]>
git.ipfire.org Git - thirdparty/babel.git/log
Christopher Lenz [Mon, 14 Jul 2008 13:02:16 +0000 (13:02 +0000)]
More graceful `__version__` initialization when Babel is not installed in a way Setuptools knows about it.
Armin Ronacher [Mon, 14 Jul 2008 12:48:21 +0000 (12:48 +0000)]
Some refactoring in plural.py
Armin Ronacher [Fri, 11 Jul 2008 13:51:20 +0000 (13:51 +0000)]
Added pluralization support module for the plural definitions from the latest CLDR
Christopher Lenz [Fri, 11 Jul 2008 13:47:28 +0000 (13:47 +0000)]
Improve CLDR import of currency-related data to ignore unsupported features such as symbol choice patterns and pluralized display names. See #93.
Jeroen Ruigrok van der Werven [Fri, 11 Jul 2008 08:58:04 +0000 (08:58 +0000)]
Clarify first_week_day, weekend_start, and weekend_end to point out that
0 is Monday.
Christopher Lenz [Wed, 9 Jul 2008 16:36:54 +0000 (16:36 +0000)]
Fix dates in ChangeLog.
Christopher Lenz [Tue, 8 Jul 2008 20:58:02 +0000 (20:58 +0000)]
Use item access to catalog messages in tests, so that they can be easily ported back to the 0.9.x branch.
Christopher Lenz [Tue, 8 Jul 2008 20:50:07 +0000 (20:50 +0000)]
Use item access to catalog messages in tests, so that they can be easily ported back to the 0.9.x branch.
Christopher Lenz [Mon, 7 Jul 2008 20:55:15 +0000 (20:55 +0000)]
Fix conversion of timestamp to `datetime` in `format_datetime()`
Christopher Lenz [Mon, 7 Jul 2008 14:49:16 +0000 (14:49 +0000)]
Implement support for aliases in the CLDR data. Closes #68. Also, update to CLDR 1.6, and a much improved `dump_data` script.
Armin Ronacher [Tue, 1 Jul 2008 09:17:34 +0000 (09:17 +0000)]
Added changelog entry for new `get_plural` function.
Armin Ronacher [Tue, 1 Jul 2008 09:16:30 +0000 (09:16 +0000)]
Added babel.messages.plurals.get_plural which returns a special tuple with the plural information.
Pedro Algarvio [Mon, 30 Jun 2008 16:37:29 +0000 (16:37 +0000)]
Remove some unused imports.
Pedro Algarvio [Mon, 30 Jun 2008 16:35:50 +0000 (16:35 +0000)]
Include fix of [407] in `pybabel`.
Join the `InitCatalog` testcases into a single one.
Make pylint a bit happier on `babel/messages/tests/checkers.py`.
Pedro Algarvio [Mon, 30 Jun 2008 12:23:28 +0000 (12:23 +0000)]
We no longer neglect `catalog.plurals`. Added tests for it. Fixes #120.
Added tests for the plural forms checker which will all now pass due to the above fix.
Pedro Algarvio [Thu, 26 Jun 2008 21:16:42 +0000 (21:16 +0000)]
Make sure the fix also works for multiple translator comments. Refs #119.
Pedro Algarvio [Thu, 26 Jun 2008 21:02:20 +0000 (21:02 +0000)]
Test and respective fix for gettext calls that spawn multiple lines. Fixes #119.
Pedro Algarvio [Thu, 26 Jun 2008 16:39:21 +0000 (16:39 +0000)]
Fix typo and add a test for translator comments with leading spaces.
Armin Ronacher [Wed, 25 Jun 2008 17:56:19 +0000 (17:56 +0000)]
Fixed a typo in the message checkers (tests will follow)
Armin Ronacher [Wed, 18 Jun 2008 12:39:01 +0000 (12:39 +0000)]
Added documentation to format method in babel.js
Armin Ronacher [Wed, 18 Jun 2008 12:29:00 +0000 (12:29 +0000)]
Added contrib folder with a JavaScript support file for babel.
Armin Ronacher [Wed, 18 Jun 2008 10:42:07 +0000 (10:42 +0000)]
Fixed logic error in the python format checker.
Armin Ronacher [Tue, 17 Jun 2008 22:11:09 +0000 (22:11 +0000)]
Convert format strings to str for nicer error messages (no u prefix) in the python_format checker.
Armin Ronacher [Tue, 17 Jun 2008 21:56:50 +0000 (21:56 +0000)]
Updated catalog test case for last commit, flags are now tested too in the clone test case.
Armin Ronacher [Tue, 17 Jun 2008 21:55:14 +0000 (21:55 +0000)]
Message.clone doesn't return a shallow copy any longer. This fixes a bug with update where flags where shared.
