1. jrhite
  2. babel-py3k

Commits

Show all
Author Commit Message Date Builds
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Commit the resulting files after a 2to3 -nvw run.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Copy over trunk.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix Serbian plural forms (ticket #213). Submitted by: Patrice Neff
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix typos.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Document the Python 2.7 ElementTree fix.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix the ElementTree find()/findall() syntax to be compatible with Python 2.7.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix test to run under Python 2.3.
cml...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Rename the ''CVS'' directory in the test data to ''ignored'', as quite a number of tools are configured or even hardcoded to ignore CVS directories. For example, I couldn't find a way to include the directory in the sdist tarball.
cml...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Workaround for a weird bug with newer versions of pytz, which fails when `utcoffset()` gets passed a `None` argument. Which it shouldn't be getting in the first place.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Add the 0.9.5 release notes.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix some more typos.
palg...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Typo correction.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Document the square bracket and CLDR 1.7 work.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Forgot to bump the copyright line.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Actually make the msgid_plural be a plural as per typical use case.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
More than two plural forms work nowadays. Adjust the test.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Put the square brackets test after all the normal plural tests. It's more logical.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Add a test for a locale with one plural form.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix testcase name typo. Switch to use a locale that uses a different singular and plural form instead of just one form for both singular and plural.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Make sure to only strip on the first occurence of ]. Submitted by: Benoit Boissinot
pje...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
remove sorted and don't assume dict ordering (Python 2.3 & Jython compat)
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Current CDLR has implemented the full type specifiers on the dateTimeFormats, as a result you cannot call None on the dictionary anymore since it's no longer a key for it.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
The date now returns 52 with the current data, adjust the test accordingly.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Revert part of the commits with relation to the de locale's use of \xa0 as thousands separator, this is still a proposed change and not the active one.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Fix the import script to skip alt or draft items in the numbers/symbols subtree of a locale.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Add extra doctest for formatting a number in de_DE since it uses a non-breaking space (nbsp).
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Adjust tests to match the data in CLDR: German doesn't use a . as thousands separator, but a non-breaking whitespace.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Mark up two docstrings as raw (r""") in order to get the escaped characters properly picked up in the doctest.
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Add a __ne__ method for the Locale class. Submitted by: mitsuhiko
jrui...@59ecc08e-a131-0410-9ea7-d4c0f28ac310
Make sure to note CLDR 1.7 in installation details.
  1. Prev
  2. Next