aboutsummaryrefslogtreecommitdiffstats
path: root/activerecord
Commit message (Collapse)AuthorAgeFilesLines
* Remove ability update locking_column valuebogdanvlviv2017-03-164-46/+4
|
* Fix fragile test (`AssociationProxyTest#test_save_on_parent_saves_children`)Fumiaki MATSUSHIMA2017-03-161-36/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we run only following tests: - test/cases/scoping/default_scoping_test.rb - test/cases/associations_test.rb ``` $ cat Rakefile.test require "rake/testtask" ENV["ARCONN"] = "postgresql" Rake::TestTask.new do |t| t.libs << "test" t.test_files = %w( test/cases/scoping/default_scoping_test.rb test/cases/associations_test.rb ) end ``` a test will fail: ``` $ bundle exec rake test -f Rakefile.test /app/activesupport/lib/active_support/core_ext/enumerable.rb:20: warning: method redefined; discarding old sum Using postgresql Run options: --seed 11830 # Running: .........................................................................................F................ Finished in 6.939055s, 15.2759 runs/s, 27.9577 assertions/s. 1) Failure: AssociationProxyTest#test_save_on_parent_saves_children [/app/activerecord/test/cases/associations_test.rb:185]: Expected: 1 Actual: 2 106 runs, 194 assertions, 1 failures, 0 errors, 0 skips rake aborted! Command failed with status (1) /usr/local/bin/bundle:22:in `load' /usr/local/bin/bundle:22:in `<main>' Tasks: TOP => test (See full trace by running task with --trace) ``` In #28083, change `self.use_transactional_tests` to `false` but we forget to clean-up fixture. However we don't have to disable transaction except a few tests.
* Remove unused `@txn` variableEugene Kenny2017-03-151-1/+0
| | | | | | This was added in c24c885209ac2334dc6f798c394a821ee270bec6, removed in b89ffe7f0047eb614e42232a21201b317b880755, and then (unintentionally?) reintroduced in 2d7ae1b08ee2a10b12cbfeef3a6cc6da55b57df6.
* Revert "Merge pull request #28369 from ↵Matthew Draper2017-03-141-2/+2
| | | | | | | | | mylake/reduce-postgresql-adapter-memory-bloat" This reverts commit 192db64452d148c7b51713979459e38407380dc6, reversing changes made to 9893955363cf6358556ed3b36f4538d5b54e9d17. We can't sacrifice correctness for performance.
* Merge pull request #28369 from mylake/reduce-postgresql-adapter-memory-bloatRafael França2017-03-131-2/+2
|\ | | | | 500x memory reduction of 10k schemas for postgresql adapter
| * Use “distinct” to filter redundant types from pg_typeto reduce memory ↵mylake2017-03-101-2/+2
| | | | | | | | bloat especially in multi-schema structure database
* | Merge pull request #28068 from kamipo/refactor_data_sourcesRafael França2017-03-138-205/+156
|\ \ | | | | | | Extract `data_source_sql` to refactor data source statements
| * | Prefer `USING (column_list)` for joining information_schemaRyuta Kamizono2017-02-251-3/+1
| | |
| * | Extract `data_source_sql` to refactor data source statementsRyuta Kamizono2017-02-208-202/+155
| | |
* | | Merge pull request #28017 from mtsmfm/suppress-dep-warnMatthew Draper2017-03-133-0/+6
|\ \ \ | | | | | | | | Suppress deprecation warning `implementing to_yaml is deprecated`
| * | | Suppress deprecation warning `implementing to_yaml is deprecated`Fumiaki MATSUSHIMA2017-02-153-0/+6
| | | |
* | | | Merge pull request #28371 from kamipo/simplify_countAndrew White2017-03-121-7/+4
|\ \ \ \ | | | | | | | | | | Simply forward `Calculations#count` to `Enumerable#count`
| * | | | Simply forward `Calculations#count` to `Enumerable#count`Ryuta Kamizono2017-03-101-7/+4
| | |_|/ | |/| | | | | | | | | | | | | | | | | | | | | | Follow up of #24203. Since b644964b `ActiveRecord::Relation` includes `Enumerable` so it is enough to call `super` simply.
* | | | Merge pull request #28372 from kamipo/delegate_as_jsonAndrew White2017-03-123-6/+2
|\ \ \ \ | | | | | | | | | | Simply delegate `as_json` to `records`
| * | | | Simply delegate `as_json` to `records`Ryuta Kamizono2017-03-103-6/+2
| |/ / /
* / / / Remove conditions parameter from `NullRelation#delete_all`Ryuta Kamizono2017-03-101-1/+1
|/ / / | | | | | | | | | Follow up of e7381d289e4f8751dcec9553dcb4d32153bd922b.
