- Introduced dynamic character budget calculation for document formatting based on model's context window.
- Updated `format_documents_for_context` to respect character limits and improve output quality.
- Added `max_input_tokens` parameter to various functions to facilitate context-aware processing.
- Enhanced error handling for context overflow in LLM router service.
- Introduced should_summarize parameter in TaskDispatcher and CeleryTaskDispatcher to control summary generation.
- Updated InlineTaskDispatcher to support the new parameter for document processing.
- Replaced direct embedding calls with a utility function across various components to streamline embedding logic.
- Added enable_summary flag to several models and routes to control summary generation behavior.
- Introduced a TaskDispatcher abstraction to decouple the upload endpoint from Celery, allowing for easier testing with synchronous implementations.
- Updated the create_documents_file_upload function to utilize the new dispatcher for task management.
- Removed direct Celery task imports from the upload function, enhancing modularity.
- Added integration tests for document upload, including page limit enforcement and file size restrictions.
cast(document_metadata["key"], String) generates CAST(col->'key' AS TEXT)
which preserves JSON string quotes (e.g. '"Fix login bug"'), causing
case-insensitive comparisons to never match.
Replace with .astext which generates the ->> operator (unquoted text
extraction), making issue lookups by title and identifier work correctly.
- Add rollback on exception to prevent dirty database state
- Fix block verification threshold from >=1 to >=80%
- Remove technical 'failed' language from user messages
- Add empty content validation with warning
- Add NotionKBSyncService for immediate KB updates after page changes
- Implement block ID verification to ensure content freshness
- Refactor duplicate block processing logic to shared utils
- Add user-friendly status messages
- Include debug logging for troubleshooting
