- Updated rowToMessage to include a parseJson utility for handling both parsed objects and JSON strings from raw SQL queries.
- Ensured that data fields such as data, assistantMeta, modelMeta, and stats are correctly parsed when necessary.
- Added documentation to clarify the handling of JSON columns in Drizzle ORM.
- Added detailed explanations and examples for cursor semantics in `api-types.md`, clarifying the exclusive nature of cursors in pagination.
- Updated `CursorPaginationParams` interface to emphasize the cursor's role as an exclusive boundary in responses.
- Refactored `BranchMessagesQueryParams` to extend `CursorPaginationParams`, aligning with the new pagination logic.
- Modified `MessageService` to implement cursor-based pagination semantics, ensuring accurate message retrieval and response structure.
- Enhanced documentation throughout to provide clearer guidance on pagination behavior and usage patterns.
- Introduced `OffsetPaginationParams`, `CursorPaginationParams`, and their corresponding response types to standardize pagination handling across the API.
- Updated existing API types and hooks to support both offset and cursor-based pagination, improving data fetching capabilities.
- Enhanced documentation with detailed usage examples for pagination, including request parameters and response structures, to aid developers in implementing pagination effectively.
- Refactored related components to utilize the new pagination types, ensuring consistency and clarity in data management.
- Changed logger context strings in ApiServer, MiddlewareEngine, IpcAdapter, MessageService, TestService, and TopicService to follow a consistent naming convention with the prefix 'DataApi:'.
- This refactor enhances clarity and organization in logging across the API services.
- Introduced a new method `runCustomMigrations` in `DbService` to execute custom SQL statements that Drizzle cannot manage, such as triggers and virtual tables.
- Updated `database-patterns.md` and `README.md` to document the handling of custom SQL and its importance in maintaining database integrity during migrations.
- Refactored `messageFts.ts` to define FTS5 virtual table and associated triggers as idempotent SQL statements for better migration management.
- Introduced a new SQL migration script to create the `__new_message` table with updated schema, including foreign key constraints and check constraints for `role` and `status`.
- Migrated existing data from the old `message` table to the new structure and renamed the table to `message`.
- Added indexes for improved query performance on `parent_id`, `topic_id`, and `trace_id`.
- Updated metadata to reflect the new migration version and breakpoints for debugging.
- Updated CreateMessageDto to include detailed behavior for the parentId field, allowing for auto-resolution based on topic state, explicit root creation, or attachment to a specified parent message.
- Refactored MessageService to implement the new parentId logic, ensuring proper validation and error handling for message creation based on the topic's current state and existing messages.
- Enhanced transaction safety and clarity in the message insertion process by resolving parentId before inserting new messages.
- Added optional fields `traceId` and `stats` to the `UpdateMessageDto` interface for enhanced message tracking and statistics.
- Updated `MessageService` to handle the new fields during message updates, ensuring they are correctly processed in the database.
- Enhanced duplicate message ID handling by updating parentId references and ensuring transaction safety.
- Implemented a smart selection logic for determining activeNodeId, prioritizing original active nodes and foldSelected messages.
- Updated documentation to reflect changes in duplicate handling and active node selection strategies.
- Implemented detailed preparation, execution, and validation phases for migrating chat topics and messages from Dexie to SQLite.
- Added robust logging and error handling to track migration progress and issues.
- Introduced data transformation strategies to convert old message structures into a new tree format, ensuring data integrity and consistency.
- Updated migration guide documentation to reflect changes in migrator registration and detailed comments for maintainability.
- Introduced type-safe access methods for shared cache in the Main process, including `getShared`, `setShared`, `hasShared`, and `deleteShared`.
- Implemented `getAllShared` for initializing new Renderer windows with the complete shared cache state.
- Updated IPC communication to support bidirectional synchronization of shared cache between Main and Renderer processes.
- Enhanced cache management with absolute timestamps for TTL, ensuring precise expiration handling across windows.
- Added ready state tracking in Renderer for improved synchronization feedback during initialization.
- Refactored related documentation to reflect new features and usage patterns for shared cache.
- Revised the README files for shared data and main data layers to improve clarity and structure.
- Consolidated documentation on shared data types and API types, removing the now-deleted `api-design-guidelines.md`.
- Streamlined directory structure descriptions and updated links to relevant documentation.
- Enhanced quick reference sections for better usability and understanding of the data architecture.
- Enhanced `getTree` method to utilize Common Table Expressions (CTEs) for fetching active paths and tree structures in a single query, reducing database load.
- Updated `getBranchMessages` to implement a similar optimization, allowing for efficient retrieval of message paths without loading all messages.
- Refactored `getPathToNode` to use a recursive CTE for fetching ancestors, addressing the N+1 query problem in deep message trees.
- Introduced transaction handling in `create` and `update` methods to ensure atomic operations and data integrity during message modifications.
- Introduced `ActiveNodeStrategy` type to define strategies for updating `activeNodeId` when a message is deleted.
- Updated `DeleteMessageResponse` to include `newActiveNodeId` for tracking changes to the active node after deletion.
- Modified the `DELETE` endpoint to accept `activeNodeStrategy` as a query parameter, allowing for flexible handling of active node updates.
- Enhanced the `delete` method in `MessageService` to implement the new strategies, ensuring consistent behavior during message deletions.
- Added sections on basic usage of foreign keys, self-referencing foreign keys, and circular foreign key references.
- Provided TypeScript code examples to illustrate best practices and avoid common pitfalls.
