• Skip cache when only_need_prompt is set
• Update only_need_context condition logic
• Prevent cache bypass in prompt-only mode

• Remove get_conversation_turns utility
• Pass history_messages to LLM directly
• Clean up prompt template formatting

- Replace json.dumps with line-by-line format
- Apply to entities, relations, text units
- Update truncation key functions
- Maintain ensure_ascii=False setting
- Improve context readability

- Delete history section from prompt
- Simplify user query response format
- Remove {history} placeholder variable

• Clarify max_tokens vs max_completion_tokens
• Add Gemini exception note
• Update parameter descriptions
• Add new completion tokens option

…parison

- Implement description length comparison in gleaning merge logic (extract_entities)
- Apply same logic to knowledge graph reconstruction (_rebuild_knowledge_from_chunks)
- Prioritize entities/relationships with longer descriptions for better quality
- Use list() instead of extend() for performance optimization when replacing

* Support `( and `（ start patterns
* Support )` and ）` end patterns
* Graceful fallback to warning logs
* Strip 2 chars for backtick variants
* Maintain existing bracket logic

- Expand bracket tolerance to support additional characters: < > " '
- Implement symmetric handling for both leading and trailing characters
- Replace simple string matching with robust regex-based pattern detection
- Maintain full backward compatibility with existing bracket formats

…LLM is used

…elimiter to <|SEP|>

• Add robust delimiter fixing logic
• Update prompts for single-line format

• Consolidate 6 regex patterns into 3
• More efficient pattern matching
• Clearer comments and examples
• Same functionality, less code
• Better maintainability

- Extract regex fixes to utils module
- Add case-insensitive delimiter handling

- Track timestamps for cache hits/misses
- Add timestamp to entity/relationship objects
- Sort descriptions by timestamp order
- Preserve temporal ordering in merges

…pserts

…port both format for safty

…|#: -> <|#|> pattern

• Fix completion_delimiter reference bug
• Add case check before lowercase conversion
• Improve delimiter corruption handling
• Optimize redundant processing logic

…ption docstring

…ption docstring

• Fix typo in error log message
• Clarify format requirements in prompts
• Make extraction instructions clearer
• Improve user prompt consistency

• Fix typo in error log message
• Clarify format requirements in prompts
• Make extraction instructions clearer
• Improve user prompt consistency

- Add debug logs for malformed records
- Fix regex for consecutive delimiters
- Handle missing closing brackets

* Remove trailing whitespace
* Fix function signature ellipsis style

…tent only

- Add debug logs for malformed records
- Fix regex for consecutive delimiters
- Handle missing closing brackets

…tent only

- Add QueryDataResponse model
- Implement /query/data endpoint
- Add aquery_data method to LightRAG
- Return entities, relationships, chunks

- Remove processing_info generation from _convert_to_user_format function
- Move all metadata generation (keywords, processing_info) to kg_query and naive_query functions
- Simplify _convert_to_user_format to focus only on data format conversion

…preserved in aquery_data responses