aboutsummaryrefslogtreecommitdiffstats
path: root/activerecord/lib/active_record
Commit message (Collapse)AuthorAgeFilesLines
* Fix typoRafael Mendonça França2013-03-251-1/+1
|
* The sql_type method called here is fromRafael Mendonça França2013-03-252-7/+6
| | | | | | | ActiveRecord::ConnectionAdapters::Column See https://github.com/rails/rails/blob/28b8ca766e3e7c6c43d3ae900c99f8377153c62/activerecord/lib/active_record/connection_adapters/column.rb#L16
* Move away from column.sql_type in untested code tooMartin Schürrer2013-03-251-5/+6
|
* Make Postgres point type correspond to ruby array with two floats insideMartin Schürrer2013-03-254-6/+43
|
* Created a layer of abstraction for the valid type checking in schema dumper. ↵Ranjay Krishna2013-03-254-1/+13
| | | | Now, connection handles the check for valid types so that each database can handle the changes individually.
* Merge pull request #9489 from obrie/fix/overridden_defaults_changed_attributesJeremy Kemper2013-03-241-3/+11
|\ | | | | Fix ActiveRecord locking column defaults not getting persisted
| * Fix ActiveRecord locking column defaults not getting persistedAaron Pfeifer2013-03-181-3/+11
| | | | | | | | | | | | | | | | | | | | | | When partial inserts are enabled, overridden db defaults are ignored. This results in locking columns having a nil value for new records if the db default is null. This happens because the list of changed attributes for new records is always assumed to be empty. Solution: When a new record's default attributes are set, also initialize the list of changed attributes by comparing current values against what's stored as the column defaults in the database.
* | Merge pull request #8501 from charliesome/version-to-sRafael Mendonça França2013-03-231-6/+7
|\ \ | | | | | | Add #to_s method to VERSION modules
| * | Add version method to top level modulesCharlie Somerville2013-03-211-6/+7
| | |
* | | Merge branch 'jlxw-patch-2'Rafael Mendonça França2013-03-231-2/+2
|\ \ \ | | | | | | | | | | | | Closes #9184
| * | | Fix regex to strip quotations from hstore valuesjlxw2013-03-231-2/+2
| | | | | | | | | | | | | | | | Previously regex did not strip quotation marks where hstore values were multi-line strings.
* | | | Merge pull request #9891 from doabit/additional_index_supportRafael Mendonça França2013-03-234-4/+31
|\ \ \ \ | |/ / / |/| | | | | | | | | | | Custom index type support with :using. Closes #9451
| * | | Custom index type support with :using.doabit2013-03-244-4/+31
| | | |
* | | | Merge pull request #9889 from neerajdotname/unscopedRafael Mendonça França2013-03-231-8/+0
|\ \ \ \ | |/ / / |/| | | Unscoped works with other named scope even without block form
| * | | unscoped works with named scope [ci skip]Neeraj Singh2013-03-231-8/+0
| | | | | | | | | | | | | | | | | | | | Update comment to reflect that unscoped works with named scope even when named scope is using without block form
* | | | Merge pull request #9880 from wangjohn/refactoring_attribute_assignmentRafael Mendonça França2013-03-231-5/+5
|\ \ \ \ | | | | | | | | | | Refactoring validate_missing_parameters to validate_required_parameters.
