aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* we have `with_env_tz` as global test helper. Remove duplicate.Yves Senn2014-01-162-14/+0
|
* isolate class attribute assignment in `migration_test.rb`Yves Senn2014-01-161-22/+33
| | | | | | | | This makes the tests inside `migration_test.rb` order independent. The assignments to `Reminder.table_name_prefix` and `Reminder.table_name_suffix` previously leaked and had impact on following test cases. This patch isolates the assignments on a throw-away subclass.
* use `teardown` for cleanup, not `setup`.Yves Senn2014-01-161-5/+2
| | | | | | | | | | | | This solves order dependent issues where the last test leaked the query cache state. This resulted in the following error: ``` QueryCacheExpiryTest#test_cache_is_expired_by_habtm_delete [test/cases/query_cache_test.rb:275]: not all expectations were satisfied unsatisfied expectations: - expected exactly twice, invoked once: #<ActiveRecord::ConnectionAdapters::SQLite3Adapter:0x7f93e0c65838>.clear_query_cache(any_parameters) ```
* tests without transactional fixtures need to cleanup afterwards.Yves Senn2014-01-161-0/+7
| | | | This fixes test order issues.
* no need to `return skip` in tests. `skip` is enough.Yves Senn2014-01-164-4/+4
|
* methods are defined right after the module_eval, so we don't need to doAaron Patterson2014-01-151-2/+2
| | | | any line number maths
* Get rid of unused TransactionError constantCarlos Antonio da Silva2014-01-151-3/+0
| | | | | | This has been added 9 years ago in a677da209b16f43198b3485dda89dce862fb9bfb, and removed 6 years ago in 38f8252e2d0a109d1b833d6b289cd989e7bfffe4.
* Avoid converting :on option to array twice when defining commit/rollback ↵Carlos Antonio da Silva2014-01-151-3/+2
| | | | callbacks
* Unify changelog entries about single quotes [ci skip]Carlos Antonio da Silva2014-01-151-5/+1
|
* Merge pull request #13713 from chulkilee/use-single-quotesCarlos Antonio da Silva2014-01-158-20/+24
|\ | | | | Use single quotes in generated files
| * Use single quotes in generated filesChulki Lee2014-01-148-20/+24
| |
* | rm dup codeAaron Patterson2014-01-141-13/+0
| | | | | | | | https://github.com/rails/rails/blob/e64a83cdbd3ce836f3339c600c1006b20c23d8fd/activerecord/test/cases/relation/merging_test.rb#L80-91
* | directly create the ARel ASTAaron Patterson2014-01-141-6/+4
| | | | | | | | rather than allocating a new Relation, just make the AST directly
* | Updated comment to mention the enum mapping class method [ci skip]Godfrey Chan2014-01-141-2/+3
| |
* | "string" isn't a valid column type for ActiveRecordAaron Patterson2014-01-141-1/+1
|/ | | | | Conflicts: activerecord/test/cases/adapters/sqlite3/sqlite3_adapter_test.rb
* don't establish a new connection when testing with `sqlite3_mem`.Yves Senn2014-01-141-15/+21
| | | | | | | This fixes broken `rake test_sqlite3_mem` suite for Active Record. The problem is that that the old database with the schema is lost when establishing a new connection. Upon reconnting we are left with a blank database and tests down the line start failing.
* Merge pull request #13350 from ccutrer/sqlite-partial-indexesYves Senn2014-01-143-1/+25
|\ | | | | sqlite >= 3.8.0 supports partial indexes
| * sqlite >= 3.8.0 supports partial indexesCody Cutrer2014-01-143-1/+25
|/
* Merge pull request #13710 from ujjwalt/hotfix/stiAndrew White2014-01-145-8/+42
|\ | | | | Active Record's subclass_from_attributes shouldn't assume :type is for STI unless there is a type column.
| * Don't try to get the subclass if the inheritance column doesn't existUjjwal Thaakar2014-01-145-8/+42
|/ | | | | | | The `subclass_from_attrs` method is called even if the column specified by the `inheritance_column` setting doesn't exist. This prevents setting associations via the attributes hash if the association name clashes with the value of the setting, typically `:type`. This worked previously in Rails 3.2.
* Enum mappings are now exposed via class methods instead of constants.Godfrey Chan2014-01-143-10/+31
| | | | | | | | | | | | | | | | Example: class Conversation < ActiveRecord::Base enum status: [ :active, :archived ] end Before: Conversation::STATUS # => { "active" => 0, "archived" => 1 } After: Conversation.statuses # => { "active" => 0, "archived" => 1 }
* Merge pull request #13705 from mess110/controller_generator_route_quotesCarlos Antonio da Silva2014-01-145-11/+15
|\ | | | | Single quotes for controller generated routes. This is more consistent with all other Rails generated code.
| * single quotes for controller generated routesCristian Mircea Messel2014-01-145-11/+15
| | | | | | | | | | | | | | | | | | | | Write routes in route.rb with single quotes get 'welcome/index' instead of get "welcome/index"
* | Merge pull request #13706 from prathamesh-sonpatki/issue-13125Yves Senn2014-01-141-3/+10
|\ \ | | | | | | Fix fields_for documentation with index option [ci skip]
| * | Fix fields_for documentation with index option [ci skip]Prathamesh Sonpatki2014-01-141-3/+10
| |/ | | | | | | | | | | | | - fields_for documentation with index option was wrong. - It does not work with passing model as it is. - Changed the example by passing id of the address object. - Fixes #13125.
