ansible/test/runner/lib/cover.py

281 lines
9.1 KiB
Python
Raw Normal View History

"""Code coverage utilities."""
from __future__ import absolute_import, print_function
import os
import re
from lib.target import (
walk_module_targets,
walk_compile_targets,
)
from lib.util import (
display,
ApplicationError,
run_command,
common_environment,
)
from lib.config import (
CoverageConfig,
CoverageReportConfig,
)
from lib.executor import (
Delegate,
install_command_requirements,
)
COVERAGE_DIR = 'test/results/coverage'
COVERAGE_FILE = os.path.join(COVERAGE_DIR, 'coverage')
COVERAGE_GROUPS = ('command', 'target', 'environment', 'version')
def command_coverage_combine(args):
"""Patch paths in coverage files and merge into a single file.
:type args: CoverageConfig
:rtype: list[str]
"""
coverage = initialize_coverage(args)
modules = dict((t.module, t.path) for t in list(walk_module_targets()) if t.path.endswith('.py'))
coverage_files = [os.path.join(COVERAGE_DIR, f) for f in os.listdir(COVERAGE_DIR) if '=coverage.' in f]
ansible_path = os.path.abspath('lib/ansible/') + '/'
root_path = os.getcwd() + '/'
counter = 0
groups = {}
if args.all or args.stub:
sources = sorted(os.path.abspath(target.path) for target in walk_compile_targets())
else:
sources = []
if args.stub:
groups['=stub'] = dict((source, set()) for source in sources)
for coverage_file in coverage_files:
counter += 1
display.info('[%4d/%4d] %s' % (counter, len(coverage_files), coverage_file), verbosity=2)
original = coverage.CoverageData()
group = get_coverage_group(args, coverage_file)
if group is None:
display.warning('Unexpected name for coverage file: %s' % coverage_file)
continue
if os.path.getsize(coverage_file) == 0:
display.warning('Empty coverage file: %s' % coverage_file)
continue
try:
original.read_file(coverage_file)
except Exception as ex: # pylint: disable=locally-disabled, broad-except
display.error(u'%s' % ex)
continue
for filename in original.measured_files():
arcs = set(original.arcs(filename) or [])
if not arcs:
# This is most likely due to using an unsupported version of coverage.
display.warning('No arcs found for "%s" in coverage file: %s' % (filename, coverage_file))
continue
if '/ansible_modlib.zip/ansible/' in filename:
AnsiballZ improvements Now that we don't need to worry about python-2.4 and 2.5, we can make some improvements to the way AnsiballZ handles modules. * Change AnsiballZ wrapper to use import to invoke the module We need the module to think of itself as a script because it could be coded as: main() or as: if __name__ == '__main__': main() Or even as: if __name__ == '__main__': random_function_name() A script will invoke all of those. Prior to this change, we invoked a second Python interpreter on the module so that it really was a script. However, this means that we have to run python twice (once for the AnsiballZ wrapper and once for the module). This change makes the module think that it is a script (because __name__ in the module == '__main__') but it's actually being invoked by us importing the module code. There's three ways we've come up to do this. * The most elegant is to use zipimporter and tell the import mechanism that the module being loaded is __main__: * https://github.com/abadger/ansible/blob/5959f11c9ddb7b6eaa9c3214560bd85e631d4055/lib/ansible/executor/module_common.py#L175 * zipimporter is nice because we do not have to extract the module from the zip file and save it to the disk when we do that. The import machinery does it all for us. * The drawback is that modules do not have a __file__ which points to a real file when they do this. Modules could be using __file__ to for a variety of reasons, most of those probably have replacements (the most common one is to find a writable directory for temporary files. AnsibleModule.tmpdir should be used instead) We can monkeypatch __file__ in fom AnsibleModule initialization but that's kind of gross. There's no way I can see to do this from the wrapper. * Next, there's imp.load_module(): * https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151 * imp has the nice property of allowing us to set __name__ to __main__ without changing the name of the file itself * We also don't have to do anything special to set __file__ for backwards compatibility (although the reason for that is the drawback): * Its drawback is that it requires the file to exist on disk so we have to explicitly extract it from the zipfile and save it to a temporary file * The last choice is to use exec to execute the module: * https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175 * The code we would have to maintain for this looks pretty clean. In the wrapper we create a ModuleType, set __file__ on it, read the module's contents in from the zip file and then exec it. * Drawbacks: We still have to explicitly extract the file's contents from the zip archive instead of letting python's import mechanism handle it. * Exec also has hidden performance issues and breaks certain assumptions that modules could be making about their own code: http://lucumr.pocoo.org/2011/2/1/exec-in-python/ Our plan is to use imp.load_module() for now, deprecate the use of __file__ in modules, and switch to zipimport once the deprecation period for __file__ is over (without monkeypatching a fake __file__ in via AnsibleModule). * Rename the name of the AnsiBallZ wrapped module This makes it obvious that the wrapped module isn't the module file that we distribute. It's part of trying to mitigate the fact that the module is now named __main)).py in tracebacks. * Shield all wrapper symbols inside of a function With the new import code, all symbols in the wrapper become visible in the module. To mitigate the chance of collisions, move most symbols into a toplevel function. The only symbols left in the global namespace are now _ANSIBALLZ_WRAPPER and _ansiballz_main. revised porting guide entry Integrate code coverage collection into AnsiballZ. ci_coverage ci_complete
