- Add batch progress reporting. Thanks to @gee-forr via ##729.
- Use correct method for clearing query cache. Thanks to @EtienneDepaulis via ##719.
- Use @@max_allowed_packet session variable instead of querying SHOW VARIABLES. Thanks to @diclophis via #706.
- Add option :track_validation_failures. When this is set to true, failed_instances will be an array of arrays, with each inner array having the form [:index_in_dataset, :object_with_errors]. Thanks to @rorymckinley via #684.
- Prevent mass-assignment errors in Rails strict mode. Thanks to @diclophis via ##709.
- Handle after_initialize callbacks. Thanks to @AhMohsen46 via #691 and #692.
- Fix regression introduced in 1.0.4. Explicitly allow adapters to support on duplicate key update. Thanks to @dsobiera, @jkowens via #696.
- Allow serialized attributes to be returned from import. Thanks to @timanovsky, @jkowens via #660.
- Return ActiveRecord::Connection from ActiveRecord::Base#establish_connection. Thanks to @reverentF via #663.
- Support PostgreSQL array. Thanks to @ujihisa via #669.
- Skip loading association ids when column changed. Thanks to @Aristat via #673.
- Use prepend pattern for ActiveRecord::Base#establish_connection patching. Thanks to @dombesz via #648.
- Fix NoMethodError when using PostgreSQL ENUM types. Thanks to @sebcoetzee via #651.
- Fix issue updating optimistic lock in Postgres. Thanks to @timanovsky via #656.
- Add support for ActiveRecord 6.1.0.alpha. Thanks to @imtayadeway via #642.
- Return an empty array for results instead of nil when importing empty array. Thanks to @gyfis via #636.
- Add support for CockroachDB adapter. Thanks to @willie via #605.
- Add support for ActiveRecord 6.0.0.rc1. Thanks to @madeindjs, @bill-filler, @jkowens via #619, #623.
- Fixes NoMethodError when attempting to use nil logger. Thanks to @MattMecel, @khiav22357.
- Fix issue validating STI models. Thanks to @thejbsmith, @jkowens via #626.
- Raise an error with a helpful message if array of values exceeds the number of columns for a table. Thanks to @golddranks via #589.
- Properly check if model responds to import before creating alias. Thanks to @jcw- via #591.
- No longer pass :returning option to child associations on recursive import. Thanks to @dmitriy-kiriyenko via #595.
- Fix import issue for models with Postgresql json/jsonb fields. Thanks to @stokarenko via #594.
- Fix issue importing models with timestamps that contain timezone information. Thanks to @dekaikiwi, @jkowens via #598.
- Ignore :no_returning when using :recursive option. Thanks to @dgollahon, @jkowens via #599.
- Move ActiveRecord::Dirty changes to previous_changes after import. Thanks to @stokarenko via #584.
- Previously :on_duplicate_key_update was enabled by default for MySQL. The update timestamp columns (updated_at, updated_on) would be updated on duplicate key. This was behavior is inconsistent with the other database adapters and could also be considered surprising. Going forward it must be explicitly enabled. See #548.
- Fix issue where validations where not working in certain scenarios. Thanks to @CASIXx1 via #579.
- Fix issue where ActiveRecord presence validations were being mutated. Limited custom presence validation to bulk imports.
- Allow updated timestamps to be manually set.Thanks to @Rob117, @jkowens via #570.
- Fix validating presence of belongs_to associations. Existence of the parent record is not validated, but the foreign key field cannot be empty. Thanks to @Rob117, @jkowens via #575.
- Add "secret" option validate_uniqueness to enable uniqueness validators when validating import. This is not a recommended approach (See #228), but is being added back in for projects that depended on this feature. Thanks to @jkowens via #554.
- Add on_duplicate_key_update for SQLite. Thanks to @jkowens via #542.
- Add option to update all fields on_duplicate_key_update. Thanks to @aimerald, @jkowens via #543.
- Handle deeply frozen options hashes. Thanks to @jturkel via #546.
- Switch from FactoryGirl to FactoryBot. Thanks to @koic via #547.
- Allow import to work with ProxySQL. Thanks to @GregFarrell via #550.
- Add support for makara_postgis adapter. Thanks to @chadwilken via #527.
- Skip validating presence of belongs_to associations. Thanks to @Sohair63, @naiyt, @jkowens via #528.
- Add missing require for ActiveSupport.on_load. Thanks to @onk via #529.
- Support setting attribute values in before_validation callbacks. Thanks to @SirRawlins, @jkowens via #531.
- Ignore virtual columns. Thanks to @dbourguignon, @jkowens via #530.
- Use the association primary key when importing. Thanks to @dpogue via #512.
