aboutsummaryrefslogtreecommitdiffstats
path: root/activerecord/lib/active_record/connection_adapters
Commit message (Collapse)AuthorAgeFilesLines
...
* | Fix sqlite3 collation parsing when using decimal columns.Martin Schuster2019-06-041-2/+2
| | | | | | | | | | | | If an sqlite3 table contains a decimal column behind columns with a collation definition, then parsing the collation of all preceeding columns will fail -- the collation will be missed without notice.
* | Merge pull request #36384 from ↵Ryuta Kamizono2019-06-031-1/+1
|\ \ | |/ |/| | | | | guigs/fix-invalid-schema-when-pk-column-has-comment Fix invalid schema dump when primary key column has a comment
| * Fix invalid schema dump when primary key column has a commentGuilherme Goettems Schneider2019-06-031-1/+1
| | | | | | | | | | | | | | | | Before this fix it would either generate an invalid schema, passing `comment` option twice to `create_table`, or it move the comment from primary key column to the table if table had no comment when the dump was generated. The situation now is that a comment on primary key will be ignored (not present on schema). Fixes #29966
* | Refactor `create_table`'s options separationRyuta Kamizono2019-06-032-12/+11
|/ | | | | | | | `create_table` and `t.column` have the same named options (e.g. `:comment`, `:primary_key`), so it should be separated table options from column options. Related #36373.
* Fix table comment also being applied to the primary key columnGuilherme Goettems Schneider2019-05-311-1/+1
|
* `:datetime` and `:time` columns allows `:precision` option [ci skip]Ryuta Kamizono2019-05-281-2/+3
|
* Use WeakRef to avoid leaking connection poolsJohn Hawthorn2019-05-271-11/+21
| | | | | | | | | | | | | | | | | | Prior to 3e2e8eeb9ea552bd4782538cf9348455f3d0e14a the Reaper thread would hold a reference to connection pools indefinitely, preventing the connection pool from being garbage collected, and also leaking the Thread. Since 3e2e8eeb9ea552bd4782538cf9348455f3d0e14a, there is only one Reaper Thread for all pools, however all pools are still stored in a class variable, preventing them from being garbage collected. This commit instead holds reference to the pools through a WeakRef. This way, connection pools referenced elsewhere will be reaped, any others will be able to be garbage collected. I don't love resorting to WeakRef to solve this, but I believe it's the simplest way to accomplish the the desired behaviour.
* Put all `explain` methods into `DatabaseStatements` moduleRyuta Kamizono2019-05-224-18/+14
| | | | | Almost all database statements methods except `explain` was moved into `DatabaseStatements` at #35922. This moves the last one method.
* Merge pull request #36296 from jhawthorn/dont_fear_the_reaperAaron Patterson2019-05-181-7/+25
|\ | | | | Use a single thread for all ConnectionPool Reapers
| * Use a single thread for all ConnectionPool ReapersJohn Hawthorn2019-05-171-7/+25
| | | | | | | | | | Previously we would spawn one thread per connection pool, which ends up being wasteful for apps with several connection pools.
* | Remove SQLite version support caveats [ci skip]Eugene Kenny2019-05-123-5/+3
|/ | | | | | | | | Since d1a74c1e012ed96f7179e53b9190b7da0a369e11, Active Record requires SQLite version 3.8.0 or greater, so savepoints and partial indexes are always available. That commit also added a runtime version check, so we can remove the minimum version requirement from the internal adapter documentation.
* Remove ignored_sql from SQLCounter by adding "TRANSACTION" to log nameYasuo Honda2019-05-084-12/+12
| | | | | | | | | | This commit adds "TRANSACTION" to savepoint and commit, rollback statements because none of savepoint statements were removed by #36153 since they are not "SCHEMA" statements. Although, only savepoint statements can be labeled as "TRANSACTION" I think all of transaction related method should add this label. Follow up #36153
* Should attempt `committed!`/`rolledback!` to all enrolled records in the ↵Ryuta Kamizono2019-05-071-4/+12
| | | | | | | | | | | | | | | | | | | | | | transaction Currently, `committed!`/`rolledback!` will only be attempted for the first enrolled record in the transaction, that will cause some problematic behaviors. The first one problem, `clear_transaction_record_state` won't be called even if the transaction is finalized except the first enrolled record. This means that de-duplicated records in the transaction won't refer latest state (e.g. won't happen rolling back record state). The second one problem, the enrolled order is not always the same as the order in which the actions actually happened, the first enrolled record may succeed no actions (e.g. `destroy` has already succeeded on another record during `before_destroy`), it will lose to fire any transactional callbacks. To avoid both problems, we should attempt `committed!`/`rolledback!` to all enrolled records in the transaction.
