Changelog

February 22, 2026 · View on GitHub

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[v1.3.0]

Changed

Go version updated to 1.26
MongoDB driver updated to v2
All module dependencies updated

[v1.2.0]

Enhanced

Improved ObjectID Detection - Now detects any field ending with _id (not just _id itself)
Smart ObjectID Conversion - Only attempts conversion for 24-character hex strings matching ^[0-9a-fA-F]{24}$
Graceful Fallback - Falls back to string search instead of erroring when ObjectID conversion fails
Removed Strict Validation - ID fields now support all query patterns (regex, wildcards, ranges, comparisons)

Changed

ID Field Detection - isIDField() now matches any field ending with _id (e.g., user_id, order_id, product_id)
ObjectID Conversion - convertToObjectID() now validates hex pattern before attempting conversion
Error Handling - Removed strict validation errors; ID fields now gracefully fallback to string behavior
Field Value Processing - fieldValueToBSONWithContext() no longer errors on invalid ObjectIDs

Features Implemented

user_id:507f1f77bcf86cd799439011 - Converts to {"user_id": ObjectID("507f1f77bcf86cd799439011")}
order_id:invalid - Falls back to {"order_id": "invalid"} (string search)
id:/pattern/ - Falls back to {"_id": {"$regex": "^pattern$"}} (regex search)
id:*pattern* - Falls back to wildcard pattern (wildcard search)
id:[start TO end] - Falls back to range pattern (range search)
id:>value - Falls back to comparison pattern (comparison search)

Technical Improvements

Enhanced isIDField() method to detect any *_id field
Removed validateIDFieldValue() method entirely
Updated convertToObjectID() with regex validation and graceful fallback
Modified fieldValueToBSONWithContext() to handle fallback scenarios
Updated all test cases to reflect new behavior

Documentation

Updated README.md with new ID field detection behavior
Added examples of *_id field detection and fallback scenarios
Documented removal of strict validation restrictions
Updated configuration documentation

Testing

Updated unit tests to expect success instead of errors for ID field patterns
Added test cases for *_id fields (user_id, order_id, product_id)
Updated integration tests to verify fallback behavior
Added comprehensive test coverage for new detection logic

[v1.1.0]

Added

ID field conversion - Automatic conversion of id field names to _id for MongoDB compatibility
ObjectID support - Automatic conversion of string values to primitive.ObjectID for _id fields
Nested ID field conversion - Support for nested ID fields (e.g., user.id → user._id)
ID field validation - Strict validation for _id fields with clear error messages
Configuration options for ID field handling:
- WithReplaceIDWithMongoID(bool) - Enable/disable id to _id conversion (default: true)
- WithAutoConvertIDToObjectID(bool) - Enable/disable ObjectID conversion (default: true)

Features Implemented

id:507f1f77bcf86cd799439011 - Converts to {"_id": ObjectID("507f1f77bcf86cd799439011")}
user.id:507f1f77bcf86cd799439011 - Converts to {"user._id": ObjectID("507f1f77bcf86cd799439011")}
id:invalid-hex - Returns validation error for invalid ObjectID hex strings
id:/pattern/ - Returns error for unsupported regex patterns on _id fields
id:*pattern* - Returns error for unsupported wildcard patterns on _id fields
id:[start TO end] - Returns error for unsupported range queries on _id fields
id:>value - Returns error for unsupported comparison operators on _id fields

Technical Improvements

Enhanced MongoFormatter with ID field conversion logic
Added convertFieldName() method for field name transformation
Added isIDField() method for ID field detection
Added validateIDFieldValue() method for strict validation
Added convertToObjectID() method for ObjectID conversion
Updated error handling to propagate validation errors through the parsing pipeline
Added comprehensive test coverage for all ID field conversion scenarios

Documentation

Updated README.md with ID field conversion examples and configuration
Added new "ID Field Conversion" section with usage examples
Documented configuration options and restrictions
Updated feature list to include ID field conversion

Testing

Added unit tests for ID field conversion in formatter/mongo/formatter_test.go
Added integration tests in tests/lucene-mongo/unit_test.go
Added MongoDB integration tests in tests/lucene-mongo/integration_test.go
All tests consolidated into existing test functions for better organization
Comprehensive error case testing for invalid ObjectID and unsupported patterns

[v0.10.0-beta.1]

Fixed

NOT operator with wildcards and regex - Fixed bug where NOT expressions with wildcards (e.g., NOT name:jo*) or regex patterns (e.g., NOT name:/^john.*/) were using $ne operator instead of $not. MongoDB's $ne operator cannot wrap query operators like $regex, $gt, $lt, etc. The fix now correctly uses $not for query operators and $ne for simple values.

