The unit test infrastructure will remain for things that are mocked out and testable with out filesystem
side effects, and a few cases of things that might not be quite so much (like inventory) that can still
benefit from heavy access to the API.
See the 'tests_new/integration' directory, this will soon fold into tests_new.
We run into some problems because tar --diff will take into account the file ownership and fail if they don't match.
The real-world implication of this is that we could be doing more unarchives then we need to be doing.
There is a bit going on with the changes here. Most of the changes are cleanup of files so that they line up with the standard files.
PR #5136 was merged into the current devel and brought up to working order. A few bug fixes had to be done to get the code to test correctly. Thanks out to @pib!
Issue #5431 was not able to be confirmed as it behaved as expected with a sudo user.
Tests were added via a playbook with archive files to verify functionality.
All tests fire clean including custom playbooks across multiple linux and solaris systems.
Using
```
assert 'changed' in result
```
doesn't actually check if something is changed, which is presumably
the reason for the assertion. What is actually needed is
```
assert result.get('changed')
```
which checks that changed is set and not False. Tests still pass after
this change
This adds two parameters to the git module:
bare (boolean)
Indicates this is to be a bare repositori
reference (string)
Indicates the path or url to the reference repo.
Check out the "--reference" option in the "git clone"
man page
Added appropriate tests.
The validate option is constructed similarly to the template command's
validate option. TestRunner.py has been updated to include two new
tests, one for passing and one for failing validation.
Still compatible with user: but deprecating it so we can have
a matching remote_user: in tasks, cannot be user: because of the
module of the same name. #3932
Signed-off-by: Brian Coca <briancoca+dev@gmail.com>
The 'always_run' task clause allows one to execute a task even in
check mode.
While here implement Runner.noop_on_check() to check if a runner
really should execute its task, with respect to check mode option
and 'always_run' clause.
Also add the optional 'jinja2' argument to check_conditional() :
it allows to give this function a jinja2 expression without exposing
the 'jinja2_compare' implementation mechanism.
Tests `test_playbook_undefined_varsX_fail` check if ansible detects
undefined variables when `error_on_undefined_vars` is enabled. These
tests fail without "Improve behavior with error_on_undefined_vars
enabled" patch.
Tests `test_playbook_undefined_varsX_ignore` check if ansible ignores
undefined variables when `error_on_undefined_vars` is disabled.
Also modify PlayBook._run_task_internal() so error_on_undefined_vars is
testable.
ansible.constants was calling expanduser (by way of shell_expand_path)
on the entire configured value for the library and *_plugins
configuration values, but these values have always been interpreted as
multiple directories separated by os.pathsep. Thus, if you supplied
multiple directories for one of these values, typically only the first
(at least on *nix) would have e.g. "~" expanded to HOME.
Now PluginLoader does expansion on each individual path in each of
these variables.
If someone has a " #" in a quoted var string, it
will interpret that as a comment and refuse to
load the inventory file due to an unbalanced
quote. Noisy failure > unexpected behavior.
PluginLoader._get_paths, as of 391fb98e, was only finding plug-ins that
were in a subdirectory of one of the basedirs (i.e. in a category
directory). For example, action_plugins/foo.py would never be loaded,
but action_plugins/bar/foo.py would work.
This makes it so that "uncategorized" plug-ins in the top level of a
directory such as action_plugins will be loaded, though plug-ins in a
"category" subdirectory will still be preferred. For example,
action_plugins/bar/foo.py would be preferred over action_plugins/foo.py.
The copy action accepts force=no, which tells it not to replace an
existing file even if it differs from the source. The copy action
plug-in wasn't respecting this option when operated in check mode, so it
would report that changes are necessary in check mode even though copy
would make no changes when run normally.
Runner._remote_md5 was changed to make the logic for setting rc perhaps
a little more clear, and to make sure that rc=0 when the file does not
exist.
As documented in #2623, early variable substitution causes when_
tests to fail and possibly other side effects.
I can see the reason for this early substitution, likely introduced
in 1dfe60a6, to allow many playbook parameters to be templated.
This is a valid goal, but the recursive nature of the utils.template
function means that it goes too far.
At this point removing tasks from the list of parameters to be
substituted seems sufficient to make my tests pass. It may be the
case that other parameters should be excluded, but I suspect not.
