Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | ActiveSupport::BacktraceCleaner#remove_filters! allows for completely ↵ | Doug Alcorn | 2010-04-10 | 1 | -1/+7 |
| | | | | | | untouched backtrace [#4079 state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | Update versions (otherwise you install a gem from source as beta3 but ↵ | José Valim | 2010-04-10 | 1 | -1/+1 |
| | | | | internally it's beta2) and update CHANGELOG. | ||||
* | Refactor for readability | Jeremy Kemper | 2010-04-09 | 1 | -30/+24 |
| | |||||
* | Fixed method sign error | Santiago Pastorino | 2010-04-09 | 1 | -1/+1 |
| | | | | Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | fix stack trace lines on class_eval | Santiago Pastorino | 2010-04-09 | 5 | -6/+6 |
| | | | | Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | Make tidy_bytes work on 1.9 and improve its performance. [#4350 state:resolved] | Norman Clarke | 2010-04-09 | 1 | -18/+67 |
| | | | | Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
* | fisting stack trace [#4349 state:resolved] | Aaron Patterson | 2010-04-08 | 1 | -1/+1 |
| | | | | Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
* | Make sure MemCacheStore logger exists | Joshua Peek | 2010-04-07 | 1 | -3/+3 |
| | |||||
* | Changed the way inflections for uncountables work for 'funky jeans' [#3576 ↵ | Rolf Bjaanes | 2010-04-05 | 2 | -2/+2 |
| | | | | | | state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | moves Object#singleton_class to Kernel#singleton_class to match Ruby also ↵ | Xavier Noria | 2010-04-05 | 8 | -9/+9 |
| | | | | there, same for #class_eval to simplify, and adds coverage for class_eval | ||||
* | Parameterize should accept malformed utf8 characters [#4323 state:resolved] | Kristopher Murata | 2010-04-04 | 1 | -1/+4 |
| | | | | Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
* | delegate unknown timezones to TZInfo | Santiago Pastorino | 2010-04-02 | 1 | -5/+12 |
| | |||||
* | utc_offset is no longer required on TimeZone and if it's not supplied we ↵ | Santiago Pastorino | 2010-04-02 | 1 | -61/+18 |
| | | | | delegate to TZInfo | ||||
* | Fix class_attribute so that it works with singleton classes. | Carl Lerche | 2010-04-01 | 1 | -1/+1 |
| | |||||
* | Prep for beta2, depend on latest Bundler | David Heinemeier Hansson | 2010-04-01 | 1 | -1/+1 |
| | |||||
* | adds missing require for blank? and present? | Xavier Noria | 2010-03-31 | 1 | -0/+1 |
| | |||||
* | Don't cache the utc_offset we are already caching the timezone [#4301 ↵ | Santiago Pastorino | 2010-03-30 | 1 | -2/+5 |
| | | | | | | state:committed] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
* | New assertion: assert_present [#4299 state:committed] | Juanjo Bazan | 2010-03-30 | 1 | -0/+7 |
| | | | | Signed-off-by: Xavier Noria <fxn@hashref.com> | ||||
* | new assertion: assert_blank | Juanjo Bazan | 2010-03-30 | 1 | -0/+7 |
| | | | | Signed-off-by: Xavier Noria <fxn@hashref.com> | ||||
* | Replace the placeholder base_hook API with on_load. To specify some code that | wycats | 2010-03-29 | 4 | -25/+13 |
| | | | | | | | | should run during framework load do: ActiveSupport.on_load(:action_controller) do # Code run in the context of AC::Base end | ||||
* | Revert "Hash#symbolize_keys(!) optimizations" | Jeremy Kemper | 2010-03-29 | 1 | -1/+1 |
| | | | | | | | | Was slower in common case. [#3891 state:open] This reverts commit 2060977b767061a42eb8db2d5c3a30d205a94123. | ||||
* | Provide a better error if the parsed REXML document has no root [#3803 ↵ | wycats | 2010-03-28 | 1 | -1/+7 |
| | | | | state:resolved] | ||||
* | Ruby 1.9.2: marshaling round-trips Time#zone | Jeremy Kemper | 2010-03-28 | 3 | -27/+57 |
| | |||||
* | Drop expires argument from call to @data in MemCacheStore so it works with ↵ | Vladimir Andrijevik | 2010-03-28 | 1 | -1/+1 |
| | | | | | memcache-client and memcached gems, as advertised [#3672 state:resolved] Signed-off-by: wycats <wycats@gmail.com> | ||||
* | adds missing requires for Object#blank? and Object#present? | Xavier Noria | 2010-03-28 | 2 | -0/+2 |
| | |||||
* | Actually check the value of the ivar | wycats | 2010-03-27 | 1 | -4/+2 |