2018-06-20 18:23:59 +00:00
# Rewrite the module_utils path from the remote host to match the controller. Ansible 2.6 and earlier.
new_name = re.sub('^.*/ansible_modlib.zip/ansible/', ansible_path, filename)
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
AnsiballZ improvements Now that we don't need to worry about python-2.4 and 2.5, we can make some improvements to the way AnsiballZ handles modules. * Change AnsiballZ wrapper to use import to invoke the module We need the module to think of itself as a script because it could be coded as: main() or as: if __name__ == '__main__': main() Or even as: if __name__ == '__main__': random_function_name() A script will invoke all of those. Prior to this change, we invoked a second Python interpreter on the module so that it really was a script. However, this means that we have to run python twice (once for the AnsiballZ wrapper and once for the module). This change makes the module think that it is a script (because __name__ in the module == '__main__') but it's actually being invoked by us importing the module code. There's three ways we've come up to do this. * The most elegant is to use zipimporter and tell the import mechanism that the module being loaded is __main__: * https://github.com/abadger/ansible/blob/5959f11c9ddb7b6eaa9c3214560bd85e631d4055/lib/ansible/executor/module_common.py#L175 * zipimporter is nice because we do not have to extract the module from the zip file and save it to the disk when we do that. The import machinery does it all for us. * The drawback is that modules do not have a __file__ which points to a real file when they do this. Modules could be using __file__ to for a variety of reasons, most of those probably have replacements (the most common one is to find a writable directory for temporary files. AnsibleModule.tmpdir should be used instead) We can monkeypatch __file__ in fom AnsibleModule initialization but that's kind of gross. There's no way I can see to do this from the wrapper. * Next, there's imp.load_module(): * https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151 * imp has the nice property of allowing us to set __name__ to __main__ without changing the name of the file itself * We also don't have to do anything special to set __file__ for backwards compatibility (although the reason for that is the drawback): * Its drawback is that it requires the file to exist on disk so we have to explicitly extract it from the zipfile and save it to a temporary file * The last choice is to use exec to execute the module: * https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175 * The code we would have to maintain for this looks pretty clean. In the wrapper we create a ModuleType, set __file__ on it, read the module's contents in from the zip file and then exec it. * Drawbacks: We still have to explicitly extract the file's contents from the zip archive instead of letting python's import mechanism handle it. * Exec also has hidden performance issues and breaks certain assumptions that modules could be making about their own code: http://lucumr.pocoo.org/2011/2/1/exec-in-python/ Our plan is to use imp.load_module() for now, deprecate the use of __file__ in modules, and switch to zipimport once the deprecation period for __file__ is over (without monkeypatching a fake __file__ in via AnsibleModule). * Rename the name of the AnsiBallZ wrapped module This makes it obvious that the wrapped module isn't the module file that we distribute. It's part of trying to mitigate the fact that the module is now named __main)).py in tracebacks. * Shield all wrapper symbols inside of a function With the new import code, all symbols in the wrapper become visible in the module. To mitigate the chance of collisions, move most symbols into a toplevel function. The only symbols left in the global namespace are now _ANSIBALLZ_WRAPPER and _ansiballz_main. revised porting guide entry Integrate code coverage collection into AnsiballZ. ci_coverage ci_complete
2018-06-20 18:23:59 +00:00
elif re.search(r'/ansible_[^/]+_payload\.zip/ansible/', filename):