Adding a test case. I would prefer to analyse not just the aggregate
statistics but also whether the results are as expected - I can't
see an easy way to do that with the available callbacks at present.
When operating on a unicode string in python 2.6, shlex.split returns
a result that does not work with the file constructor.
To reproduce this requires a task include that is templated (this is
because the templated string is a unicode result, whereas a non-
templated string is a non-unicode string)
[will@centos6.3] $ python
Python 2.6.6 (r266:84292, Sep 11 2012, 08:34:23)
[GCC 4.4.6 20120305 (Red Hat 4.4.6-4)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import shlex
>>> shlex.split(u'abc')
['a\x00\x00\x00b\x00\x00\x00c\x00\x00\x00']
[will@fedora17] $ python
Python 2.7.3 (default, Jul 24 2012, 10:05:38)
[GCC 4.7.0 20120507 (Red Hat 4.7.0-5)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import shlex
>>> shlex.split(u'abc')
['abc']
The proposed fix (coercing the include parameters to string before the
shlex.split) may not be ideal but it does fix the bug for my test case.
- added cron_file attribute: if specified, the file with appropriate
job is created in /etc/cron.d directory. Also, you can store multiple
jobs in one file. state='absent' attribute is handled in the following
way in this case: if after the deletion of the job from the file specified
by cron_file variable the file is empty, the file is deleted, otherwise
not.
- fixed the behaviour, when the backupfile is saved forever in /tmp
folder, even if the backup= atribute is not set (os.unlink() is called if
backup is not True).
- added some comments to the unobvious places
Instead of having to remember when to use which one, rename template_ds
to template and move the last bit of code from template to varReplace
(which gets used for all string replacements, in the end).
This means that you can template any data type without worrying about
whether it's a string or not, and the right thing will happen.
Pretty straightforward. Give it a URL with an exported GPG key for signing an
Apt repository. It downloads it and will install it using apt-key. It's even
smart enough to tell if it's already there (i.e. actually tells you if it
changed or not).
At the moment Ansible prefers yes/no for module booleans, however booleans in playbooks are still using True/False, rather than yes/no. This changes modifies boolean uses in playbooks (and man pages) to favor yes/no rather than True/False.
This change includes:
- Adaptation of documentation and examples to favor yes/no
- Modification to manpage output to favor yes/no (the docsite output already favors yes/no)
This allows patterns such as webservers:!debian:&datacenter1 to target
hosts in the webservers group, that are not in the debian group, but are
in the datacenter1 group. It also parses patterns left to right.
Test for when environment variable and configuration file
variable both set now tests that the environment variable takes
precedence
Removed logic that would never be triggered from
lib/ansible/constants.py
pipes.quote is a bit overzealous for what we want to do, quoting ;
and other characters that you most likely want to use in your shell
invocations. The regexp is the best I could come up with to be able
to only replace the parts of the arguments that shouldn't be
executed.
Automatic quoting of variables in only_if breaks existing playbooks
where entire statements are put in a variable, and other cases. See
issue #1120 for details.
This fixes a few issues,
- ${foo}${bar} would be parsed as a variable named foo}${bar,
which wouldn't be easily fixed without breaking ${foo.${bar}}
- allows escaping . in variable parts so e.g.
${hostvars.{test.example.com}.foo} works
This is slower than using re. 3 million templating calls take about
about twice as long to complete with this compared to the regexp,
from ~65 seconds to ~115 seconds on my laptop.
This fixes e.g. only_if: ${task.changed} which would always
evaluate to true due to it having been replaced by a string for its
boolean value. Also adds a test case to ensure it doesn't get
missed again.
When the output of a command is stored in a register, this will create a
stdout_lines field in the result object that contains stdout split into a list
of lines. This list can then be iterated over using with_items.
date still issues warning and ignores TZ
- Updated tests to work inside bsd jails (127 addresses are an issue)
Signed-off-by: Brian Coca <briancoca+ansible@gmail.com>
on_skipped callback being passed an extra item parameter that it wasn't
expecting.
Fixed it so that on_skipped in TestCallbacks accepts and ignores the
extra parameter
Extra parameter was added in 4b9b9a8a5b
but not really clear why from commit message
as it is /bin/false on many systems but /usr/bin/false on OS X
test/playbook1.yml now just does command true, rather than command /bin/true,
again so that it works on OS X
Changed from using which false to just using command false to
make it simpler and also match how playbook1.yml works