| | |||||
* | LINE on class_eval need + 1, also removed comments [#4281 state:committed] | Santiago Pastorino | 2010-03-27 | 2 | -44/+47 |
| | | | | Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | flatten not needed here | Santiago Pastorino | 2010-03-27 | 1 | -2/+2 |
| | | | | Signed-off-by: José Valim <jose.valim@gmail.com> | ||||
* | Remove stray instance variable to resolve serialization problem [#3769 ↵ | wycats | 2010-03-27 | 1 | -3/+9 |
| | | | | state:resolved] (ht: Joe Rafaniello) | ||||
* | Don't write to /dev/null on mingw. [#3921 state:resolved] (ht: Patrick ↵ | wycats | 2010-03-27 | 1 | -1/+1 |
| | | | | Daryll Glandien) | ||||
* | Limit Array#extract_options! to directl instances of Hash and HWIA. Add ↵ | wycats | 2010-03-27 | 2 | -1/+20 |
| | | | | extractable_options? to Hash so that subclasses of Hash can opt-into extractable behavior. This fixes an issue where respond_with wasn't working with subclasses of Hash that were provided by other libraries (such as CouchDB or Mashie) [#4145 state:resolved] | ||||
* | fixing inconsistency with cattr_reader and matter_reader [#4172 state:resolved] | Kabari Hendrick | 2010-03-27 | 2 | -10/+17 |
| | | | Signed-off-by: wycats <wycats@gmail.com> | ||||
* | Merge branch 'master' into docrails | wycats | 2010-03-26 | 20 | -23/+69 |
|\ | |||||
| * | Move application configuration to the application configuration object, ↵ | José Valim | 2010-03-26 | 1 | -4/+2 |
| | | | | | | | | remove railtie_name and engine_name and allow to set the configuration object. | ||||
| * | Ensure to reference the proper TZInfo namespace [#4268 state:resolved]. | José Valim | 2010-03-26 | 1 | -2/+2 |
| | | |||||
| * | Add support for a type=binary with an optional encoding=base64. If the ↵ | Josh Franklin | 2010-03-26 | 1 | -0/+9 |
| | | | | | | | | | | | | encoding attribute is absent, the data is considered unencoded. [#2966 state:resolved] | ||||
| * | avoid active_support/core_ext/time/conversions.rb warnings | Santiago Pastorino | 2010-03-21 | 1 | -0/+1 |
| | | | | | | | | | | | | [#4250 state:committed] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
| * | Singleton classes returns parent's methods with instance_methods(false) and ↵ | José Valim | 2010-03-19 | 1 | -1/+3 |
| | | | | | | | | this makes remove_method in Module#delegate fail. Add a test case and fix the bug. | ||||
| * | AS datetime conversions now require AS time conversions | snusnu | 2010-03-18 | 1 | -0/+1 |
| | | |||||
| * | Merge branch 'master' of gitproxy:rails/rails | José Valim | 2010-03-17 | 1 | -1/+2 |
| |\ | |||||
| | * | Require AS singleton_class code in AS output_safety | snusnu | 2010-03-17 | 1 | -1/+2 |
| | | | | | | | | | | | | Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
| * | | Mark bang instrumentations as something that you shuold not be listening to. | José Valim | 2010-03-17 | 1 | -1/+1 |
| |/ | |||||
| * | fixing activemodel tests. [#4210 state:resolved] | Aaron Patterson | 2010-03-17 | 1 | -0/+1 |
| | | | | | | | | Signed-off-by: wycats <wycats@gmail.com> | ||||
| * | Move require closer to home | Jeremy Kemper | 2010-03-17 | 1 | -0/+1 |
| | | |||||
| * | fixing activerecord tests [#4205 state:resolved] | Aaron Patterson | 2010-03-17 | 1 | -2/+2 |
| | | | | | | | | Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> | ||||
| * | Any reason we can't just use the slow 1.8 path for transliteration in 1.9? | wycats | 2010-03-17 | 1 | -2/+2 |
| | | |||||
| * | Fix some more warnings on 1.9 | wycats | 2010-03-17 | 1 | -0/+2 |
| | | |||||
| * | Remove some 1.9 warnings (resulting in some fixed bugs). Remaining AM ↵ | wycats | 2010-03-17 | 9 | -10/+29 |
| | | | | | | | | warnings are in dependencies. | ||||
| * | Eliminate warnings for AM on 1.8 | wycats | 2010-03-16 | 3 | -2/+13 |
| | | |||||
| * | Improve performance of multibyte utils. | James Golick | 2010-03-16 | 1 | -1/+1 |
| | | | | | | | | | | | | | | Switch from using String#match to using String#=~. There's no need to generate a MatchData for each iteration since we're not using it. Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net> |