bitbake: toaster/contrib: adding TTS squashed patch

In order to move the Toaster Test System in Toaster itself,
we create a contrib directory.

The TTS is added as a squashed patch with no history.

It contains code contributed by Ke Zou <ke.zou@windriver.com>.

(Bitbake rev: 7d24fea2b5dcaac6add738b6fb4700d698824286)

Signed-off-by: Alexandru DAMIAN <alexandru.damian@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This commit is contained in:
Alexandru DAMIAN
2015-05-13 13:21:33 +01:00
committed by Richard Purdie
parent 23f5b009e0
commit 80ca4f00f8
18 changed files with 2829 additions and 0 deletions

View File

@@ -0,0 +1,6 @@
contrib directory for toaster
This directory holds code that works with Toaster, without being an integral part of the Toaster project.
It is intended for testing code, testing fixtures, tools for Toaster, etc.
NOTE: This directory is NOT a Python module.

View File

@@ -0,0 +1,41 @@
Toaster Testing Framework
Yocto Project
Rationale
------------
As Toaster contributions grow with the number of people that contribute code, verifying each patch prior to submitting upstream becomes a hard-to-scale problem for humans. We devised this system in order to run patch-level validation, trying to eliminate common problems from submitted patches, in an automated fashion.
The Toaster Testing Framework is a set of Python scripts that provides an extensible way to write smoke and regression tests that will be run on each patch set sent for review on the toaster mailing list.
Usage
------------
There are three main executable scripts in this directory.
* runner.py is designed to be run from the command line. It requires, as mandatory parameter, a branch name on poky-contrib, branch which contains the patches to be tested. The program will auto-discover the available tests residing in this directory by looking for unittest classes, and will run the tests on the branch dumping the output to the standard output. Optionally, it can take parameters inhibiting the branch checkout, or specifying a single test to be run, for debugging purposes.
* launcher.py is a designed to be run from a crontab or similar scheduling mechanism. It looks up a backlog file containing branches-to-test (named tasks in the source code), select the first one in FIFO manner, and launch runner.py on it. It will await for completion, and email the standard output and standard error dumps from the runner.py execution
* recv.py is an email receiver, designed to be called as a pipe from a .forward file. It is used to monitor a mailing list, for example, and add tasks to the backlog based on review requests coming on the mailing list.
Installation
------------
As prerequisite, we expect a functioning email system on a machine with Python2.
The broad steps to installation
* set up the .forward on the receiving email account to pipe to the recv.py file
* edit config.py and settings.json to alter for local installation settings
* on email receive, verify backlog.txt to see that the tasks are received and marked for processing
* execute launcher.py in command line to verify that a test occurs with no problems, and that the outgoing email is delivered
* add launcher.py
Contribute
------------
What we need are tests. Add your own tests to either tests.py file, or to a new file.
Use "config.logger" to write logs that will make it to email.
Commonly used code should be going to shellutils, and configuration to config.py.
Contribute code by emailing patches to the list: toaster@yoctoproject.org (membership required)

View File

@@ -0,0 +1,9 @@
We need to implement tests:
automated link checker; currently
$ linkchecker -t 1000 -F csv http://localhost:8000/
integrate the w3c-validation service; currently
$ python urlcheck.py

View File

@@ -0,0 +1,2 @@
michaelw/toaster-frontend-cleanups|PENDING
adamian/test_0|PENDING