Armin Ronacher [Tue, 17 Jun 2008 21:17:44 +0000 (21:17 +0000)]
Fix scientific notation for 0. closes #99
Armin Ronacher [Tue, 17 Jun 2008 20:40:36 +0000 (20:40 +0000)]
Fixed #59 by falling back silently on invalid location comments.
Armin Ronacher [Tue, 17 Jun 2008 20:30:02 +0000 (20:30 +0000)]
Refactored the checker system. It's now possible to partially validate translations on a per-message level.
Armin Ronacher [Tue, 17 Jun 2008 20:07:08 +0000 (20:07 +0000)]
Moved PYTHON_FORMAT back to catalog.
Armin Ronacher [Tue, 17 Jun 2008 19:59:02 +0000 (19:59 +0000)]
Fixed a small bug in the python format string checker that caused the wrong exception to be thrown.
Armin Ronacher [Tue, 17 Jun 2008 19:54:34 +0000 (19:54 +0000)]
The builtin checkers don't require setuptools any longer, validate_format and python_format from the checkers module are merged into one now.
Armin Ronacher [Tue, 17 Jun 2008 19:32:09 +0000 (19:32 +0000)]
Added `validate_format helper function to `babel.support`.
Christopher Lenz [Mon, 16 Jun 2008 16:49:56 +0000 (16:49 +0000)]
More work on msgctxt support (#54).
Christopher Lenz [Mon, 16 Jun 2008 16:05:35 +0000 (16:05 +0000)]
Fixes for timezone calculations in time formatting (#83).
Christopher Lenz [Mon, 16 Jun 2008 12:24:04 +0000 (12:24 +0000)]
Update to CLDR 1.5.1, which split out the metazone mappings into a separate supplemental file.
Christopher Lenz [Mon, 16 Jun 2008 11:37:52 +0000 (11:37 +0000)]
Fix for incorrect month context lookup in date formatting. Closes #75. Thanks to Andrew Stromnov for reporting the problem and providing a patch.
Christopher Lenz [Mon, 16 Jun 2008 10:14:38 +0000 (10:14 +0000)]
Remove leftover unused code from [365].
Armin Ronacher [Mon, 16 Jun 2008 09:53:44 +0000 (09:53 +0000)]
Fixed a bug in pofile (in_msgctxt was not defined). Test follows.
Armin Ronacher [Sat, 14 Jun 2008 22:07:41 +0000 (22:07 +0000)]
JavaScript lexer falls back silently now on syntax errors and tries to recover.
Armin Ronacher [Sat, 14 Jun 2008 19:00:35 +0000 (19:00 +0000)]
added some newlines to extract and jslexer to stay consistent with the rest of the sourcecode.
Armin Ronacher [Thu, 12 Jun 2008 16:26:52 +0000 (16:26 +0000)]
Added !JavaScript extractor
Armin Ronacher [Thu, 12 Jun 2008 16:24:25 +0000 (16:24 +0000)]
Stripping of comment tags is optional now. If enabled it will strip the tags from all lines of a comment now.
Christopher Lenz [Wed, 11 Jun 2008 22:12:09 +0000 (22:12 +0000)]
Fix iterkeys/iteritems/itervalues/pop/popitem methods on the `odict` utility class. Thanks to Armin Ronacher for the patch.
Christopher Lenz [Wed, 11 Jun 2008 20:00:21 +0000 (20:00 +0000)]
Fix handling of default value of `locales` parameter of the `Translations.load()` method. Thanks to Armin Ronacher for reporting the problem.
Christopher Lenz [Wed, 11 Jun 2008 18:56:27 +0000 (18:56 +0000)]
More preparation for msgctxt support (#54).
Christopher Lenz [Tue, 10 Jun 2008 17:05:52 +0000 (17:05 +0000)]
Add basic MO file reading in preparation for #54.
Christopher Lenz [Tue, 10 Jun 2008 16:57:15 +0000 (16:57 +0000)]
Change Catalog class to retain the plural forms set in the MIME headers.
Christopher Lenz [Tue, 10 Jun 2008 10:32:33 +0000 (10:32 +0000)]
Add test file needed for [368].
Christopher Lenz [Fri, 6 Jun 2008 22:06:33 +0000 (22:06 +0000)]
Fix message catalog compilation for locales with more than two plural forms. Closes #95. Many thanks to Victor Safronovich for the patch.
Christopher Lenz [Fri, 6 Jun 2008 21:51:48 +0000 (21:51 +0000)]
Fix for #97, compilation of message catalogs for locales with more than two plural forms where the translations were empty was failing.
Christopher Lenz [Fri, 6 Jun 2008 21:22:29 +0000 (21:22 +0000)]
Allow extraction method specification to use a dot instead of the colon for separating module and function names. See #105.