* Merge pull request #35998 from itsWill/add_documentation_for_add_indexEileen M. Uchitelle2019-04-191-0/+20
|\ | | | | Document algorithm: concurrent option for PostgreSQL [ci skip]
| * Document algorithm: concurrent options for PostgreSQL [ci skip]Guilherme Mansur2019-04-161-0/+20
| |
* | Merge pull request #35946 from alimi/cache-full-mysql-database-versionKasper Timm Hansen2019-04-163-5/+14
|\ \ | |/ |/| Cache full MySQL version in schema cache
| * Make changes per PR feedbackAli Ibrahim2019-04-122-9/+3
| | | | | | | | | | | | | | * Remove AbstractMysqlAdapter::Version since full_version_string will always be set. * Remove nodoc on private methods because private methods are not exposed in the docs.
| * Cache full MySQL version in schema cacheAli Ibrahim2019-04-113-6/+21
| | | | | | | | | | | | | | | | | | * The database version is cached in all the adapters, but this didn't include the full MySQL version. Anything that uses the full MySQL version would need to query the database to get that data even if they're using the schema cache. * Now the full MySQL version will be cached in the schema cache via the Version object.
* | make change_column_comment and change_table_comment invertibleYoshiyuki Kinjo2019-04-153-6/+21
| | | | | | | | | | | | | | | | | | We can revert migrations using `change_column_comment` or `change_table_comment` at current master. However, results are not what we expect: comments are remained in new status. This change tells previous comment to these methods in a way like `change_column_default`.
* | use PostgreSQL's bulk_alter_table implementationYoshiyuki Kinjo2019-04-133-41/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Running this migration on mysql at current master fails because `add_references_for_alter` is missing. ``` change_table :users, bulk: true do |t| t.references :article end ``` This is also true for postgresql adapter, but its `bulk_alter_table` implementation can fallback in such case. postgresql's implementation is desirable to prevent unknown failure like this.
* | Refactor around sql_type metadata and columnRyuta Kamizono2019-04-124-33/+35
| | | | | | | | | | | | * remove useless `@type_metadata` and `@array` * move the compatibility code (for array) into column * etc.
* | Merge pull request #35922 from ↵Rafael França2019-04-112-80/+80
|\ \ | |/ |/| | | | | michaelglass/move-sqlite-3-database-statements-into-database-statements make SQLite3 `last_inserted_id` private and organize DatabaseStatement methods
| * moves sqlite3 methods that mirror Abstract::DatabaseStatements into ↵Michael Glass2019-04-102-80/+80
| | | | | | | | Sqlite3::DatabaseStatements and make those that are private in Abstract::DatabaseStatements private for sqlite3
* | Adding type option example to the documentation [ci skip] (#35917)Roberto Miranda2019-04-101-0/+1
|/ | | | | | | | | | | | * Adding type option example to the documentation [ci skip] It was hard for me looking https://api.rubyonrails.org/ to find that there was a type option. Adding this to the doc would be helpful especially for application with old tables where the references are still an integer not bigint * Update activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb Co-Authored-By: robertomiranda <rjmaltamar@gmail.com>
* Accidentally lost `comment` in `Column#==` and `Column#hash`Ryuta Kamizono2019-04-101-2/+4
| | | | Refer #35875.
* Remove unused `sequence_name` in `sql_for_insert`Ryuta Kamizono2019-04-102-3/+3
| | | | | | | | | | All adapters (sqlite3, mysql2, postgresql, oracle-enhanced, sqlserver) doesn't use `sequence_name` in `sql_for_insert`. https://github.com/rsim/oracle-enhanced/blob/4e0db270a93859c9713fd079dbb315b9fe550e57/lib/active_record/connection_adapters/oracle_enhanced/database_statements.rb#L79-L85 https://github.com/rails-sqlserver/activerecord-sqlserver-adapter/blob/959fe8f49744460b876bc205c73259f8d4f37629/lib/active_record/connection_adapters/sqlserver/database_statements.rb#L226-L249 It can be handled in `exec_insert` like postgresql adapter if we want.