View File

@@ -0,0 +1,70 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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 the configuration/single module for tts
# everything that would be a global variable goes here
import os, sys, logging
LOGDIR = "log"
SETTINGS_FILE = os.path.join(os.path.dirname(__file__), "settings.json")
TEST_DIR_NAME = "tts_testdir"
OWN_PID = os.getpid()
OWN_EMAIL_ADDRESS = "Toaster Testing Framework <alexandru.damian@intel.com>"
REPORT_EMAIL_ADDRESS = "alexandru.damian@intel.com"
# make sure we have the basic logging infrastructure
logger = logging.getLogger("toastertest")
__console = logging.StreamHandler(sys.stdout)
__console.setFormatter(logging.Formatter("%(asctime)s %(levelname)s: %(message)s"))
logger.addHandler(__console)
logger.setLevel(logging.DEBUG)
# singleton file names
LOCKFILE="/tmp/ttf.lock"
BACKLOGFILE=os.path.join(os.path.dirname(__file__), "backlog.txt")
# task states
def enum(*sequential, **named):
enums = dict(zip(sequential, range(len(sequential))), **named)
reverse = dict((value, key) for key, value in enums.iteritems())
enums['reverse_mapping'] = reverse
return type('Enum', (), enums)
class TASKS:
PENDING = "PENDING"
INPROGRESS = "INPROGRESS"
DONE = "DONE"
@staticmethod
def next_task(task):
if task == TASKS.PENDING:
return TASKS.INPROGRESS
if task == TASKS.INPROGRESS:
return TASKS.DONE
raise Exception("Invalid next task state for %s" % task)
# TTS specific
CONTRIB_REPO = "git@git.yoctoproject.org:poky-contrib"

View File

@@ -0,0 +1,100 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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.
# Program to run the next task listed from the backlog.txt; designed to be
# run from crontab.
from __future__ import print_function
import sys, os, config, shellutils
from shellutils import ShellCmdException
# Import smtplib for the actual sending function
import smtplib
# Import the email modules we'll need
from email.mime.text import MIMEText
DEBUG=True
def _take_lockfile():
return shellutils.lockfile(shellutils.mk_lock_filename())
def read_next_task_by_state(task_state, task_name = None):
if not os.path.exists(os.path.join(os.path.dirname(__file__), config.BACKLOGFILE)):
return None
os.rename(config.BACKLOGFILE, config.BACKLOGFILE + ".tmp")
task = None
with open(config.BACKLOGFILE + ".tmp", "r") as f_in:
with open(config.BACKLOGFILE, "w") as f_out:
for line in f_in.readlines():
if task is None:
fields = line.strip().split("|", 2)
if fields[1] == task_state:
if task_name is None or task_name == fields[0]:
task = fields[0]
print("Updating %s %s to %s" % (task, task_state, config.TASKS.next_task(task_state)))
line = "%s|%s\n" % (task, config.TASKS.next_task(task_state))
f_out.write(line)
os.remove(config.BACKLOGFILE + ".tmp")
return task
def send_report(task_name, plaintext, errtext = None):
if errtext is None:
msg = MIMEText(plaintext)
else:
if plaintext is None:
plaintext=""
msg = MIMEText("--STDOUT dump--\n\n%s\n\n--STDERR dump--\n\n%s" % (plaintext, errtext))
msg['Subject'] = "[review-request] %s - smoke test results" % task_name
msg['From'] = config.OWN_EMAIL_ADDRESS
msg['To'] = config.REPORT_EMAIL_ADDRESS
s = smtplib.SMTP("localhost")
s.sendmail(config.OWN_EMAIL_ADDRESS, [config.REPORT_EMAIL_ADDRESS], msg.as_string())
s.quit()
if __name__ == "__main__":
# we don't do anything if we have another instance of us running
lf = _take_lockfile()
if lf is None:
if DEBUG:
print("Concurrent script in progress, exiting")
sys.exit(1)
next_task = read_next_task_by_state(config.TASKS.PENDING)
if next_task is not None:
print("Next task is", next_task)
errtext = None
out = None
try:
out = shellutils.run_shell_cmd("./runner.py %s" % next_task)
pass
except ShellCmdException as e:
print("Failed while running the test runner: %s", e)
errtext = e.__str__()
send_report(next_task, out, errtext)
read_next_task_by_state(config.TASKS.INPROGRESS, next_task)
else:
print("No task")
shellutils.unlockfile(lf)