| * | | | Refactoring validate_missing_parameters to validate_required_parameters.wangjohn2013-03-231-5/+5
| |/ / /
* / / / Autoload the AlterTable classRafael Mendonça França2013-03-231-0/+1
|/ / /
* | | add uuid primary key supportAaron Patterson2013-03-224-3/+20
| | |
* | | separate primary key from column typeAaron Patterson2013-03-221-10/+11
| | |
* | | push the mysql add_column up to the abstract adapterAaron Patterson2013-03-224-22/+35
| | |
* | | allow multiple add columnsAaron Patterson2013-03-223-6/+6
| | |
* | | pull add_column_options! off the pg connection classAaron Patterson2013-03-222-13/+13
| | |
* | | add a pg visitor for dealing with schema modificationAaron Patterson2013-03-221-4/+21
| | |
* | | push alter table add column sql in to the schema modification visitorAaron Patterson2013-03-223-14/+51
| | |
* | | there is no reason to check for an already defined columnAaron Patterson2013-03-221-1/+1
| | |
* | | push column initialization down to the factory methodAaron Patterson2013-03-221-13/+13
| | |
* | | @columns list is no longer necessaryAaron Patterson2013-03-221-8/+4
| | |
* | | keep ivars private, do not manipulate them outside their owner objectAaron Patterson2013-03-222-1/+5
| | |
* | | factory methods should not alter object stateAaron Patterson2013-03-221-5/+7
| | |
* | | push SQL generation inside the schema creation objectAaron Patterson2013-03-224-12/+20
| | |
* | | mostly decouple TableDefinition from the database connectionAaron Patterson2013-03-223-5/+5
| | |
* | | remove to_sql from TableDefinitionAaron Patterson2013-03-223-9/+5
| | |
* | | decouple column definition from the database connectionAaron Patterson2013-03-222-8/+8
| | |
* | | stop depending on sql_type in pgAaron Patterson2013-03-221-6/+8
| | |
* | | remove knowledge of SQL from the column definition objectAaron Patterson2013-03-222-21/+43
| | |
* | | When we pass id to update_attributes it will try to set new id for that recordDmitry Vorotilin2013-03-222-2/+2
| | |
* | | Allows setting fixture_path to a PathnameAlexander Murmann and Can Berk Güder2013-03-211-1/+1
| | |
* | | Adding a bang to method name of raise_on_type_mismatch.wangjohn2013-03-216-8/+8
|/ /
* | Update other counter caches on destroyIan Young2013-03-204-2/+17
| |
* | use `connect_poll` on pg so that reaping does not hurt the connectionAaron Patterson2013-03-202-3/+4
| |
* | Revert "default the reaping frequency to 10 seconds"Aaron Patterson2013-03-201-1/+1
| | | | | | | | | | | | | | mysql can't handle a parallel thread pinging the connection, so we can get wrong results or segvs This reverts commit 7cc588b684f6d1af3e7fab1edfa6715e269e41a2.
* | Properly named variable inside blockAnupam Choudhury2013-03-201-2/+2
| | | | | | | | Closes #9824.
* | Fix some typos in AR- CHANGELOG, tests, method doc. fixedVipul A M2013-03-202-2/+2
| |
* | Changed the call to .sort.last to .max when computing the migration version. Youwangjohn2013-03-191-1/+1
| | | | | | | | | | do not actually need to sort everything (incurs more cost than just a simple max).
* | Calling find() on an association with an inverse will now check to seewangjohn2013-03-191-2/+3
|/ | | | | if the association already holds that record in memory before checking the database for the specified ids.
* Revert "Merge pull request #9784 from ↵Carlos Antonio da Silva2013-03-181-4/+4
| | | | | | | | | | vipulnsward/change_from_blank_to_empty_on_string" This reverts commit 9c4c05fc82e997b722dec4068c3aa27eaee69eb8, reversing changes made to 4620bdcefd0c88905a005f191496df887877b8f3. Reason: They're not completely interchangeable, since blank? will also check for strings containing spaces.
* Merge pull request #9784 from vipulnsward/change_from_blank_to_empty_on_stringCarlos Antonio da Silva2013-03-181-4/+4
|\ | | | | Change from blank? to empty? on obvious string values to save extra calls
| * change from blank? to empty? on obvious string values to save extra method ↵Vipul A M2013-03-191-4/+4
| | | | | | | | calls
* | Merge pull request #9782 from vipulnsward/change_from_each_to_each_key_habtmRafael Mendonça França2013-03-181-1/+1
|\ \ | | | | | | Change from each to each_value;drop assignment in habtm