- Allow association ids to be updated. Thanks to @Aristat via #515.
- Rename
importmethod tobulk_importand alias toimport. Thanks to @itay-grudev, @jkowens via #498. - Increment lock_version on duplicate key update. Thanks to @aimerald via #500.
- Fix import_without_validations_or_callbacks exception if array is empty. Thanks to @doloopwhile via #508.
- Add support for importing hashes thru a has many association. Thanks to @jkowens via #483.
- Fix validation logic for recursive import. For those on Rails 5.0 and 5.1,
this change requires models with polymorphic associations to specify the
inverse_ofargument (See issue #495). Thanks to @eric-simonton-sama, @jkowens via #489.
- Allow SQL subqueries (objects that respond to .to_sql) to be passed as values. Thanks to @jalada, @jkowens via #471
- Raise an ArgumentError when importing an array of hashes if any of the hash objects have different keys. Thanks to @mbell697 via #465.
- Fix issue loading incorrect foreign key value when syncing belongs_to associations with custom foreign key columns. Thanks to @marcgreenstock, @jkowens via #470.
- Fix issue importing models with polymorphic belongs_to associations. Thanks to @zorab47, @jkowens via #476.
- Fix issue importing STI models with ActiveRecord 4.0. Thanks to @kazuki-st, @jkowens via #478.
- Unscope model when synchronizing with database. Thanks to @indigoviolet via #455.
- Prevent :on_duplicate_key_update args from being modified. Thanks to @joshuamcginnis, @jkowens via #451.
- Allow returning columns to be specified for PostgreSQL. Thanks to @tjwp via #433.
- Fixes an issue when bypassing uniqueness validators. Thanks to @vmaxv via #444.
- For AR < 4.2, prevent type casting for binary columns on Postgresql. Thanks to @mwalsher via #446.
- Fix issue logging class name on import. Thanks to @sophylee, @jkowens via #447.
- Copy belongs_to association id to foreign key column before importing. Thanks to @jkowens via #448.
- Reset model instance on validate. Thanks to @vmaxv via #449.
- Fix a regression where models weren't properly being marked clean. Thanks to @tjwp via #434.
- Raise ActiveRecord::Import::ValueSetTooLargeError when a record being inserted exceeds the
max_allowed_packetfor MySQL. Thanks to @saizai, @jkowens via #437. - Fix issue concatenating column names array with primary key. Thanks to @keeguon via #440.
- For PostgreSQL, add option to set WHERE condition in conflict_action. Thanks to @Saidbek via #423.
- Fix issue importing saved records with serialized fields. Thanks to @Andreis13, @jkowens via #425.
- Fix issue importing records that have columns defined with default values that are functions or expressions. Thanks to @Andreis13, @jkowens via #428.
- Set models new_record attribute to false when importing with :on_duplicate_key_ignore. Thanks to @nijikon, @jkowens via #416.
- Enable custom validate callbacks when validating import. Thanks to @afn via #410.
- Prevent wrong IDs being set on models when using :on_duplicate_key_ignore. Thanks to @afn, @jkowens via #412.
- Fix to enable validation callbacks (before_validation, after_validation). Thanks to @sinsoku, @jkowens via #406.
- Uniqueness validation is bypassed when validating models since it cannot be guaranteed if there are duplicates in a batch. Thanks to @jkowens via #301.
- Allow for custom timestamp columns. Thanks to @mojidabckuu, @jkowens via #401.
- Fix ActiveRecord 5 issue coercing boolean values when serializing for the database. Thanks to @rjrobinson, @jkowens via #403.
- Fix issue where PostgreSQL cannot recognize columns if names include mixed case characters. Thanks to @hugobgranja via #379.
- Fix an issue for ActiveRecord 5 where serialized fields with default values were not being typecast. Thanks to @whistlerbrk, @jkowens via #386.
- Add option :force_single_insert for MySQL to make sure a single insert is attempted instead of performing multiple inserts based on max_allowed_packet. Thanks to @mtparet via #387.
- Along with setting id on models for adapters that support it, add created_at and updated_at timestamps. Thanks to @jacob-carlborg via #364.
- Properly set returned ids when using composite_primary_keys. Thanks to @guigs, @jkowens via #371.
- Add support for composite_primary_keys gem. Thanks to @jkowens via #350.
- Add support for importing an array of hashes. Thanks to @jkowens via #352.
- Add JDBC SQLite3 support. Thanks to @jkowens via #356.
- Remove support for SQLite recursive imports. See #351.
- Improve import speed for Rails 5. Thanks to @ranchodeluxe, @jkowens via #359.
