aboutsummaryrefslogtreecommitdiffstats
path: root/mediagoblin
Commit message (Collapse)AuthorAgeFilesLines
...
* | | Switch "sqlite_refcheck" keyword arg to "migrations" which Elrond thinks is ↵Christopher Allan Webber2013-04-262-4/+16
| | | | | | | | | | | | | | | | | | cleaner Also, if migrations is true, *explicitly* say that foreign key checking is off
* | | Don't turn on sqlite refcheck stuff during migrationsChristopher Allan Webber2013-04-262-3/+3
| | |
* | | Merge remote-tracking branch 'refs/remotes/rodney757/issue643' into mergetestChristopher Allan Webber2013-04-262-2/+42
|\ \ \
| * | | Check for duplicate collection slugs and make them unique. Add unique ↵Rodney Ewing2013-04-242-2/+42
| | | | | | | | | | | | | | | | constraint to collection.slug model
* | | | Removed an unused importRodney Ewing2013-04-261-2/+0
| | | |
* | | | Unit test for ReallyLazyProxy.Elrond2013-04-251-0/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | LazyProxy caches its value. So create a unit test for lazy_pass_to_ugettext to make sure its returned proxy actually gives a different translation when the locale is switched.
* | | | No lazy_* needed here; Add ReallyLazyProxy.__repr__.Elrond2013-04-252-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | One should use lazy_* only if you have a good reason. This one found by our unit tests! For example add_message adds the message to the session, the session needs to be serialized (with json) and well, LazyProxy is not serializable. To aid in debugging, gave our ReallyLazyProxy a __repr__.
* | | | Adding ReallyLazyProxy, a proxy that does what we expect :)Christopher Allan Webber2013-04-241-4/+13
| | | |
* | | | Switching non-forms back to using normal pass_to_ugettextChristopher Allan Webber2013-04-245-5/+5
| | | |
* | | | Fix-bug-667-Use-lazy_pass_to_ugettext-for-forms.Aditi Mittal2013-04-2412-15/+15
| | | |
* | | | Create set_thread_locale and use it.Elrond2013-04-242-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | Setting up a new template env should not directly touch the translation tools, but go through a function in translations.py.
* | | | resize_image: Change calling style a bit.Elrond2013-04-241-5/+4
|/ / / | | | | | | | | | | | | | | | - entry: not used, removed. - proc_state: going to be used, added. - size_limits: not used, removed.
* | | mediagoblin/submit/lib.py: fix typoAlon Levy2013-04-231-1/+1
| | | | | | | | | | | | Signed-off-by: Alon Levy <alon@pobox.com>
* | | Create add_media_to_collection()Elrond2013-04-232-12/+23
| | | | | | | | | | | | | | | | | | | | | | | | The ideas is by Alon Levy. Not only media_collect, but also other places might want to add media to a collection. So refactor this into a function for easier usage.
* | | Doc string for ProcessingState.Elrond2013-04-221-0/+8
| | |
* | | resize_image: Refactor config loading a small bit.Elrond2013-04-211-7/+4
| | | | | | | | | | | | Well, get the config into a local variable, for easier access.
* | | Cleanup storage after test, and test .delete_dir().Elrond2013-04-212-7/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The storage tests work in the system's tmpdir. The python docs say, we should clean up after using things. Yes the directory should be cleaned up on reboot, but if running tests a lot, the tmpdir could fill up, so we should really cleanup. So use the new .delete_dir() on the storage interface to cleanup test dirs and get them finally removed with os.rmdir. All nicely packed into cleanup_storage().
* | | Merge remote-tracking branch 'remotes/npigeon/ticket-601'Christopher Allan Webber2013-04-201-1/+1
|\ \ \
| * | | Changed the max character length for tags on media to 255. Only had to ↵nattily pigeonfowl2013-04-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | change one line in mediagoblin/config_spec.ini in response to isses.mediagoblin.org/ticket/601
* | | | bumping image quality to 90, after conversation with ElrondChristopher Allan Webber2013-04-201-1/+1
| | | |
* | | | Bumping the default image quality up to 85.Christopher Allan Webber2013-04-201-1/+1
| | | |
* | | | image quality can be set in config filedunkyp2013-04-202-3/+6
| | | | | | | | | | | | | | | | | | | | (cherry-picked from https://gitorious.org/~dunkyp/mediagoblin/dunkyps-mediagoblin/commits/image_quality)
* | | | Refactor test configs.Elrond2013-04-201-24/+13
| | | | | | | | | | | | | | | | All the hook_* tests use the same config, so refactor it.