* | Merge pull request #13709 from akshay-vishnoi/typoYves Senn2014-01-141-1/+1
|\ \ | | | | | | [ci skip] Grammar correction
| * | [ci skip] Grammar correctionAkshay Vishnoi2014-01-141-1/+1
| | |
* | | quick pass through Active Record CHANGELOG. [ci skip]Yves Senn2014-01-141-7/+9
| | |
* | | Merge pull request #13669 from chulkilee/name-error-nameYves Senn2014-01-143-2/+8
|\ \ \ | | | | | | | | Set NameError#name
| * | | Set NameError#nameChulki Lee2014-01-133-2/+8
| | | |
* | | | Merge pull request #13707 from udaykadaboina/docModificationDamien Mathieu2014-01-131-2/+2
|\ \ \ \ | |_|_|/ |/| | | [ci skip] Added alias to CSRF
| * | | [ci skip] Added alias to CSRFUday Kadaboina2014-01-141-2/+2
|/ / /
* | | unscope should remove bind values associated with the whereAaron Patterson2014-01-132-0/+11
| | |
* | | reverse_order_value= is not private, so no need to sendAaron Patterson2014-01-131-1/+1
| | | | | | | | | | | | It's not public API, but is not private scoped.
* | | avoid more dynamic symbolsAaron Patterson2014-01-131-1/+1
| | |
* | | no need to to_symAaron Patterson2014-01-131-2/+2
| | |
* | | Merge pull request #13704 from waynn/patch-4Rafael Mendonça França2014-01-131-2/+2
|\ \ \ | | | | | | | | "serie" => "series"
| * | | "serie" => "series"Waynn Lue2014-01-131-2/+2
| | | |
* | | | recover from test runs that leave the database in a bad stateAaron Patterson2014-01-131-1/+1
| | | |
* | | | updates screenshot data [ci skip]Xavier Noria2014-01-131-1/+1
|/ / /
* / / revises a few things in the getting started guide [ci skip]Xavier Noria2014-01-135-121/+129
|/ / | | | | | | More to come.
* | Merge pull request #13703 from schneems/schneems/rail-secrets-keyRafael Mendonça França2014-01-132-3/+3
|\ \ | | | | | | Favor canonical environment variables for secrets
| * | Favor canonical environment variables for secretsschneems2014-01-132-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | Prefixing an environment variable with `RAILS_` should be used when there is otherwise a conflict, such as `RAILS_ENV` or if it is being used for a very Rails specific value. If we remove the prefix of `RAILS_` in the case of `RAILS_SECRET_KEY_BASE` then we can push for a pseudo standard among other frameworks that will accept a common environment key `SECRET_KEY_BASE` to keep your app secure. This is beneficial for containerized deployments such as docker, Heroku, etc. So that the container need to know one less thing about your app (it can just set it no-matter what language or framework you are using). This change also allows the defaults to be consistent with the way the secret key is accessed so `secrets.secret_key_base` is aliased to `SECRET_KEY_BASE` instead of `RAILS_SECRET_KEY_BASE`.
* | | Merge pull request #13702 from kuldeepaggarwal/docs-fixRafael Mendonça França2014-01-131-1/+1
|\ \ \ | |/ / |/| | removed extra comma [ci skip]
| * | removed extra comma [ci skip]Kuldeep Aggarwal2014-01-141-1/+1
|/ /
* | Only lookup `config.log_level` for stdlib `::Logger`. Closes #11665.Yves Senn2014-01-133-2/+26
| | | | | | | | | | | | | | | | This prevents Rails from assigning meaningless log levels to third party loggers like log4r. If `Rails.logger` is not `kind_of?(::Logger)` we simply assign the `config.log_level` as is. This bug was introduced by #11665.
* | Merge pull request #13699 from rystraum/dependent-restrict-removalRafael Mendonça França2014-01-131-0/+4
|\ \ | | | | | | Updated Changelog to reflect removal of :dependent => :restrict
| * | Updated Changelog to reflect removal of :dependent => :restrictRystraum Gamonez2014-01-141-0/+4
|/ / | | | | | | | | Based on https://github.com/rails/rails/issues/2502, the option has been removed.
* | Merge pull request #13693 from gaurish/typoYves Senn2014-01-137-6/+39
|\ \ | | | | | | [JRuby] Fix incorrect RUBY_ENGINE value
| * | standardize on jruby_skip & rbx_skipGaurish Sharma2014-01-137-6/+39
| | | | | | | | | | | | | | | | | | This Adds helpers(jruby_skip & rbx_skip). In Future, Plan is to use these helpers instead of calls directly to RUBY_ENGINE/RbConfig/JRUBY_VERSION
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-27 04:02:09 +0000