Skip to content
Commits on Oct 26, 2010
  1. @josevalim

    Bump to 0.4.2.

    josevalim committed Oct 26, 2010
  2. @josevalim @josevalim
  3. @josevalim

    Update lockfile.

    josevalim committed Oct 26, 2010
Commits on Oct 16, 2010
  1. bundlerize test setup for ci

    committed Oct 16, 2010
  2. make sure the cache backend does not try to store proc objects (resul…

    …ting in marshalling errors since rails 3.x)
    committed Oct 16, 2010
Commits on Oct 10, 2010
  1. @lawrencepit @josevalim

    expose I18n::VERSION

    lawrencepit committed with josevalim Aug 4, 2010
  2. @josevalim

    change wiki link to new site

    Steven Jenkins committed with josevalim Sep 15, 2010
  3. @icco @josevalim

    1.3.5, because you don't need 1.3.6

    icco committed with josevalim Sep 23, 2010
Commits on Sep 29, 2010
  1. @polyras @josevalim

    In Ruby < 1.9 the following is true: { :foo => 1, :bar => 2 }.hash ==…

    … { :foo => 2, :bar => 1 }.hash
    
    Therefore we must use the hash of the inspect string instead to avoid cache key colisions for Ruby version older than 1.9.
    polyras committed with josevalim Sep 29, 2010
Commits on Sep 21, 2010
  1. @davidw @josevalim

    Added test and code to throw an error if a translation of an empty st…

    …ring is attempted.
    davidw committed with josevalim Sep 21, 2010
Commits on Aug 29, 2010
  1. @tomash @josevalim

    README updated with note about issue#20 occuring when I18n gem is use…

    …d with Rails < 2.3.6 and thus recommending Rails 2.3.6 or higher.
    tomash committed with josevalim Aug 29, 2010
Commits on Aug 3, 2010
  1. @lawrencepit @josevalim

    speedup cache_key

    lawrencepit committed with josevalim Aug 3, 2010
Commits on Jul 26, 2010
  1. @josevalim

    Make tests pass on both 1.8.7 and 1.9.2. Raise an error if locale fil…

    …es do not return a hash.
    josevalim committed Jul 26, 2010
Commits on Jul 22, 2010
  1. @spastorino @josevalim
Commits on Jul 13, 2010
  1. @tenderlove @josevalim
Commits on Jul 11, 2010
  1. @radar

    Remove ugly caller trace.

    radar committed with Jul 10, 2010
Commits on Jun 23, 2010
  1. @josevalim
Commits on Jun 19, 2010
  1. @knapo

    Fix #44 git ls-files returned empty list in i18n.gemspec. Skip listin…

    …g app dir, as it no longer exists
    knapo committed Jun 19, 2010
Commits on Jun 8, 2010
  1. @josevalim
  2. @lawrencepit @josevalim

    More tests for I18n.available_locales

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 8, 2010
  3. @josevalim
  4. @lawrencepit @josevalim

    Fix memoization of available_locales

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 7, 2010
  5. @josevalim

    Tidy up previous commits.

    josevalim committed Jun 8, 2010
  6. @lawrencepit @josevalim

    Added ability to run tests using mysql (use -a mysql).

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 6, 2010
  7. @lawrencepit @josevalim

    No need to test for AR version: use methods returning a scope instead…

    … of using +named_scope+ and +scope+.
    
    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 6, 2010
  8. @lawrencepit @josevalim

    Should be able to lookup native +false+ and +true+ values.

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 6, 2010
  9. @josevalim
  10. @josevalim

    Chain backend will now return {} if {} supplied as default

    Signed-off-by: José Valim <jose.valim@gmail.com>
    Matt committed with josevalim Jun 8, 2010
  11. @josevalim
  12. @lawrencepit @josevalim

    Fix bulk lookups in Chain backend.

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 7, 2010
  13. @josevalim
  14. @lawrencepit @josevalim

    Missing Backend flattening keys properly.

    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 7, 2010
  15. @josevalim
  16. @lawrencepit @josevalim

    The default option takes precedence over fallback locales only when i…

    …t's a Symbol. When it's a String or a Proc it is evaluated last after all the fallback locales have been tried.
    
    Signed-off-by: José Valim <jose.valim@gmail.com>
    lawrencepit committed with josevalim Jun 7, 2010
  17. @lawrencepit @josevalim

    Fallback backend: Returns the :de translation for a missing :'de-DE' …

    …when defaults is a Symbol (which exists in :en).
    
    Added fallbacks_test for clarity: non-String defaults take precedence before falling back.
    lawrencepit committed with josevalim Jun 5, 2010
Something went wrong with that request. Please try again.