View File

@@ -0,0 +1,51 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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.
# Program to receive review requests by email and log tasks to backlog.txt
# Designed to be run by the email system from a .forward file:
#
# cat .forward
# |[full/path]/recv.py
from __future__ import print_function
import sys, os, config, shellutils
from shellutils import ShellCmdException
from email.parser import Parser
def recv_mail(datastring):
headers = Parser().parsestr(datastring)
return headers['subject']
if __name__ == "__main__":
lf = shellutils.lockfile(shellutils.mk_lock_filename(), retry = True)
subject = recv_mail(sys.stdin.read())
subject_parts = subject.split()
if "[review-request]" in subject_parts:
task_name = subject_parts[subject_parts.index("[review-request]") + 1]
with open(os.path.join(os.path.dirname(__file__), config.BACKLOGFILE), "a") as fout:
line = "%s|%s\n" % (task_name, config.TASKS.PENDING)
fout.write(line)
shellutils.unlockfile(lf)

View File

@@ -0,0 +1,200 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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 the main test execution controller. It is designed to be run
# manually from the command line, or to be called from a different program
# that schedules test execution.
#
# Execute runner.py -h for help.
from __future__ import print_function
import optparse
import sys, os
import unittest, inspect, importlib
import logging, pprint, json
from shellutils import *
import config
# we also log to a file, in addition to console, because our output is important
__log_file_name =os.path.join(os.path.dirname(__file__),"log/tts_%d.log" % config.OWN_PID)
mkdirhier(os.path.dirname(__log_file_name))
__log_file = open(__log_file_name, "w")
__file_handler = logging.StreamHandler(__log_file)
__file_handler.setFormatter(logging.Formatter("%(asctime)s %(levelname)s: %(message)s"))
config.logger.addHandler(__file_handler)
# set up log directory
try:
if not os.path.exists(config.LOGDIR):
os.mkdir(config.LOGDIR)
else:
if not os.path.isdir(config.LOGDIR):
raise Exception("Expected log dir '%s' is not actually a directory." % config.LOGDIR)
except OSError as e:
raise e
# creates the under-test-branch as a separate directory
def set_up_test_branch(settings, branch_name):
testdir = "%s/%s.%d" % (settings['workdir'], config.TEST_DIR_NAME, config.OWN_PID)
# creates the host dir
if os.path.exists(testdir):
raise Exception("Test dir '%s'is already there, aborting" % testdir)
os.mkdir(testdir)
# copies over the .git from the localclone
run_shell_cmd("cp -a '%s'/.git '%s'" % (settings['localclone'], testdir))
# add the remote if it doesn't exist
crt_remotes = run_shell_cmd("git remote -v", cwd = testdir)
remotes = [word for line in crt_remotes.split("\n") for word in line.split()]
if not config.CONTRIB_REPO in remotes:
remote_name = "tts_contrib"
run_shell_cmd("git remote add %s %s" % (remote_name, config.CONTRIB_REPO), cwd = testdir)
else:
remote_name = remotes[remotes.index(config.CONTRIB_REPO) - 1]
# do the fetch
run_shell_cmd("git fetch %s -p" % remote_name, cwd=testdir)
# do the checkout
run_shell_cmd("git checkout origin/master && git branch -D %s; git checkout %s/%s -b %s && git reset --hard" % (branch_name,remote_name,branch_name,branch_name), cwd=testdir)
return testdir
def __search_for_tests():
# we find all classes that can run, and run them
tests = []
for dir_name, dirs_list, files_list in os.walk(os.path.dirname(os.path.abspath(__file__))):
for f in [f[:-3] for f in files_list if f.endswith(".py") and not f.startswith("__init__")]:
config.logger.debug("Inspecting module %s", f)
current_module = importlib.import_module(f)
crtclass_names = vars(current_module)
for v in crtclass_names:
t = crtclass_names[v]
