2.5.0

[2.5.0]

Added

• The new match_phrase query (#73)
• Ability to optionally set more parameters for a few queries (#68)
• Index aliases, useful for zero downtime indexing

Changed

• Connection configuration now can include API ID and key (#74)

Deprecated

• The elastic:create and elastic:delete are deprecated in favour of scout:index and scout:delete-index

v2.4.1

[2.4.1]

Fixed

• Use lazy instead of cursor when mapping lazy collections

v2.4.0

[2.4.0]

Added

• The scout:index and scout:delete commands are now implemented for Elasticsearch
• A DocumentAdapterInterface with an adapter that only deals with documents
• The Query String and Simple Query String syntax
• The ability to fake Elasticsearch responses, allowing integration tests with this package

Changed

• The IndexAdapterInterface now only focuses on indices

Deprecated

• If you rely on the old IndexAdapterInterface, use the DeprecatedElasticAdapterInterface instead of the IndexAdapterInterface to keep the functionality working as it previously was

v2.3.0

[2.3.0]

Added

• Support for term aggregations

v2.2.0

[2.2.0]

Added

• Wildcard syntax

Fixed

• Pagination now works with the default Laravel Scout paginate() method

v2.1.1

Fix engine implementation for Laravel Scout 9.1 (#44, #45)

v2.1.0

[2.1.0]

Added

• DisjunctionMax, Exists and Invert Query Syntaxes

Changed

• Updated Laravel Scout support to include v9

v2.0.0

[2.0.0]

Added

• Indices are now configured through a IndexConfiguration class and repository
• Both Match and MultiMatch queries may now specify a 'fuzziness', the default stays 'auto'
• Using the field() on the search builder you may now define specifically which field(s) should be retrieved for the documents
• New function score compound queries that can replace the default boolean compound query
• A static debug method to help you with the last executed query
• Laravel Scout's take() method can be used to set the max amount of results
• Text analysis (analyzers, tokenizers, filters, etc.) is now possible through index settings.
• Prepare (parts of) your data before letting Elasticsearch index it.

Changed

• Sorting now uses the default Scout orderBy() method
• The MultiMatch now accepts a fields array with the fields to search in
• The Engine now uses Elastic's bulk operations to speed up updates of models

Fixed

• Running scout:flush now actually deletes the contents of an index

v1.2.0

[1.2.0]

Added

• Added the newCompound() function to customize the compound query

Fixed

• Boosts may now completely be left out
• Console commands now use the configured connection
• Client used to boot with three connections instead of one

Changed

• BoolQueries can now also be nested

v1.1.0

[1.1.0]

Added

• Ability to query nested property types

Fixed

• Boosts are now null by default.