# Rewrite the module_utils path from the remote host to match the controller. Ansible 2.7 and later.
new_name = re.sub(r'^.*/ansible_[^/]+_payload\.zip/ansible/', ansible_path, filename)
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
elif '/ansible_module_' in filename:
AnsiballZ improvements Now that we don't need to worry about python-2.4 and 2.5, we can make some improvements to the way AnsiballZ handles modules. * Change AnsiballZ wrapper to use import to invoke the module We need the module to think of itself as a script because it could be coded as: main() or as: if __name__ == '__main__': main() Or even as: if __name__ == '__main__': random_function_name() A script will invoke all of those. Prior to this change, we invoked a second Python interpreter on the module so that it really was a script. However, this means that we have to run python twice (once for the AnsiballZ wrapper and once for the module). This change makes the module think that it is a script (because __name__ in the module == '__main__') but it's actually being invoked by us importing the module code. There's three ways we've come up to do this. * The most elegant is to use zipimporter and tell the import mechanism that the module being loaded is __main__: * https://github.com/abadger/ansible/blob/5959f11c9ddb7b6eaa9c3214560bd85e631d4055/lib/ansible/executor/module_common.py#L175 * zipimporter is nice because we do not have to extract the module from the zip file and save it to the disk when we do that. The import machinery does it all for us. * The drawback is that modules do not have a __file__ which points to a real file when they do this. Modules could be using __file__ to for a variety of reasons, most of those probably have replacements (the most common one is to find a writable directory for temporary files. AnsibleModule.tmpdir should be used instead) We can monkeypatch __file__ in fom AnsibleModule initialization but that's kind of gross. There's no way I can see to do this from the wrapper. * Next, there's imp.load_module(): * https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151 * imp has the nice property of allowing us to set __name__ to __main__ without changing the name of the file itself * We also don't have to do anything special to set __file__ for backwards compatibility (although the reason for that is the drawback): * Its drawback is that it requires the file to exist on disk so we have to explicitly extract it from the zipfile and save it to a temporary file * The last choice is to use exec to execute the module: * https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175 * The code we would have to maintain for this looks pretty clean. In the wrapper we create a ModuleType, set __file__ on it, read the module's contents in from the zip file and then exec it. * Drawbacks: We still have to explicitly extract the file's contents from the zip archive instead of letting python's import mechanism handle it. * Exec also has hidden performance issues and breaks certain assumptions that modules could be making about their own code: http://lucumr.pocoo.org/2011/2/1/exec-in-python/ Our plan is to use imp.load_module() for now, deprecate the use of __file__ in modules, and switch to zipimport once the deprecation period for __file__ is over (without monkeypatching a fake __file__ in via AnsibleModule). * Rename the name of the AnsiBallZ wrapped module This makes it obvious that the wrapped module isn't the module file that we distribute. It's part of trying to mitigate the fact that the module is now named __main)).py in tracebacks. * Shield all wrapper symbols inside of a function With the new import code, all symbols in the wrapper become visible in the module. To mitigate the chance of collisions, move most symbols into a toplevel function. The only symbols left in the global namespace are now _ANSIBALLZ_WRAPPER and _ansiballz_main. revised porting guide entry Integrate code coverage collection into AnsiballZ. ci_coverage ci_complete
2018-06-20 18:23:59 +00:00
# Rewrite the module path from the remote host to match the controller. Ansible 2.6 and earlier.
module_name = re.sub('^.*/ansible_module_(?P<module>.*).py$', '\\g<module>', filename)
if module_name not in modules:
display.warning('Skipping coverage of unknown module: %s' % module_name)
continue
new_name = os.path.abspath(modules[module_name])
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
AnsiballZ improvements Now that we don't need to worry about python-2.4 and 2.5, we can make some improvements to the way AnsiballZ handles modules. * Change AnsiballZ wrapper to use import to invoke the module We need the module to think of itself as a script because it could be coded as: main() or as: if __name__ == '__main__': main() Or even as: if __name__ == '__main__': random_function_name() A script will invoke all of those. Prior to this change, we invoked a second Python interpreter on the module so that it really was a script. However, this means that we have to run python twice (once for the AnsiballZ wrapper and once for the module). This change makes the module think that it is a script (because __name__ in the module == '__main__') but it's actually being invoked by us importing the module code. There's three ways we've come up to do this. * The most elegant is to use zipimporter and tell the import mechanism that the module being loaded is __main__: * https://github.com/abadger/ansible/blob/5959f11c9ddb7b6eaa9c3214560bd85e631d4055/lib/ansible/executor/module_common.py#L175 * zipimporter is nice because we do not have to extract the module from the zip file and save it to the disk when we do that. The import machinery does it all for us. * The drawback is that modules do not have a __file__ which points to a real file when they do this. Modules could be using __file__ to for a variety of reasons, most of those probably have replacements (the most common one is to find a writable directory for temporary files. AnsibleModule.tmpdir should be used instead) We can monkeypatch __file__ in fom AnsibleModule initialization but that's kind of gross. There's no way I can see to do this from the wrapper. * Next, there's imp.load_module(): * https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151 * imp has the nice property of allowing us to set __name__ to __main__ without changing the name of the file itself * We also don't have to do anything special to set __file__ for backwards compatibility (although the reason for that is the drawback): * Its drawback is that it requires the file to exist on disk so we have to explicitly extract it from the zipfile and save it to a temporary file * The last choice is to use exec to execute the module: * https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175 * The code we would have to maintain for this looks pretty clean. In the wrapper we create a ModuleType, set __file__ on it, read the module's contents in from the zip file and then exec it. * Drawbacks: We still have to explicitly extract the file's contents from the zip archive instead of letting python's import mechanism handle it. * Exec also has hidden performance issues and breaks certain assumptions that modules could be making about their own code: http://lucumr.pocoo.org/2011/2/1/exec-in-python/ Our plan is to use imp.load_module() for now, deprecate the use of __file__ in modules, and switch to zipimport once the deprecation period for __file__ is over (without monkeypatching a fake __file__ in via AnsibleModule). * Rename the name of the AnsiBallZ wrapped module This makes it obvious that the wrapped module isn't the module file that we distribute. It's part of trying to mitigate the fact that the module is now named __main)).py in tracebacks. * Shield all wrapper symbols inside of a function With the new import code, all symbols in the wrapper become visible in the module. To mitigate the chance of collisions, move most symbols into a toplevel function. The only symbols left in the global namespace are now _ANSIBALLZ_WRAPPER and _ansiballz_main. revised porting guide entry Integrate code coverage collection into AnsiballZ. ci_coverage ci_complete
2018-06-20 18:23:59 +00:00
elif re.search(r'/ansible_[^/]+_payload(_[^/]+|\.zip)/__main__\.py$', filename):
# Rewrite the module path from the remote host to match the controller. Ansible 2.7 and later.
# AnsiballZ versions using zipimporter will match the `.zip` portion of the regex.
# AnsiballZ versions not using zipimporter will match the `_[^/]+` portion of the regex.
module_name = re.sub(r'^.*/ansible_(?P<module>[^/]+)_payload(_[^/]+|\.zip)/__main__\.py$', '\\g<module>', filename).rstrip('_')
if module_name not in modules:
display.warning('Skipping coverage of unknown module: %s' % module_name)
continue
new_name = os.path.abspath(modules[module_name])
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
elif re.search('^(/.*?)?/root/ansible/', filename):
AnsiballZ improvements Now that we don't need to worry about python-2.4 and 2.5, we can make some improvements to the way AnsiballZ handles modules. * Change AnsiballZ wrapper to use import to invoke the module We need the module to think of itself as a script because it could be coded as: main() or as: if __name__ == '__main__': main() Or even as: if __name__ == '__main__': random_function_name() A script will invoke all of those. Prior to this change, we invoked a second Python interpreter on the module so that it really was a script. However, this means that we have to run python twice (once for the AnsiballZ wrapper and once for the module). This change makes the module think that it is a script (because __name__ in the module == '__main__') but it's actually being invoked by us importing the module code. There's three ways we've come up to do this. * The most elegant is to use zipimporter and tell the import mechanism that the module being loaded is __main__: * https://github.com/abadger/ansible/blob/5959f11c9ddb7b6eaa9c3214560bd85e631d4055/lib/ansible/executor/module_common.py#L175 * zipimporter is nice because we do not have to extract the module from the zip file and save it to the disk when we do that. The import machinery does it all for us. * The drawback is that modules do not have a __file__ which points to a real file when they do this. Modules could be using __file__ to for a variety of reasons, most of those probably have replacements (the most common one is to find a writable directory for temporary files. AnsibleModule.tmpdir should be used instead) We can monkeypatch __file__ in fom AnsibleModule initialization but that's kind of gross. There's no way I can see to do this from the wrapper. * Next, there's imp.load_module(): * https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151 * imp has the nice property of allowing us to set __name__ to __main__ without changing the name of the file itself * We also don't have to do anything special to set __file__ for backwards compatibility (although the reason for that is the drawback): * Its drawback is that it requires the file to exist on disk so we have to explicitly extract it from the zipfile and save it to a temporary file * The last choice is to use exec to execute the module: * https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175 * The code we would have to maintain for this looks pretty clean. In the wrapper we create a ModuleType, set __file__ on it, read the module's contents in from the zip file and then exec it. * Drawbacks: We still have to explicitly extract the file's contents from the zip archive instead of letting python's import mechanism handle it. * Exec also has hidden performance issues and breaks certain assumptions that modules could be making about their own code: http://lucumr.pocoo.org/2011/2/1/exec-in-python/ Our plan is to use imp.load_module() for now, deprecate the use of __file__ in modules, and switch to zipimport once the deprecation period for __file__ is over (without monkeypatching a fake __file__ in via AnsibleModule). * Rename the name of the AnsiBallZ wrapped module This makes it obvious that the wrapped module isn't the module file that we distribute. It's part of trying to mitigate the fact that the module is now named __main)).py in tracebacks. * Shield all wrapper symbols inside of a function With the new import code, all symbols in the wrapper become visible in the module. To mitigate the chance of collisions, move most symbols into a toplevel function. The only symbols left in the global namespace are now _ANSIBALLZ_WRAPPER and _ansiballz_main. revised porting guide entry Integrate code coverage collection into AnsiballZ. ci_coverage ci_complete