if isinstance(t, type(unittest.TestCase)) and issubclass(t, unittest.TestCase):
tests.append((f,v))
break
return tests
# boilerplate to self discover tests and run them
def execute_tests(dir_under_test, testname):
if testname is not None and "." in testname:
tests = []
tests.append(tuple(testname.split(".", 2)))
else:
tests = __search_for_tests()
# let's move to the directory under test
crt_dir = os.getcwd()
os.chdir(dir_under_test)
# execute each module
try:
config.logger.debug("Discovered test clases: %s" % pprint.pformat(tests))
suite = unittest.TestSuite()
loader = unittest.TestLoader()
result = unittest.TestResult()
for m,t in tests:
suite.addTest(loader.loadTestsFromName("%s.%s" % (m,t)))
config.logger.info("Running %d test(s)", suite.countTestCases())
suite.run(result)
if len(result.errors) > 0:
map(lambda x: config.logger.error("Exception on test: %s" % pprint.pformat(x)), result.errors)
if len(result.failures) > 0:
map(lambda x: config.logger.error("Failed test: %s:\n%s\n" % (pprint.pformat(x[0]), "\n".join(["-- %s" % x for x in eval(pprint.pformat(x[1])).split("\n")]))), result.failures)
config.logger.info("Test results: %d ran, %d errors, %d failures" % (result.testsRun, len(result.errors), len(result.failures)))
except Exception as e:
import traceback
config.logger.error("Exception while running test. Tracedump: \n%s", traceback.format_exc(e))
finally:
os.chdir(crt_dir)
return len(result.failures)
# verify that we had a branch-under-test name as parameter
def validate_args():
from optparse import OptionParser
parser = OptionParser(usage="usage: %prog [options] branch_under_test")
parser.add_option("-t", "--test-dir", dest="testdir", default=None, help="Use specified directory to run tests, inhibits the checkout.")
parser.add_option("-s", "--single", dest="singletest", default=None, help="Run only the specified test")
(options, args) = parser.parse_args()
if len(args) < 1:
raise Exception("Please specify the branch to run on. Use option '-h' when in doubt.")
return (options, args)
# load the configuration options
def read_settings():
if not os.path.exists(config.SETTINGS_FILE) or not os.path.isfile(config.SETTINGS_FILE):
raise Exception("Config file '%s' cannot be openend" % config.SETTINGS_FILE);
return json.loads(open(config.SETTINGS_FILE, "r").read())
# cleanup !
def clean_up(testdir):
# TODO: delete the test dir
#run_shell_cmd("rm -rf -- '%s'" % testdir)
pass
if __name__ == "__main__":
(options, args) = validate_args()
settings = read_settings()
need_cleanup = False
testdir = None
no_failures = 1
try:
if options.testdir is not None and os.path.exists(options.testdir):
testdir = options.testdir
config.logger.info("No checkout, using %s" % testdir)
else:
need_cleanup = True
testdir = set_up_test_branch(settings, args[0]) # we expect a branch name as first argument
config.testdir = testdir # we let tests know where to run
no_failures = execute_tests(testdir, options.singletest)
except ShellCmdException as e :
import traceback
config.logger.error("Error while setting up testing. Traceback: \n%s" % traceback.format_exc(e))
finally:
if need_cleanup and testdir is not None:
clean_up(testdir)
sys.exit(no_failures)

View File

@@ -0,0 +1,20 @@
import config
# Code testing section
def _code_test():
def callback_writeeventlog(opt, opt_str, value, parser):
if len(parser.rargs) < 1 or parser.rargs[0].startswith("-"):
value = ""
else:
value = parser.rargs[0]
del parser.rargs[0]
setattr(parser.values, opt.dest, value)
parser = optparse.OptionParser()
parser.add_option("-w", "--write-log", help = "Writes the event log of the build to a bitbake event json file.",
action = "callback", callback=callback_writeeventlog, dest = "writeeventlog")
options, targets = parser.parse_args(sys.argv)
print (options, targets)