Jeroen Ruigrok van der Werven [Fri, 2 May 2008 22:13:17 +0000 (22:13 +0000)]
Reinstate changeset r362, but this time properly wrap the exception list in
braces.
Submitted by: cboos
Jeroen Ruigrok van der Werven [Thu, 24 Apr 2008 09:25:09 +0000 (09:25 +0000)]
Back out last change, this broke extraction.
Christopher Lenz [Fri, 11 Apr 2008 12:51:12 +0000 (12:51 +0000)]
Make dependency on the parser module optional.
Jeroen Ruigrok van der Werven [Wed, 19 Mar 2008 15:32:18 +0000 (15:32 +0000)]
Actually use UTC to be more correct.
Jeroen Ruigrok van der Werven [Wed, 19 Mar 2008 15:03:07 +0000 (15:03 +0000)]
Replace Europe/Paris with Universal.
The doctest was, and still is, not respecting local clock changes for
summer/winter time, which cause the test to fail for half of the year.
Jeroen Ruigrok van der Werven [Wed, 19 Mar 2008 14:58:15 +0000 (14:58 +0000)]
Fix typo.
Philip Jenvey [Mon, 18 Feb 2008 18:09:30 +0000 (18:09 +0000)]
fix invalid message extraction methods causing:
UnboundLocalError: local variable 'func' referenced before assignment
thanks cramm
Christopher Lenz [Mon, 4 Feb 2008 10:06:49 +0000 (10:06 +0000)]
Prepare ChangeLog for [milestone:0.9.2] release.
Christopher Lenz [Mon, 4 Feb 2008 09:51:38 +0000 (09:51 +0000)]
Documentation fixes/updates, closes #70.
Christopher Lenz [Fri, 1 Feb 2008 15:30:49 +0000 (15:30 +0000)]
Fix for #60.
Christopher Lenz [Fri, 1 Feb 2008 15:23:13 +0000 (15:23 +0000)]
Fix for #79 (location lines wrapping at hyphens).
Christopher Lenz [Fri, 1 Feb 2008 14:46:32 +0000 (14:46 +0000)]
Merging catalogs would sometimes mix translations from different runs.
Christopher Lenz [Fri, 1 Feb 2008 13:38:09 +0000 (13:38 +0000)]
Fix catalog updating with fuzzy matches. Closes #82.
Jeroen Ruigrok van der Werven [Sun, 20 Jan 2008 15:32:56 +0000 (15:32 +0000)]
Add a Plural-Forms for Cymraeg (Welsh).
See #71
Christopher Lenz [Mon, 14 Jan 2008 12:47:45 +0000 (12:47 +0000)]
Fix for unicode problem when the previous message id is included as a comment in PO serialization. Closes #78.
Christopher Lenz [Mon, 7 Jan 2008 16:04:38 +0000 (16:04 +0000)]
Same bug fixed in [339] was also in the setuptools `compile_catalog` command.
Christopher Lenz [Mon, 7 Jan 2008 15:10:46 +0000 (15:10 +0000)]
Fixed bug where the commandline interface to the compile catalog command actually used the output file for the input file option. Thanks to mitsuhiko for reporting this.
Jeroen Ruigrok van der Werven [Mon, 31 Dec 2007 11:04:28 +0000 (11:04 +0000)]
Assume that, since Chuvash is a Turkic language, Chuvash follows the same
plurality rules as Turkish.
For the interested: Chuvash is spoken by approximately 2 million people in
Russia and uses an augmented Cyrillic alphabet to write their Oghuric-Turkic
language.
Jeroen Ruigrok van der Werven [Mon, 31 Dec 2007 10:58:56 +0000 (10:58 +0000)]
Given Dzongkha's relation to Tibetan for now assume the same plurality rules
are in effect.
For the interested: Dzongkha is the official language of Bhutan.
Jeroen Ruigrok van der Werven [Mon, 31 Dec 2007 10:55:28 +0000 (10:55 +0000)]
Add commented ISO 639-1 codes up to 'e'.
Jeroen Ruigrok van der Werven [Sun, 30 Dec 2007 16:03:10 +0000 (16:03 +0000)]
Add a definition for Lao.
Jeroen Ruigrok van der Werven [Thu, 13 Dec 2007 15:10:12 +0000 (15:10 +0000)]
Use consistent English names for all language names.
Jeroen Ruigrok van der Werven [Thu, 13 Dec 2007 11:51:36 +0000 (11:51 +0000)]
When using sys.stdout with a pipe or redirection the sys.stdout.encoding value
will be set to None instead of what the environment first specified through
any locale setting. As such, fall back to locale.getpreferredencoding() before
the ultimate fallback to ASCII.
