlepage.xsl
echo
echo " #### Success! $PDF ready. ####"
echo

View File

@@ -3,7 +3,7 @@
#
# This is a copy on write dictionary and set which abuses classes to try and be nice and fast.
#
# Copyright (C) 2006 Tim Amsell
# Copyright (C) 2006 Tim Amsell
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License version 2 as
@@ -18,31 +18,31 @@
# with this program; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#Please Note:
#Please Note:
# Be careful when using mutable types (ie Dict and Lists) - operations involving these are SLOW.
# Assign a file to __warn__ to get warnings about slow operations.
#
from __future__ import print_function
from inspect import getmro
import copy
import types
ImmutableTypes = (
types.NoneType,
bool,
complex,
float,
int,
long,
tuple,
frozenset,
basestring
)
types.ImmutableTypes = tuple([ \
types.BooleanType, \
types.ComplexType, \
types.FloatType, \
types.IntType, \
types.LongType, \
types.NoneType, \
types.TupleType, \
frozenset] + \
list(types.StringTypes))
MUTABLE = "__mutable__"
class COWMeta(type):
pass
class COWDictMeta(COWMeta):
__warn__ = False
__hasmutable__ = False
@@ -61,12 +61,12 @@ class COWDictMeta(COWMeta):
__call__ = cow
def __setitem__(cls, key, value):
if not isinstance(value, ImmutableTypes):
if not isinstance(value, types.ImmutableTypes):
if not isinstance(value, COWMeta):
cls.__hasmutable__ = True
key += MUTABLE
setattr(cls, key, value)
def __getmutable__(cls, key, readonly=False):
nkey = key + MUTABLE
try:
@@ -79,10 +79,10 @@ class COWDictMeta(COWMeta):
return value
if not cls.__warn__ is False and not isinstance(value, COWMeta):
print("Warning: Doing a copy because %s is a mutable type." % key, file=cls.__warn__)
print >> cls.__warn__, "Warning: Doing a copy because %s is a mutable type." % key
try:
value = value.copy()
except AttributeError as e:
except AttributeError, e:
value = copy.copy(value)
setattr(cls, nkey, value)
return value
@@ -100,13 +100,13 @@ class COWDictMeta(COWMeta):
value = getattr(cls, key)
except AttributeError:
value = cls.__getmutable__(key, readonly)
# This is for values which have been deleted
# This is for values which have been deleted
if value is cls.__marker__:
raise AttributeError("key %s does not exist." % key)
return value
except AttributeError as e:
except AttributeError, e:
if not default is cls.__getmarker__:
return default
@@ -120,9 +120,6 @@ class COWDictMeta(COWMeta):
key += MUTABLE
delattr(cls, key)
def __contains__(cls, key):
return cls.has_key(key)
def has_key(cls, key):
value = cls.__getreadonly__(key, cls.__marker__)
if value is cls.__marker__:
@@ -132,7 +129,7 @@ class COWDictMeta(COWMeta):
def iter(cls, type, readonly=False):
for key in dir(cls):
if key.startswith("__"):
continue
continue
if key.endswith(MUTABLE):
key = key[:-len(MUTABLE)]
@@ -158,11 +155,11 @@ class COWDictMeta(COWMeta):
return cls.iter("keys")
def itervalues(cls, readonly=False):
if not cls.__warn__ is False and cls.__hasmutable__ and readonly is False:
print("Warning: If you arn't going to change any of the values call with True.", file=cls.__warn__)
print >> cls.__warn__, "Warning: If you arn't going to change any of the values call with True."
return cls.iter("values", readonly)
def iteritems(cls, readonly=False):
if not cls.__warn__ is False and cls.__hasmutable__ and readonly is False:
print("Warning: If you arn't going to change any of the values call with True.", file=cls.__warn__)
print >> cls.__warn__, "Warning: If you arn't going to change any of the values call with True."
return cls.iter("items", readonly)
class COWSetMeta(COWDictMeta):
@@ -181,13 +178,13 @@ class COWSetMeta(COWDictMeta):
def remove(cls, value):
COWDictMeta.__delitem__(cls, repr(hash(value)))
def __in__(cls, value):
return COWDictMeta.has_key(repr(hash(value)))
def iterkeys(cls):
raise TypeError("sets don't have keys")
def iteritems(cls):
raise TypeError("sets don't have 'items'")
@@ -204,120 +201,120 @@ if __name__ == "__main__":
import sys
COWDictBase.__warn__ = sys.stderr
a = COWDictBase()
print("a", a)
print "a", a
a['a'] = 'a'
a['b'] = 'b'
a['dict'] = {}
b = a.copy()
print("b", b)
print "b", b
b['c'] = 'b'
print()
print
print("a", a)
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems():
print(x)
print()
print x
print
b['dict']['a'] = 'b'
b['a'] = 'c'
print("a", a)
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems():
print(x)
print()
print x
print
try:
b['dict2']
except KeyError as e:
print("Okay!")
except KeyError, e:
print "Okay!"
a['set'] = COWSetBase()
a['set'].add("o1")
a['set'].add("o1")
a['set'].add("o2")
print("a", a)
print "a", a
for x in a['set'].itervalues():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b['set'].itervalues():
print(x)
print()
print x
print
b['set'].add('o3')
print("a", a)
print "a", a
for x in a['set'].itervalues():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b['set'].itervalues():
print(x)
print()
print x
print
a['set2'] = set()
a['set2'].add("o1")
a['set2'].add("o1")
a['set2'].add("o2")
print("a", a)
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print(x)
print()
print x
print
del b['b']
try:
print(b['b'])
print b['b']
except KeyError:
print("Yay! deleted key raises error")
print "Yay! deleted key raises error"
if b.has_key('b'):
print("Boo!")
print "Boo!"
else:
print("Yay - has_key with delete works!")
print("a", a)
print "Yay - has_key with delete works!"
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print(x)
print()
print x
print
b.__revertitem__('b')
print("a", a)
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print(x)
print()
print x
print
b.__revertitem__('dict')
print("a", a)
print "a", a
for x in a.iteritems():
print(x)
print("--")
print("b", b)
print x
print "--"
print "b", b
for x in b.iteritems(readonly=True):
print(x)
print()
print x
print

File diff suppressed because it is too large Load Diff

Some files were not shown because too many files have changed in this diff Show More