View File

@@ -0,0 +1,5 @@
{
"repo": "git@git.yoctoproject.org:poky-contrib",
"localclone": "/home/ddalex/ssd/yocto/poky",
"workdir": "/home/ddalex/ssd/yocto"
}

View File

@@ -0,0 +1,139 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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.
# Utilities shared by tests and other common bits of code.
import sys, os, subprocess, fcntl, errno
import config
from config import logger
# License warning; this code is copied from the BitBake project, file bitbake/lib/bb/utils.py
# The code is originally licensed GPL-2.0, and we redistribute it under still GPL-2.0
# End of copy is marked with #ENDOFCOPY marker
def mkdirhier(directory):
"""Create a directory like 'mkdir -p', but does not complain if
directory already exists like os.makedirs
"""
try:
os.makedirs(directory)
except OSError as e:
if e.errno != errno.EEXIST:
raise e
def lockfile(name, shared=False, retry=True):
"""
Use the file fn as a lock file, return when the lock has been acquired.
Returns a variable to pass to unlockfile().
"""
config.logger.debug("take lockfile %s" % name)
dirname = os.path.dirname(name)
mkdirhier(dirname)
if not os.access(dirname, os.W_OK):
logger.error("Unable to acquire lock '%s', directory is not writable",
name)
sys.exit(1)
op = fcntl.LOCK_EX
if shared:
op = fcntl.LOCK_SH
if not retry:
op = op | fcntl.LOCK_NB
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+')
fileno = lf.fileno()
fcntl.flock(fileno, op)
statinfo = os.fstat(fileno)
if os.path.exists(lf.name):
statinfo2 = os.stat(lf.name)
if statinfo.st_ino == statinfo2.st_ino:
return lf
lf.close()
except Exception:
try:
lf.close()
except Exception:
pass
pass
if not retry:
return None
def unlockfile(lf):
"""
Unlock a file locked using lockfile()
"""
try:
# If we had a shared lock, we need to promote to exclusive before
# removing the lockfile. Attempt this, ignore failures.
fcntl.flock(lf.fileno(), fcntl.LOCK_EX|fcntl.LOCK_NB)
os.unlink(lf.name)
except (IOError, OSError):
pass
fcntl.flock(lf.fileno(), fcntl.LOCK_UN)
lf.close()
#ENDOFCOPY
def mk_lock_filename():
our_name = os.path.basename(__file__)
our_name = ".%s" % ".".join(reversed(our_name.split(".")))
return config.LOCKFILE + our_name
class ShellCmdException(Exception):
pass
def run_shell_cmd(command, cwd = None):
if cwd is None:
cwd = os.getcwd()
config.logger.debug("_shellcmd: (%s) %s" % (cwd, command))
p = subprocess.Popen(command, cwd = cwd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
(out,err) = p.communicate()
p.wait()
if p.returncode:
if len(err) == 0:
err = "command: %s \n%s" % (command, out)
else:
err = "command: %s \n%s" % (command, err)
config.logger.warn("_shellcmd: error \n%s\n%s" % (out, err))
raise ShellCmdException(err)
else:
#config.logger.debug("localhostbecontroller: shellcmd success\n%s" % out)
return out

View File

@@ -0,0 +1,57 @@
#!/usr/bin/python
# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
#
# Copyright (C) 2015 Alexandru Damian for Intel Corp.
#
# 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.
# Test definitions. The runner will look for and auto-discover the tests
# no matter what they file are they in, as long as they are in the same directory
# as this file.
import unittest
from shellutils import *
class TestPyCompilable(unittest.TestCase):
''' Verifies that all Python files are syntactically correct '''
def test_compile_file(self):
try:
out = run_shell_cmd("find . -name *py -type f -print0 | xargs -0 -n1 -P20 python -m py_compile", config.testdir)
except ShellCmdException as e:
self.fail("Error compiling python files: %s" % (e))
except Exception as e:
self.fail("Unknown error: %s" % e)
class TestPySystemStart(unittest.TestCase):
''' Attempts to start Toaster, verify that it is succesfull, and stop it '''
def setUp(self):
run_shell_cmd("bash -c 'rm -f build/*log'")
def test_start_interactive_mode(self):
try:
run_shell_cmd("bash -c 'source %s/oe-init-build-env && source toaster start && source toaster stop'" % config.testdir, config.testdir)
except ShellCmdException as e:
self.fail("Failed starting interactive mode: %s" % (e))
def test_start_managed_mode(self):
try:
run_shell_cmd("./poky/bitbake/bin/toaster webport=56789 & sleep 10 && curl http://localhost:56789/ && kill -2 %1")
pass
except ShellCmdException as e:
self.fail("Failed starting managed mode: %s" % (e))