- Fixes issue clearing query cache on wrong connection when using multiple databases. Thanks to @KentoMoriwaki via #337
- Raises an ArgumentError on incorrect usage of nested arrays. Thanks to @Nitrodist via #340
- Fixes issue that prevented uuid primary keys from being set manually. Thanks to @Dclusin-og, @jkowens via #342
- Fixes issue with missing error messages on failed instances when importing using arrays of columns and values. Thanks to @Fivell via #332
- Update so SQLite only return ids if table has a primary key field via #333
- Add partial index upsert support for PostgreSQL. Thanks to @luislew via #305
- Add UUID primary key support for PostgreSQL. Thanks to @jkowens via #312
- Add store accessor support for JSON, JSON, and HSTORE data types. Thanks to @jkowens via #322
- Log warning if database does not support :on_duplicate_key_update. Thanks to @jkowens via #324
- Add option :on_duplicate_key_ignore for MySQL and SQLite. Thanks to @jkowens via #326
- Fixes issue with recursive import using same primary key for all models. Thanks to @chopraanmol1 via #309
- Fixes issue importing from STI subclass with polymorphic associations. Thanks to @JNajera via #314
- Fixes issue setting returned IDs to wrong models when some fail validation. Also fixes issue with SQLite returning wrong IDs. Thanks to @mizukami234 via #315
- An ArgumentError is now raised if when no
conflict_targetorconflict_nameis provided or can be determined when using theon_duplicate_key_updateoption for PostgreSQL. Thanks to @jkowens via #290 - Support for Rails 5.0 final release for all except the JDBC driver which is not yet updated to support Rails 5.0
- activerecord-import no longer modifies a value array inside of the given values array when called with
import(columns, values). Thanks to @jkowens via #291
raise_erroris used to raise errors for ActiveRecord 5.0. Thanks to @couragecourag via #294raise_record_invalidhas been
- JRuby/JDBCDriver with PostgreSQL will no longer raise a JDBCDriver error when using the :no_returning boolean option. Thanks to @jkowens via #287
- Support for ActiveRecord 3.1 has been dropped. Thanks to @sferik via #254
- SQLite3 has learned the :recursive option. Thanks to @jkowens via #281
- :on_duplicate_key_ignore will be ignored when imports are being done with :recursive. Thanks to @jkowens via #268
- :activerecord-import learned how to tell PostgreSQL to return no data back from the import via the :no_returning boolean option. Thanks to @makaroni4 via #276
- Polymorphic associations will not import the :type column. Thanks to @seanlinsley via #282 and #283
- ~2X speed increase for importing models with validations. Thanks to @jkowens via #266
- Benchmark HTML report has been fixed. Thanks to @jkowens via #264
- seamless_database_pool has been updated to work with AR 5.0. Thanks to @jkowens via #280
- Code cleanup, removal of redundant condition checks. Thanks to @pavlik4k via #273
- Code cleanup, removal of deprecated
alias_method_chain. Thanks to @codeodor via #271
-
Addition of :batch_size option to control the number of rows to insert per INSERT statement. The default is the total number of records being inserted so there is a single INSERT statement. Thanks to @jkowens via #245
-
Addition
import!which will raise an exception if a validation occurs. It will fail fast. Thanks to @jkowens via #246
- Fixing issue with recursive import when utilizing the
:on_duplicate_key_updateoption. Theon_duplicate_key_updateonly applies to parent models at this time. Thanks to @yuri-karpovich for reporting and @jkowens for fixing via #249
- Refactoring of fetching and assigning attributes. Thanks to @jkownes via #259
- Lots of code cleanup and addition of Rubocop linter. Thanks to @sferik via #256 and #250
- Resolving errors with the test suite when running against ActiveRecord 4.0 and 4.1. Thanks to @jkowens via #262
- Cleaning up the TravisCI settings and packages. Thanks to @sferik via #258 and #251
- PostgreSQL UPSERT support has been added. Thanks @jkowens via #218
- has_one and has_many associations will now be recursively imported regardless of :autosave being set. Thanks @sferik, @jkowens via #243, #234
- Fixing an issue with enum column support for Rails > 4.1. Thanks @aquajach via #235
- Support for em-synchrony has been removed since it appears the project has been abandoned. Thanks @sferik, @zdennis via #239
- Support for the mysql gem/adapter has been removed since it has officially been abandoned. Use the mysql2 gem/adapter instead. Thanks @sferik, @zdennis via #239
- Cleaned up TravisCI output and removing deprecation warnings. Thanks @jkowens, @zdennis #242
Never look back. What's gone is now history. But in the process make memory of events to help you understand what will help you to make your dream a true story. Mistakes of the past are lessons, success of the past is inspiration. – Dr. Anil Kr Sinha