- Introduced a `ProcessingMode` enum to differentiate between basic and premium processing modes.
- Updated `EtlRequest` to include a `processing_mode` field, defaulting to basic.
- Enhanced ETL pipeline services to utilize the selected processing mode for Azure Document Intelligence and LlamaCloud parsing.
- Modified various routes and services to handle processing mode, affecting document upload and indexing tasks.
- Improved error handling and logging to include processing mode details.
- Added tests to validate processing mode functionality and its impact on ETL operations.
- Fix timestamp conversion: String(epochMs) → new Date(epochMs).toISOString()
in use-messages-sync, use-comments-sync, use-documents, use-inbox.
Without this, date comparisons (isEdited, cutoff filters) would fail.
- Fix updated_at: undefined → null in use-inbox to match InboxItem type
- Fix ZeroProvider: skip Zero connection for unauthenticated users
- Clean 30+ stale "Electric SQL" comments in backend Python code
- Increased maximum file upload limit from 10 to 50 to improve user experience.
- Implemented batch processing for document uploads to avoid proxy timeouts, splitting files into manageable chunks.
- Enhanced garbage collection in chat streaming functions to prevent memory leaks and improve performance.
- Added memory delta tracking in system snapshots for better monitoring of resource usage.
- Updated LLM router and service configurations to prevent unbounded internal accumulation and improve efficiency.
- 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.
- Added session refresh for notifications to prevent stale data after rollbacks in multiple document processing tasks.
- Wrapped notification update logic in try-except blocks to handle potential failures gracefully and log errors without crashing the process.
- Improved error handling for notification updates in various document processing functions, enhancing overall robustness.
- Introduced a new DocumentProcessingNotificationHandler to manage notifications for document processing stages.
- Updated existing notification methods to include detailed progress updates for various stages (queued, parsing, chunking, embedding, storing, completed, failed).
- Refactored NotificationService to support the new document processing notification type and metadata schema.
- Updated multiple document processing tasks to create and manage notifications throughout the processing lifecycle.
- Adjusted UI components to reflect changes in notification types and improve user experience during document uploads and processing.
- Added streaming service support to the Q&A agent for real-time token streaming.
- Updated `answer_question` method to stream responses token-by-token to the frontend.
- Modified `handle_qna_workflow` to handle both custom and values streaming modes.
- Enhanced state management to include streaming service for improved user experience.