View File

@@ -0,0 +1,87 @@
#!/usr/bin/python
# Copyright
# DESCRIPTION
# This is script for running all selected toaster cases on
# selected web browsers manifested in toaster_test.cfg.
# 1. How to start toaster in yocto:
# $ source poky/oe-init-build-env
# $ source toaster start
# $ bitbake core-image-minimal
# 2. How to install selenium on Ubuntu:
# $ sudo apt-get install scrot python-pip
# $ sudo pip install selenium
# 3. How to install selenium addon in firefox:
# Download the lastest firefox addon from http://release.seleniumhq.org/selenium-ide/
# Then install it. You can also install firebug and firepath addon
# 4. How to start writing a new case:
# All you need to do is to implement the function test_xxx() and pile it on.
# 5. How to test with Chrome browser
# Download/install chrome on host
# Download chromedriver from https://code.google.com/p/chromedriver/downloads/list according to your host type
# put chromedriver in PATH, (e.g. /usr/bin/, bear in mind to chmod)
# For windows host, you may put chromedriver.exe in the same directory as chrome.exe
import unittest, time, re, sys, getopt, os, logging, platform
import ConfigParser
import subprocess
class toaster_run_all():
def __init__(self):
# in case this script is called from other directory
os.chdir(os.path.abspath(sys.path[0]))
self.starttime = time.strptime(time.ctime())
self.parser = ConfigParser.SafeConfigParser()
found = self.parser.read('toaster_test.cfg')
self.host_os = platform.system().lower()
self.run_all_cases()
self.collect_log()
def get_test_cases(self):
# we have config groups for different os type in toaster_test.cfg
cases_to_run = eval(self.parser.get('toaster_test_' + self.host_os, 'test_cases'))
return cases_to_run
def run_all_cases(self):
cases_temp = self.get_test_cases()
for case in cases_temp:
single_case_cmd = "python -m unittest toaster_automation_test.toaster_cases.test_" + str(case)
print single_case_cmd
subprocess.call(single_case_cmd, shell=True)
def collect_log(self):
"""
the log files are temporarily stored in ./log/tmp/..
After all cases are done, they should be transfered to ./log/$TIMESTAMP/
"""
def comple(number):
if number < 10:
return str(0) + str(number)
else:
return str(number)
now = self.starttime
now_str = comple(now.tm_year) + comple(now.tm_mon) + comple(now.tm_mday) + \
comple(now.tm_hour) + comple(now.tm_min) + comple(now.tm_sec)
log_dir = os.path.abspath(sys.path[0]) + os.sep + 'log' + os.sep + now_str
log_tmp_dir = os.path.abspath(sys.path[0]) + os.sep + 'log' + os.sep + 'tmp'
try:
os.renames(log_tmp_dir, log_dir)
except OSError :
logging.error(" Cannot create log dir(timestamp) under log, please check your privilege")
if __name__ == "__main__":
toaster_run_all()