* | | Add missing credit [ci skip]Ryuta Kamizono2017-03-101-0/+2
| | |
* | | Merge pull request #28351 from kamipo/deprecate_schema_migrations_table_nameRafael Mendonça França2017-03-093-0/+13
|\ \ \ | | | | | | | | | | | | Deprecate `Migrator.schema_migrations_table_name`
| * | | Deprecate `Migrator.schema_migrations_table_name`Ryuta Kamizono2017-03-093-0/+13
| | | | | | | | | | | | | | | | | | | | Since 67fba0cf `SchemaMigration` model was extracted. Use `SchemaMigration.table_name` instead.
* | | | Fix select with block doesn't return newly built records in has_many associationRyuta Kamizono2017-03-093-2/+14
|/ / / | | | | | | | | | | | | | | | | | | | | | The `select` in `QueryMethods` is also an enumerable method. Enumerable methods with block should delegate to `records` on `CollectionProxy`, not `scope`. Fixes #28348.
* | | Check whether `Rails.application` defined before calling itAndrew White2017-03-072-1/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In #27674 we changed the migration generator to generate migrations at the path defined in `Rails.application.config.paths` however the code checked for the presence of the `Rails` constant but not the `Rails.application` method which caused problems when using Active Record and generators outside of the context of a Rails application. Fixes #28325.
* | | Merge pull request #28305 from kamipo/extract_schema_migration_all_versionsAndrew White2017-03-063-4/+8
|\ \ \ | | | | | | | | Extract `SchemaMigration.all_versions`
| * | | Extract `SchemaMigration.all_versions`Ryuta Kamizono2017-03-063-4/+8
| | | | | | | | | | | | | | | | | | | | Use `SchemaMigration.all_versions` instead of `SchemaMigration.all.map(&:version)` to avoid to instantiate AR objects.
* | | | Fix `deserialize` with JSON arrayRyuta Kamizono2017-03-063-1/+17
|/ / / | | | | | | | | | Fixes #28285.
* | | Remove useless `Migrator.schema_migrations_table_name`Ryuta Kamizono2017-03-053-12/+8
| | | | | | | | | | | | Simply use `SchemaMigration.table_name` instead.
* | | Merge branch 'master' of github.com:rails/docrailsVijay Dev2017-03-041-1/+1
|\ \ \
| * | | Remove duplicated "does" [ci skip]Ryuta Kamizono2017-02-141-1/+1
| | | |
* | | | Don't hide method that's been documented API since Rails 2.1Andrew White2017-03-041-3/+3
| | | |
* | | | Fix `rake db:schema:load` with subdirectoriesRyuta Kamizono2017-03-053-5/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Related #25174. `db:schema:load` doesn't work with subdirectories like previous `db:migrate:status`. `Migrator.migration_files` should be used in `assume_migrated_upto_version` to fix the issue.
* | | | Fix `rake db:migrate:status` with subdirectoriesRyuta Kamizono2017-03-044-24/+75
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | `db:migrate` supports subdirectories and have a test. https://github.com/rails/rails/blob/v5.1.0.beta1/activerecord/test/cases/migrator_test.rb#L78-L85 But `db:migrate:status` doesn't work with subdirectories. It is due to `Dir.foreach(path)` is not the same with `Dir["#{path}/**/[0-9]*_*.rb"]`. I extracted `migration_files` and sharing it in the both to fix the issue. And added tests for `db:migrate:status`.
* | | | [ci skip] Add CHANGELOG entry for #28282Andrew White2017-03-041-0/+10
| | | |
* | | | Don't share `options` with a reference type columnRyuta Kamizono2017-03-042-7/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sharing `options` causes some unexpected behavior. If `limit: 2` is specified, this means that 2 bytes integer for a reference id column and 2 chars string for a reference type column. Another example, if `unsigned: true` is specified, this means that unsigned integer for a reference id column, but a invalid option for a reference type column. So `options` should not be shared with a reference type column.
* | | | remove unused parametersAaron Patterson2017-03-031-6/+6
| | | |
* | | | `join_keys` no longer needs a class passed to itAaron Patterson2017-03-033-30/+33
| | | | | | | | | | | | | | | | | | | | | | | | Reflections only use their own information to create a `join_keys` object. This means that we can call `join_keys` on a reflection object and have it be context-free.
* | | | refactor so there is only one JoinKeys factory methodAaron Patterson2017-03-031-4/+15
| | | |
* | | | ask reflection for klass join reflectionAaron Patterson2017-03-032-13/+16
| | | |
* | | | Move join scopes on to the reflection objectAaron Patterson2017-03-032-8/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Scopes can only ever be *not* reflection objects when they are passed in to the Reflection constructor. Given this fact, we can eliminate is_a checks and an intermediate array object by just asking the reflection object for join scopes.