2018-06-20 18:23:59 +00:00
# Rewrite the path of code running on a remote host or in a docker container as root.
new_name = re.sub('^(/.*?)?/root/ansible/', root_path, filename)
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
elif '/.ansible/test/tmp/' in filename:
# Rewrite the path of code running from an integration test temporary directory.
new_name = re.sub(r'^.*/\.ansible/test/tmp/[^/]+/', root_path, filename)
display.info('%s -> %s' % (filename, new_name), verbosity=3)
filename = new_name
if group not in groups:
groups[group] = {}
arc_data = groups[group]
if filename not in arc_data:
arc_data[filename] = set()
arc_data[filename].update(arcs)
output_files = []
for group in sorted(groups):
arc_data = groups[group]
updated = coverage.CoverageData()
for filename in arc_data:
if not os.path.isfile(filename):
display.warning('Invalid coverage path: %s' % filename)
continue
updated.add_arcs({filename: list(arc_data[filename])})
if args.all:
updated.add_arcs(dict((source, []) for source in sources))
if not args.explain:
output_file = COVERAGE_FILE + group
updated.write_file(output_file)
output_files.append(output_file)
return sorted(output_files)
def command_coverage_report(args):
"""
:type args: CoverageReportConfig
"""
output_files = command_coverage_combine(args)
for output_file in output_files:
if args.group_by or args.stub:
display.info('>>> Coverage Group: %s' % ' '.join(os.path.basename(output_file).split('=')[1:]))
options = []
if args.show_missing:
options.append('--show-missing')
if args.include:
options.extend(['--include', args.include])
if args.omit:
options.extend(['--omit', args.omit])
env = common_environment()
env.update(dict(COVERAGE_FILE=output_file))
run_command(args, env=env, cmd=['coverage', 'report'] + options)
def command_coverage_html(args):
"""
:type args: CoverageConfig
"""
output_files = command_coverage_combine(args)
for output_file in output_files:
dir_name = 'test/results/reports/%s' % os.path.basename(output_file)
env = common_environment()
env.update(dict(COVERAGE_FILE=output_file))
run_command(args, env=env, cmd=['coverage', 'html', '-i', '-d', dir_name])
def command_coverage_xml(args):
"""
:type args: CoverageConfig
"""
output_files = command_coverage_combine(args)
for output_file in output_files:
xml_name = 'test/results/reports/%s.xml' % os.path.basename(output_file)
env = common_environment()
env.update(dict(COVERAGE_FILE=output_file))
run_command(args, env=env, cmd=['coverage', 'xml', '-i', '-o', xml_name])
def command_coverage_erase(args):
"""
:type args: CoverageConfig
"""
initialize_coverage(args)
for name in os.listdir(COVERAGE_DIR):
if not name.startswith('coverage') and '=coverage.' not in name:
continue
path = os.path.join(COVERAGE_DIR, name)
if not args.explain:
os.remove(path)
def initialize_coverage(args):
"""
:type args: CoverageConfig
:rtype: coverage
"""
if args.delegate:
raise Delegate()
if args.requirements:
install_command_requirements(args)
try:
import coverage
except ImportError:
coverage = None
if not coverage:
raise ApplicationError('You must install the "coverage" python module to use this command.')
return coverage
def get_coverage_group(args, coverage_file):
"""
:type args: CoverageConfig
:type coverage_file: str
:rtype: str
"""
parts = os.path.basename(coverage_file).split('=', 4)
if len(parts) != 5 or not parts[4].startswith('coverage.'):
return None
names = dict(
command=parts[0],
target=parts[1],
environment=parts[2],
version=parts[3],
)
group = ''
for part in COVERAGE_GROUPS:
if part in args.group_by:
group += '=%s' % names[part]
return group