* There is no need to create `QueryAttribute` to just type cast a valueRyuta Kamizono2019-04-101-2/+1
|
* Merge pull request #35875 from Shopify/alloc-free-comparisonsRafael França2019-04-094-32/+38
|\ | | | | Improve == and hash methods on various schema cache structs to be allocation free.
| * Improve == and hash methods on various schema cache structs to be allocation ↵Jean Boussier2019-04-094-32/+38
| | | | | | | | | | | | | | | | free. The previous implementation would allocate 2 arrays per comparisons. I tried relying on Struct, but they do allocate one Hash inside `Struct#hash`.
* | Merge pull request #35909 from simi/alias-postgresql-adapterRyuta Kamizono2019-04-101-0/+1
|\ \ | | | | | | Bring back postgresql_version as an alias.
| * | Bring back postgresql_version as an alias.Josef Šimánek2019-04-091-0/+1
| |/
* / `get_database_version` is not public API [ci skip]Ryuta Kamizono2019-04-091-1/+1
|/
* Merge pull request #34800 from mqchau/mysqlCountDeleteRowInLockMatthew Draper2019-04-091-1/+3
|\ | | | | Wrap Mysql count of deleted rows in lock block to avoid conflict in test
| * Wrap Mysql count of deleted rows in lock block to avoid conflict in testQuan Chau2019-04-081-1/+3
| |
* | Merge pull request #35887 from kamipo/argument_errorRyuta Kamizono2019-04-093-6/+6
|\ \ | | | | | | Raise `ArgumentError` for invalid `:limit` and `:precision` like as other options
| * | Raise `ArgumentError` for invalid `:limit` and `:precision` like as other ↵Ryuta Kamizono2019-04-073-6/+6
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | options When I've added new `:size` option in #35071, I've found that invalid `:limit` and `:precision` raises `ActiveRecordError` unlike other invalid options. I think that is hard to distinguish argument errors and statement invalid errors since the `StatementInvalid` is a subclass of the `ActiveRecordError`. https://github.com/rails/rails/blob/c9e4c848eeeb8999b778fa1ae52185ca5537fffe/activerecord/lib/active_record/errors.rb#L103 ```ruby begin # execute any migration rescue ActiveRecord::StatementInvalid # statement invalid rescue ActiveRecord::ActiveRecordError, ArgumentError # `ActiveRecordError` except `StatementInvalid` is maybe an argument error end ``` I'd say this is the inconsistency worth fixing. Before: ```ruby add_column :items, :attr1, :binary, size: 10 # => ArgumentError add_column :items, :attr2, :decimal, scale: 10 # => ArgumentError add_column :items, :attr3, :integer, limit: 10 # => ActiveRecordError add_column :items, :attr4, :datetime, precision: 10 # => ActiveRecordError ``` After: ```ruby add_column :items, :attr1, :binary, size: 10 # => ArgumentError add_column :items, :attr2, :decimal, scale: 10 # => ArgumentError add_column :items, :attr3, :integer, limit: 10 # => ArgumentError add_column :items, :attr4, :datetime, precision: 10 # => ArgumentError ```
* | Merge pull request #35890 from kamipo/except_table_name_from_columnRyuta Kamizono2019-04-096-49/+43
|\ \ | | | | | | Except `table_name` from column objects
| * | Except `table_name` from column objectsRyuta Kamizono2019-04-086-49/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | The `table_name` was added at #23677 to detect whether serial column or not correctly. We can do that detection before initialize column object, it makes column object size smaller, and it probably helps column object de-duplication.
* | | When skipping duplicates in bulk insert on MySQL, avoid assigning id when ↵Bob Lail2019-04-081-2/+2
|/ / | | | | | | | | | | | | | | | | | | | | not specified If `id` is an `AUTONUMBER` column, then my former strategy here of assigning `no_op_column` to an arbitrary column would fail in this specific scenario: 1. `model.columns.first` is an AUTONUMBER column 2. `model.columns.first` is not assigned in the insert attributes I added three tests: the first test covers the actual error; the second test documents that this _isn't_ a problem when a value is given for the AUTONUMBER column and the third test ensures that this no-op strategy isn't secretly doing an UPSERT.