* | | | removing old callable utilities and porting stuff over.Christopher Allan Webber2013-04-195-76/+8
| | | |
* | | | Let's not use reserved keywords :)Christopher Allan Webber2013-04-191-2/+2
| | | |
* | | | Added new tests to test hook_transform()Christopher Allan Webber2013-04-194-3/+32
| | | |
* | | | we should be returning the result, not None!Christopher Allan Webber2013-04-191-1/+1
| | | |
* | | | Adjusting tests to use hook_handle and hook_runallChristopher Allan Webber2013-04-191-16/+17
| | | |
* | | | fixing hook_handle for when result is never assigned, avoid reference ↵Christopher Allan Webber2013-04-191-0/+2
| | | | | | | | | | | | | | | | without assignent
* | | | As Elrond points out, we might as well .pop() default_handlerChristopher Allan Webber2013-04-191-1/+1
| | | |
* | | | Hooks: the Next Generation!Christopher Allan Webber2013-04-191-0/+90
| | | | | | | | | | | | | | | | I have not tested any of this but it looks right ;)
* | | | And... pyblosxom code removed again :)Christopher Allan Webber2013-04-191-114/+0
| | | | | | | | | | | | | | | | We agreed on a new direction on IRC last night...
* | | | pyblosxom run_callback hoistedChristopher Allan Webber2013-04-181-0/+114
| | | |
* | | | Moving the rest of the relevant tests to have test_app set up in "def ↵Christopher Allan Webber2013-04-182-58/+50
| | | | | | | | | | | | | | | | setup()" as well
* | | | at pytest author's advice, changing the way _setup back to -> setupChristopher Allan Webber2013-04-181-37/+15
| | | | | | | | | | | | | | | | | | | | They showed me how to use @pytest.fixture, which allowed us to pass the test into setup!
* | | | piwigo: Send NotImplemented for unknown methods.Elrond2013-04-181-3/+2
| | | | | | | | | | | | | | | | That's somewhat, what piwigo does.
* | | | Use check_file_field in pwg_images_addSimple.Elrond2013-04-181-0/+4
| | | |
* | | | Start to use six for basestring.Elrond2013-04-182-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | six allows us to smoothly get more forward compatible with py3. The idea is to change things over to use six, when/if we feel a need for it.
* | | | stl:blender_render() - Remove useless arg_string.Elrond2013-04-171-2/+0
| | | |
* | | | Fixing typo in calling render_http_exceptionChristopher Allan Webber2013-04-171-1/+1
| | | |
* | | | Turning image's media manager into a new style class.Elrond2013-04-172-12/+12
| | | | | | | | | | | | | | | | | | | | images are the first media type to use the new style class based media manager.
* | | | Add sniff_handler method to mediamanagers.Elrond2013-04-171-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | The MediaManagerBase gets a fallback "return False" one. The Compat-Manager checks in the dict and calls that one or returns False.
* | | | Add MediaManagerBase class.Elrond2013-04-171-0/+16
| | | | | | | | | | | | | | | | | | | | This is the base class for all new style class based media managers.
* | | | MediaManager: Use .foo instead of ['foo'].Elrond2013-04-173-6/+7
| | | | | | | | | | | | | | | | To make .media_fetch_order work, create a property.
* | | | First step towards a MediaManager class: Compat one.Elrond2013-04-173-4/+32
| | | | | | | | | | | | | | | | | | | | | | | | To get us moving towards a MediaManager class, the first idea is to create a class that wraps our current dict based manager and makes all users happy.
* | | | PIL: Support systems with Pillow and withoutAlon Levy2013-04-177-10/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes for systems with Pillow, but leaves a "try: except ImportError" to support anything that doesn't have a PIL top level import. Signed-off-by: Alon Levy <alon@pobox.com>
* | | | pdf/processing.py: don't use check_output, only in python>=2.7Alon Levy2013-04-171-9/+14
| | | | | | | | | | | | | | | | Signed-off-by: Alon Levy <alon@pobox.com>
* | | | Refactor test resources into new resources.pyElrond2013-04-175-55/+47
| | | |
* | | | PDF: Use pytest.mark.skipif for skipping testsElrond2013-04-162-5/+4
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of leaving test early if they can not run, use the pytest.mark.skipif marked to tell the test system not to even run the test. This also adds to the stats, because skipped tests are counted differently. Thus making it obvious, that some tests did not run, because of any reason.
* | | test_timesince() doesn't need a complete test_app.Elrond2013-04-161-1/+1
| | |