aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* fixing warnings with regexps on assert_matchAaron Patterson2010-10-011-2/+2
|
* removing more duplicate codeAaron Patterson2010-10-013-30/+17
|
* delete repeated codeAaron Patterson2010-10-013-36/+18
|
* remove another lolinjectAaron Patterson2010-10-011-1/+1
|
* use a method that actually existsAaron Patterson2010-10-011-1/+1
|
* speed up index_by by removing a lolinjectAaron Patterson2010-10-011-4/+1
|
* removing AS::Testing::Default in favor of just undefing default_testAaron Patterson2010-10-014-20/+2
|
* brings csrf_meta_tags back to the generated layoutXavier Noria2010-10-022-2/+2
| | | | | | | | | | After more discussion, it has be agreed that this kind of changes within reasonable margins are OK for 3.1. That is, it is fine to change a little bit the generators even if that means examples in existing books won't be exact. (Note that the singular csrf_meta_tag exists as an alias and thus those outdated examples will run, same for existing applications.)
* Arel::Sql::Engine.new does not do anything anymoreAaron Patterson2010-09-302-2/+2
|
* type_name should check for blank because people may have messed up databasesAaron Patterson2010-09-302-1/+15
|
* build_where should be privateAaron Patterson2010-09-301-2/+2
|
* type_name is never a blank string, so use faster .nil? callAaron Patterson2010-09-301-1/+1
|
* AssociationCollection#include? working properly for objects added with build ↵Marcelo Giorgi2010-09-304-0/+39
| | | | method [#3472 state:resolved]
* speeding up object instantiation by eliminating instance_evalAaron Patterson2010-09-301-7/+21
|
* changing map and include to findAaron Patterson2010-09-301-1/+1
|
* [#5441 state:resolved] refactoring code to determine aggregate columnAaron Patterson2010-09-301-11/+14
|
* group clause must be more specificAaron Patterson2010-09-301-1/+1
|
* Tests proving #5441oleg dashevskii2010-09-301-1/+11
|
* Fix 'rake db:create' is ignore encoding when using postgres [#5717 ↵yalab2010-09-301-1/+1
| | | | | | state:resolved] Signed-off-by: Santiago Pastorino <santiago@wyeworks.com>
* no need to check for nilNeeraj Singh2010-09-301-1/+1
|
* assert on empty array is always true. fixing test.Neeraj Singh2010-09-301-1/+1
|
* Revert "removing crazy finalizer code until there is proof that we need it"José Valim2010-09-301-0/+11
| | | | This reverts commit a40e3c1a9604ab3737ad2465c8f6a6db0fe0cc78.
* Add more information to comments.José Valim2010-09-301-3/+5
|
* Merge remote branch 'drogus/engines'José Valim2010-09-3011-47/+289
|\
| * Try to guess application's directory while trying to run the server, but ↵Piotr Sarnacki2010-09-301-0/+5
| | | | | | | | only if config.ru is not present in current dir
| * Moved Rails::RoutesReloader to Rails::Application::RoutesReloaderPiotr Sarnacki2010-09-303-56/+57
| |
| * Allow running generators for Engine with usage of other application.Piotr Sarnacki2010-09-301-1/+11
| | | | | | | | | | | | | | | | After that commit, developers can set ENGINE_PATH in ENGINE/scripts/rails file and load application's ./script/rails (most of the time it will be dummy application used for testing). When running ./script/rails g it will use application to boot up, but then it will use Engine's root and configuration for generators.
| * Added config.app_generators to allow configuring application's generators ↵Piotr Sarnacki2010-09-305-5/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | from railties. With config.generators becomes a way to configure generators for current instance only. For example: module Blog class Engine < Rails::Engine config.generators do |g| g.orm :active_record end config.app_generators do |g| g.test_framework :rspec end end end such definition sets :active_record as orm for engine and :rspec as test_framework for application. The values set with app_generators can be overwritten in application using config.generators as you would normally do: module MyApp class Application < Rails::Application config.generators do |g| g.test_framework :test_unit end end end
| * Move routes_reloader to a class: RoutesReloaderPiotr Sarnacki2010-09-303-23/+57
| |
| * Properly reload routes defined in class definitionPiotr Sarnacki2010-09-304-6/+63
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sometimes it's easier to define routes inside Engine or Application class definition (e.g. one file applications). The problem with such case is that if there is a plugin that has config/routes.rb file, it will trigger routes reload on application. Since routes definition for application is not in config/routes.rb file routes_reloader will fail to reload application's routes properly. With this commit you can pass routes definition as a block to routes method, which will allow to properly reload it: class MyApp::Application < Rails::Application routes do resources :users end end
| * Allow mounting engines at '/'Piotr Sarnacki2010-09-302-17/+78
| | | | | | | | | | Without that commit script_name always become '/', which results in paths like //posts/1 instead of /posts/1
| * Do not overwrite _railtie method on namespace while creating isolated engine ↵Piotr Sarnacki2010-09-302-12/+33
|/ | | | | | | | | or application. In order to run Engine as standalone application, you will need Rails::Application instance in the same namespace that engine one. It's very important to leave _railtie bound to whatever used "namespace" method first.
* Use .find here as it is simpler and faster.José Valim2010-09-301-1/+1
|
* avoid creating a block if possibleAaron Patterson2010-09-291-2/+2
|
* fixing regexp warningsAaron Patterson2010-09-291-1/+1
|
* tag value can be false, so nil? check is necessaryAaron Patterson2010-09-291-1/+1
|
* double negative is not goodNeeraj Singh2010-09-301-1/+1
|
* another case of extra nil? checkNeeraj Singh2010-09-301-1/+1
|
* no need to check for nil?Neeraj Singh2010-09-301-2/+2
|
* no need to check for nil?Neeraj Singh2010-09-301-1/+1
|
* no need of nil checkNeeraj Singh2010-09-301-1/+1
|
* no need of nil checkNeeraj Singh2010-09-291-1/+1
|
* removing crazy finalizer code until there is proof that we need itAaron Patterson2010-09-291-9/+0
|
* removing useless codeAaron Patterson2010-09-291-1/+0
|
* dry up method checking in the request objectAaron Patterson2010-09-291-10/+9
|
* reduce function calls on ArrayAaron Patterson2010-09-291-3/+2
|
* removing more lolinjectAaron Patterson2010-09-291-8/+7
|
* removing lollerjectAaron Patterson2010-09-291-1/+1
|
* self is implied, so remove selfAaron Patterson2010-09-291-1/+1
|
* dry up method missingAaron Patterson2010-09-291-5/+2
|