OK: cmlenz
closes: #76
Jeroen Ruigrok van der Werven [Wed, 12 Dec 2007 12:01:48 +0000 (12:01 +0000)]
Add plural forms definition for Tibetan.
Discussed in private with Andrew West, well-known and respect Sino-Tibetan
scholar and Unicode contributor.
Jeroen Ruigrok van der Werven [Thu, 6 Dec 2007 10:10:28 +0000 (10:10 +0000)]
Add a plurals definition for Thai.
Philip Jenvey [Fri, 28 Sep 2007 23:47:35 +0000 (23:47 +0000)]
mention the ticket, #66, in the last changelog
Philip Jenvey [Fri, 28 Sep 2007 05:18:27 +0000 (05:18 +0000)]
woops, misplaced changelog
Philip Jenvey [Fri, 28 Sep 2007 05:16:25 +0000 (05:16 +0000)]
fix catalogs' charset values not being recognized
thanks Jose Galvez
Philip Jenvey [Fri, 28 Sep 2007 04:44:55 +0000 (04:44 +0000)]
we must mean options.locale here
Christopher Lenz [Sun, 9 Sep 2007 17:07:45 +0000 (17:07 +0000)]
Fix error in `compile_catalog` distutils command. Closes #65.
Christopher Lenz [Fri, 7 Sep 2007 10:12:35 +0000 (10:12 +0000)]
Updated ChangeLog for [milestone:0.9.1].
Christopher Lenz [Wed, 5 Sep 2007 17:13:22 +0000 (17:13 +0000)]
Fix `__repr__` of `Translations` class. Closes #63.
Christopher Lenz [Wed, 5 Sep 2007 16:40:17 +0000 (16:40 +0000)]
In `Locale.parse()`, only parse the argument if it's a string, otherwise just return the argument. This assumes that a non-string argument is either a `Locale` instance, or a proxy object pretending to be one. That may not be a safe assumption, but at least it allows you to use proxy objects such as Paste's `StackedObjectProxy` for locales.
Christopher Lenz [Wed, 5 Sep 2007 16:32:51 +0000 (16:32 +0000)]
Write PO files in binary mode. Closes #61.
Christopher Lenz [Tue, 4 Sep 2007 15:09:54 +0000 (15:09 +0000)]
Fix for updating catalog messages that changed from gettext to ngettext or vice versa.
Jonas Borgström [Sun, 26 Aug 2007 13:43:06 +0000 (13:43 +0000)]
Fixed a bug introduced by [301].
Jonas Borgström [Sun, 26 Aug 2007 13:38:58 +0000 (13:38 +0000)]
Fixed formatting bug with 12-hour clock patterns.
Christopher Lenz [Wed, 22 Aug 2007 15:16:45 +0000 (15:16 +0000)]
Fix typo in docstring.
Christopher Lenz [Wed, 22 Aug 2007 11:32:09 +0000 (11:32 +0000)]
Fix typo in documentation of the default extraction methods. Thanks to Ramiro Morales for pointing out the mistake.
Christopher Lenz [Wed, 22 Aug 2007 09:33:47 +0000 (09:33 +0000)]
Default the encoding of output by the `pybabel --list-locales` command to ASCII with errors replaced, if the output stream has no encoding set. For a reason unknown to me I got an encoding error, but only when when running the output through a pipe on the shell.
Christopher Lenz [Mon, 20 Aug 2007 09:43:26 +0000 (09:43 +0000)]
Bump up version number on trunk.
Christopher Lenz [Mon, 20 Aug 2007 09:04:35 +0000 (09:04 +0000)]
Updated ChangeLog for [milestone:0.9].
Christopher Lenz [Mon, 20 Aug 2007 08:33:44 +0000 (08:33 +0000)]
Fix minor glitch in docstring.
Christopher Lenz [Mon, 20 Aug 2007 08:28:30 +0000 (08:28 +0000)]
Apply patch for #26 to implement locale aliases.
Philip Jenvey [Fri, 17 Aug 2007 05:02:44 +0000 (05:02 +0000)]
skip messages that have less arguments than the keyword spec calls for
avoids potential IndexErrors during extraction
Philip Jenvey [Wed, 15 Aug 2007 18:36:19 +0000 (18:36 +0000)]
add extractor tests for some odder method signatures and concatenated strings
Christopher Lenz [Wed, 15 Aug 2007 13:13:24 +0000 (13:13 +0000)]
Fix typo in [271] that slipped into the check-in.
Christopher Lenz [Wed, 15 Aug 2007 13:12:25 +0000 (13:12 +0000)]
Added examples for using `get_timezone_name` to documentation.
Christopher Lenz [Wed, 15 Aug 2007 12:41:40 +0000 (12:41 +0000)]
Fail more gracefully when formatting the timezone for an unknown/invalid territory.