[v0.9.0-beta.1]

Changed

BREAKING: Regex patterns now anchored by default - All regex queries (field:/pattern/) automatically add ^ and $ anchors for exact matching, consistent with Lucene behavior
BREAKING: Multiple unquoted words use OR logic - Free text queries with multiple words (e.g., john doe) now search each word separately with OR, rather than as a single phrase, consistent with Lucene behavior
BREAKING: Mixed queries default to OR - Field:value followed by free text without explicit operators (e.g., name:john admin) now uses OR instead of AND
Case sensitivity - Free text searches are case-insensitive with $options: "i", while regex patterns and wildcard patterns are case-sensitive

Migration

Update queries to reflect new behavior:

// Regex - for partial matches, use wildcards explicitly
"name:/john/"           // Before: partial match, After: exact match "^john$"
"name:/.*john.*/"       // Use this for partial matches

// Multiple words - use quotes for exact phrases
"john doe"              // Before: phrase match, After: john OR doe
"\"john doe\""          // Use quotes for exact phrase matching

// Mixed queries - use explicit AND for AND behavior
"name:john admin"       // Before: AND, After: OR
"name:john AND admin"   // Use explicit AND operator

[v0.8.0-beta.1]

Added

Default fields support for free text queries without requiring text indexes
ParseWithDefaults() function for specifying default fields per query
WithDefaultFields() config method for parser-level default field configuration
Wildcard and regex support in default field queries

Removed

MongoDB text search support - Removed $text operator functionality, replaced with default field searches for better Lucene compatibility and performance
WithEnableTextSearch() config method

Changed

Free text queries with default fields use case-insensitive regex by default
ParseWithDefaults takes priority over config-level default fields

[v0.7.0-beta.1]

Changed

BREAKING: Renamed BSON formatter to MongoDB formatter (config.FormatterBSON → config.FormatterMongo)
Moved formatter from formatter/bson/ to formatter/mongo/ for better clarity

Added

Enhanced developer documentation with architecture overview
Reorganized test structure with language-formatter specific test directories

Migration

Update your code to use the new formatter name:

// Old
cfg := config.Default().WithFormatter(config.FormatterBSON)

// New
cfg := config.Default().WithFormatter(config.FormatterMongo)

[v0.6.0-beta.1]

Added

Regex pattern support with Lucene-style syntax field:/regex/
MongoDB regex output with case-sensitive matching (like Lucene default)
Regex with logical operators for complex query combinations

Features Implemented

name:/john/ - Basic regex patterns
email:/.*@example\\.com/ - Complex regex with escaped characters
phone:/\\d{3}-\\d{3}-\\d{4}/ - Regex with digit matching
status:/^(active|pending|inactive)$/ - Regex with alternation
name:/john/ OR email:/.*@example\\.com/ - Regex with logical operators

Technical Improvements

Added Regex field to ParticipleValue struct for regex pattern recognition
Enhanced lexer with regex token pattern /([^/\\]|\\.)*/
Added tryParseRegex() and parseRegex() methods to formatter
Reordered parser priority to check regex before wildcard patterns
Added comprehensive test coverage for regex functionality

Documentation

Updated README with regex examples and improved readability
Streamlined documentation structure and removed redundancy
Added regex feature to features list and query syntax sections

[v0.5.0-beta.1]

Added

Participle integration for robust parsing and grammar handling
Developer documentation with comprehensive architecture overview

Changed

Refactored parsing engine to use Participle for lexical analysis and AST generation
Improved error handling with better syntax error messages from Participle