* / Fix GROUP BY with calculate longer name field to respect `table_alias_length`Ryuta Kamizono2019-04-083-6/+12
|/ | | | Follow up of c9e4c848eeeb8999b778fa1ae52185ca5537fffe.
* Clear query cache when truncate table(s)Ryuta Kamizono2019-04-041-1/+2
|
* Use `execute_batch2` rather than `execute_batch` to fix performance ↵Ryuta Kamizono2019-04-042-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | regression for fixture loading d8d6bd5 makes fixture loading to bulk statements by using `execute_batch` for sqlite3 adapter. But `execute_batch` is slower and it caused the performance regression for fixture loading. In sqlite3 1.4.0, it have new batch method `execute_batch2`. I've confirmed `execute_batch2` is extremely faster than `execute_batch`. So I think it is worth to upgrade sqlite3 to 1.4.0 to use that method. Before: ``` % ARCONN=sqlite3 bundle exec ruby -w -Itest test/cases/associations/eager_test.rb -n test_eager_loading_too_may_ids Using sqlite3 Run options: -n test_eager_loading_too_may_ids --seed 35790 # Running: . Finished in 202.437406s, 0.0049 runs/s, 0.0049 assertions/s. 1 runs, 1 assertions, 0 failures, 0 errors, 0 skips ARCONN=sqlite3 bundle exec ruby -w -Itest -n test_eager_loading_too_may_ids 142.57s user 60.83s system 98% cpu 3:27.08 total ``` After: ``` % ARCONN=sqlite3 bundle exec ruby -w -Itest test/cases/associations/eager_test.rb -n test_eager_loading_too_may_ids Using sqlite3 Run options: -n test_eager_loading_too_may_ids --seed 16649 # Running: . Finished in 8.471032s, 0.1180 runs/s, 0.1180 assertions/s. 1 runs, 1 assertions, 0 failures, 0 errors, 0 skips ARCONN=sqlite3 bundle exec ruby -w -Itest -n test_eager_loading_too_may_ids 10.71s user 1.36s system 95% cpu 12.672 total ```
* Merge pull request #35795 from alimi/cache-database-versionEileen M. Uchitelle2019-04-0310-56/+72
|\ | | | | Cache database version in schema cache
| * Cache database version in schema cacheAli Ibrahim2019-04-0310-56/+72
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * The database version will get cached in the schema cache file during the schema cache dump. When the database version check happens, the version will be pulled from the schema cache and thus avoid querying the database for the version. * If the schema cache file doesn't exist, we'll query the database for the version and cache it on the schema cache object. * To facilitate this change, all connection adapters now implement #get_database_version and #database_version. #database_version returns the value from the schema cache. * To take advantage of the cached database version, the database version check will now happen after the schema cache is set on the connection in the connection pool.
* | `bytes_margin` is only needed when `previous_packet` existsRyuta Kamizono2019-04-031-5/+2
| |
* | Use official database name [ci skip]Ryuta Kamizono2019-04-031-1/+1
|/ | | | | | | | * s/Postgres/PostgreSQL/ * s/MYSQL/MySQL/, s/Mysql/MySQL/ * s/Sqlite/SQLite/ Replaced all newly added them after 6089b31.
* Extract `sanitize_as_sql_comment` from SQL visitor into connectionRyuta Kamizono2019-03-191-0/+4
| | | | Probably that is useful for any other feature as well.
* Don't expose internal `type_casted_binds` methodRyuta Kamizono2019-03-191-7/+7
| | | | Internal usage for the method as public has removed at #29623.
* Tweak `truncate_tables`Ryuta Kamizono2019-03-181-17/+17
| | | | | | * Remove redundant `table_names.empty?` * Early return in `truncate_tables` since it is already deeply nested * Move `truncate_tables` out from between `exec_delete` and `exec_update`
* SQLite3: Make fixture loading to bulk statementsRyuta Kamizono2019-03-174-47/+38
|
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-27 07:29:02 +0000