v3.0.0

BREAKING CHANGES

• Require php 8.1.
• Use new php enum instead of the home grown versions.

v2.0.2

• Add support for utf-8 characters when parsing words.

v2.0.1

• Do not truncate input in Tokenizer::scan. Removed substr($input, 0, 256) rule as we're unsure where/why it's there and seems safe to remove.

v2.0.0

BREAKING CHANGES

• Require php >=7.4
• Uses php7 type hinting throughout with declare(strict_types=1);
• Uses "ruflin/elastica": "^7.0"

v1.0.0

• Initial stable version.

v0.3.2

• BUG :: Add minimum requirement check for word matching [a-zA-Z0-9]+ and ignore empty phrases.

v0.3.1

• BUG :: Fix invalid string casting on Numbr in ElasticaQueryBuilder.

v0.3.0

BREAKING CHANGES

• Update ElasticaQueryBuilder to use "ruflin/elastica": "~5.3".
• Require php >=7.1 in composer.json.
• Add php7 type hinting and use declare(strict_types=1);.

v0.2.1

• pull #9: Respect boolean operator preceding subquery.

v0.2.0

BREAKING CHANGES

• issue #7: Update ElasticaQueryBuilder to use 2.x queries/filters. Requires "ruflin/elastica": "~3.2".
• issue #6: Make TimeZone configurable on any builders that use date nodes.
• The Number class was renamed to Numbr to prevent issue with scalar type hints in php7.