SchemaValidations extends ActiveRecord to automatically create validations by inspecting the database schema. This makes your models more DRY as you no longer need to duplicate NOT NULL, unique, numeric and varchar constraints on the model level.

State machine that allows dynamic transitions for business workflows

ActiveRecord has always stored the base class in polymorphic _type columns when using STI. This can have non-trivial performance implications in certain cases. This gem adds the 'store_base_sti_class' configuration option which controls whether ActiveRecord will store the base class or the actual class. Defaults to true for backwards compatibility.'

Sharding

Introspects your database string field maximum lengths and automatically defines length validations.

Rails way to harness the power of SQL "EXISTS" statement

Subscribe to changes on ActiveRecord models

Backported from Rails 5.0 to use with 4.x versions

Any_of allows to compute dynamic OR queries.

An Active Record extension for creating associations through lateral joins

Recursively traverse trees using a single SQL query

ActiveRecord Stand-alone migrations

It is the library to reconnect automatically when ActiveRecord is disconnected from MySQL.

This gem enables your model to query the database using the earthdistance extension. This should be much faster than using trigonometry functions over standart indexs.

Refresh ActiveRecord connection on each rack request.

Redirect ActiveRecord (Rails) reads to slave databases while ensuring all writes go to the master database.

Rails ActiveRecord connector for Google Spanner Database

This gem is only needed before Rails 7.1, where the trilogy adapter is built in. When you upgrade to a version after 7.1.a, replace this with `gem "trilogy"` to ensure that trilogy itself is available.

Dead simple snapshot versioning for ActiveRecord models and associations.

*Atomic archiving/unarchiving for ActiveRecord-based apps* We had the problem that acts_as_paranoid and similar plugins/gems always work on a record by record basis and made it very difficult to restore records atomically (or archive them, for that matter). Because the archive and unarchive methods are in transactions, and every archival record involved gets the same archive number upon archiving, you can easily restore or remove an entire set of records without having to worry about partial deletion or restoration. Additionally, other plugins generally screw with how destroy/delete work. We don't because we actually want to be able to destroy records.