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,21 @@
# Configuration file for toaster_test
# Sorted by different host type
# test browser could be: firefox; chrome; ie(still under development)
# logging_level could be: CRITICAL; ERROR; WARNING; INFO; DEBUG; NOTSET
[toaster_test_linux]
toaster_url = 'http://127.0.0.1:8000'
test_browser = 'firefox'
test_cases = [946]
logging_level = 'INFO'
[toaster_test_windows]
toaster_url = 'http://127.0.0.1:8000'
test_browser = ['ie', 'firefox', 'chrome']
test_cases = [901, 902, 903]
logging_level = 'DEBUG'

View File

@@ -0,0 +1,44 @@
from __future__ import print_function
import sys
import httplib2
import time
import config
import urllist
# TODO: spawn server here
BASEURL="http://localhost:8000/"
#def print_browserlog(url):
# driver = webdriver.Firefox()
# driver.get(url)
# body = driver.find_element_by_tag_name("body")
# body.send_keys(Keys.CONTROL + 't')
# for i in driver.get_log('browser'):
# print(i)
# driver.close()
# TODO: turn to a test
def validate_html(url):
h = httplib2.Http(".cache")
# TODO: the w3c-validator must be a configurable setting
urlrequest = "http://icarus.local/w3c-validator/check?doctype=HTML5&uri="+url
try:
resp, content = h.request(urlrequest, "HEAD")
if resp['x-w3c-validator-status'] == "Abort":
config.logger.error("FAILed call %s" % url)
else:
config.logger.error("url %s is %s\terrors %s warnings %s (check at %s)" % (url, resp['x-w3c-validator-status'], resp['x-w3c-validator-errors'], resp['x-w3c-validator-warnings'], urlrequest))
except Exception as e:
config.logger.warn("Failed validation call: %s" % e.__str__())
print("done %s" % url)
if __name__ == "__main__":
if len(sys.argv) > 1:
validate_html(sys.argv[1])
else:
for url in urllist.URLS:
validate_html(BASEURL+url)

View File

@@ -0,0 +1,53 @@
import config
URLS = [
'toastergui/landing/',
'toastergui/builds/',
'toastergui/build/1',
'toastergui/build/1/tasks/',
'toastergui/build/1/tasks/1/',
'toastergui/build/1/task/1',
'toastergui/build/1/recipes/',
'toastergui/build/1/recipe/1/active_tab/1',
'toastergui/build/1/recipe/1',
'toastergui/build/1/recipe_packages/1',
'toastergui/build/1/packages/',
'toastergui/build/1/package/1',
'toastergui/build/1/package_built_dependencies/1',
'toastergui/build/1/package_included_detail/1/1',
'toastergui/build/1/package_included_dependencies/1/1',
'toastergui/build/1/package_included_reverse_dependencies/1/1',
'toastergui/build/1/target/1',
'toastergui/build/1/target/1/targetpkg',
'toastergui/dentries/build/1/target/1',
'toastergui/build/1/target/1/dirinfo',
'toastergui/build/1/target/1/dirinfo_filepath/_/bin/bash',
'toastergui/build/1/configuration',
'toastergui/build/1/configvars',
'toastergui/build/1/buildtime',
'toastergui/build/1/cpuusage',
'toastergui/build/1/diskio',
'toastergui/build/1/target/1/packagefile/1',
'toastergui/newproject/',
'toastergui/projects/',
'toastergui/project/',
'toastergui/project/1',
'toastergui/project/1/configuration',
'toastergui/project/1/builds/',
'toastergui/project/1/layers/',
'toastergui/project/1/layer/1',
'toastergui/project/1/layer/',
'toastergui/project/1/importlayer',
'toastergui/project/1/targets/',
'toastergui/project/1/machines/',
'toastergui/xhr_build/',
'toastergui/xhr_projectbuild/1/',
'toastergui/xhr_projectinfo/',
'toastergui/xhr_projectedit/1',
'toastergui/xhr_configvaredit/1',
'toastergui/xhr_datatypeahead/1',
'toastergui/xhr_importlayer/',
'toastergui/xhr_updatelayer/',
'toastergui/project/1/buildrequest/1',
'toastergui/',
]