* | | | Remove `node` parameter to `join_constraints`Aaron Patterson2017-03-032-3/+3
| | | | | | | | | | | | | | | | | | | | I don't think we actually need this parameter anymore. Nobody seems to be using it.
* | | | Merge pull request #28242 from rails/add-fractional-support-to-datetime-changeAndrew White2017-03-021-1/+1
|\ \ \ \ | | | | | | | | | | Update `DateTime#change` to support usec and nsec
| * | | | Update `DateTime#change` to support usec and nsecAndrew White2017-03-021-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adding support for these options now allows us to update the `DateTime#end_of` methods to match the equivalent `Time#end_of` methods, e.g: datetime = DateTime.now.end_of_day datetime.nsec == 999999999 # => true Fixes #21424.
* | | | | Merge pull request #28220 from kamipo/valid_type_accept_only_supported_typesRafael França2017-03-015-42/+10
|\ \ \ \ \ | |/ / / / |/| | | | `valid_type?` should accept only supported types
| * | | | `valid_type?` should accept only supported typesRyuta Kamizono2017-02-285-42/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | `valid_type?` is used in schema dumper to determine if a type is supported. So if `valid_type?(:foobar)` is true, it means that schema dumper is allowed to create `t.foobar`. But it doesn't work. I think that `valid_type?` should accept only supported types. https://github.com/rails/rails/blob/v5.1.0.beta1/activerecord/lib/active_record/schema_dumper.rb#L135-L142 ```ruby columns.each do |column| raise StandardError, "Unknown type '#{column.sql_type}' for column '#{column.name}'" unless @connection.valid_type?(column.type) next if column.name == pk type, colspec = @connection.column_spec(column) tbl.print " t.#{type} #{column.name.inspect}" tbl.print ", #{format_colspec(colspec)}" if colspec.present? tbl.puts end ```
* | | | | Whitelist adapters that support aliases in `HAVING` clauseAndrew White2017-03-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Support for using `SELECT` column or expression aliases in the `HAVING` clause isn't part of the SQL standard so it's better to whitelist the test for adapters where we know it works and skip it on others.
* | | | | Merge pull request #25296 from kamipo/use_inspect_for_type_cast_for_schemaRafael França2017-02-282-7/+12
|\ \ \ \ \ | | | | | | | | | | | | Use `inspect` in `type_cast_for_schema` for date/time and decimal values
| * | | | | Use `inspect` in `type_cast_for_schema` for date/time and decimal valuesRyuta Kamizono2016-12-112-7/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently dumping defaults on schema is inconsistent. Before: ```ruby create_table "defaults", force: :cascade do |t| t.string "string_with_default", default: "Hello!" t.date "date_with_default", default: '2014-06-05' t.datetime "datetime_with_default", default: '2014-06-05 07:17:04' t.time "time_with_default", default: '2000-01-01 07:17:04' t.decimal "decimal_with_default", default: 1234567890 end ``` After: ```ruby create_table "defaults", force: :cascade do |t| t.string "string_with_default", default: "Hello!" t.date "date_with_default", default: "2014-06-05" t.datetime "datetime_with_default", default: "2014-06-05 07:17:04" t.time "time_with_default", default: "2000-01-01 07:17:04" t.decimal "decimal_with_default", default: "1234567890" end ```
* | | | | | Merge pull request #28210 from cefigueiredo/improve_required_testRafael França2017-02-281-7/+35
|\ \ \ \ \ \ | | | | | | | | | | | | | | Make required by default test for belongs_to association clearer
| * | | | | | Ensure that tests will reset belongs_to_requierd_by_default configCarlos Figueiredo2017-02-271-23/+27
| | | | | | |
| * | | | | | Make required by default test for belongs_to association clearerCarlos Figueiredo2017-02-271-1/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since #18937 `belongs_to` associations receive a setting to determine if it should be or not treated as `required` by default. While the tests were still passing, it was not evident that the "default" behaviour for `required` could change in fuction of a setting, that is set by default for fresh Rails5 apps, but not for upgraded apps. This commit try to relate them to make it clear what is the behaviour expected when the setting is set as `true` or not set.
* | | | | | | Merge pull request #28217 from ↵Rafael França2017-02-282-8/+17
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | kamipo/create_join_table_respects_reference_key_type `create_join_table` should respect `references` column type
| * | | | | | | `create_join_table` should respect `references` column typeRyuta Kamizono2017-02-282-8/+17
| | |_|/ / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Follow up of #26266. The default type of `primary_key` and `references` were changed to `bigint` since #26266. But `create_join_table` column type is still `integer`. It should respect `references` column type.