- Explained the rationale behind using soft references in SQLite for improved data integrity and simplified operations.
- Added `setAsActive` property to `CreateMessageDto` for controlling active node status in topics.
- Updated `messageTable` and `topicTable` schemas to include foreign key constraints and improved handling of active node references.
- Refactored message and topic service methods to utilize `.returning()` for better data retrieval after inserts and updates.
- Implemented hard delete functionality for messages and topics, replacing soft delete logic to ensure data integrity.
- Changed the default status for messages from 'success' to 'pending' to better reflect the message processing state. This aligns with recent updates to the MessageStatus type.
- Replaced `DataApiError` with `SerializedDataApiError` for improved error serialization and IPC transmission.
- Enhanced error response format with additional fields for better context and debugging.
- Updated error handling utilities to streamline error creation and retry logic.
- Removed the deprecated `errorCodes.ts` file and migrated relevant functionality to `apiErrors.ts`.
- Updated documentation to reflect changes in error handling practices and structures.
- Expanded the README with detailed database schema guidelines, including naming conventions for tables, columns, and export names.
- Introduced new column helper functions for UUID primary keys (v4 and v7) to streamline table definitions.
- Updated existing schemas (group, message, tag, topic) to utilize the new UUID primary key helpers for improved consistency and auto-generation.
- Added `siblingsGroupId` to `message` schema for better message organization.
- Introduced `activeNodeId` in `topic` schema to manage message tree structure.
- Updated `assistantMeta` and `modelMeta` fields to use specific types for improved type safety.
- Included `isNameManuallyEdited` and `sortOrder` in `topic` schema for enhanced topic management.
- Added a new entry in the migration journal for version tracking.
- Deleted batch and transaction related schemas, handlers, and IPC channels to streamline the Data API.
- Updated related type definitions and import paths to reflect the removal of batch and transaction functionalities.
- Simplified the API server and adapter logic by eliminating unused methods and handlers.
- Removed outdated API model and schema files to simplify the structure.
- Consolidated API types and schemas for better organization and clarity.
- Updated import paths across the codebase to reflect the new structure.
- Enhanced documentation in related README files to guide usage of the new API schema organization.
- Changed migration command from `yarn run migrations:generate` to `yarn run db:migrations:generate` for consistency across the project.
- Updated related documentation in `CLAUDE.md`, `migrations/README.md`, and `src/main/data/README.md` to reflect the new command.
- Added a notice in `migrations/README.md` regarding potential database structure changes before the alpha release.
- Add topicTable schema with group organization and pinning support
- Add messageTable schema with tree structure (adjacency list pattern)
- Add groupTable schema for organizing entities by type
- Add tagTable and entityTagTable schemas for tagging system
- Add FTS5 full-text search support for message content
- Update preferenceTable to use composite primary key (scope, key)
- Regenerate initial migration with all tables
Changes Summary
| Type | Files |
|-------------|---------------------------------------------------------------------|
| New schemas | topic.ts, message.ts, group.ts, tag.ts, entityTag.ts, messageFts.ts |
| Modified | preference.ts (index → composite PK) |
| Migration | Renamed 0000_solid_lord_hawal.sql → 0000_init.sql with all tables |
This is part of the data refactoring project - adding core user data table schemas.
- Renamed `MultiPreferencesResultType` to `PreferenceMultipleResultType` for clarity and consistency.
- Updated `getMultipleRaw` method in `PreferenceService` to reflect the new type name.
- Adjusted related components and services to utilize the updated type, ensuring consistent usage across the application.
- Renamed `Preference_GetMultiple` to `Preference_GetMultipleRaw` in IpcChannel for clarity.
- Introduced `MultiPreferencesResultType` to better map requested keys to their values or undefined in preferenceTypes.
- Updated `PreferenceService` to implement `getMultipleRaw` for synchronous access to multiple preferences.
- Adjusted related components and services to utilize the new method for fetching multiple preferences.
- Cleaned up imports and ensured consistent usage across the application.
- Add data-classify tools for data inventory extraction and code generation
- Include consolidated Chinese documentation (README.md)
- Update generated file path references
This temporary directory will be removed after V2 refactor is complete.
- Added an init method to DbService for database initialization, ensuring it is called before migrations.
- Updated the migrateDb and migrateSeed methods to check if the database is initialized, improving error handling.
- Called dbService.init() in the app's whenReady event to ensure proper database setup during startup.
- Introduced a new method to configure Write-Ahead Logging (WAL) mode for better concurrency during database operations.
- Ensured WAL mode is set only once, with error handling to fall back to default settings if configuration fails.
- Updated the migrateDb method to call the new configuration method on the first database operation.
- Added detailed JSDoc comments for better understanding of DbService methods and usage.
- Implemented error handling during database initialization and migration processes to ensure robustness.
- Introduced a method to check if the database is initialized before accessing it.
- Updated the migrateSeed method to throw errors on failure, improving error reporting.
- Deleted the DataRefactorMigrateService and associated HTML files, as they are no longer needed.
- Removed test components and files related to data refactor migration, streamlining the codebase.
- Updated configuration files to reflect the removal of the data refactor migration functionality.
- Replaced core types with shared types in migration files to ensure consistency across the application.
- Deleted obsolete core types file and updated imports in migrator and window components to reference the new shared types.
- Enhanced the migration process by streamlining type definitions and improving code maintainability.
- Added 'migration_completed' stage to the migration process for better tracking of completion.
- Updated relevant components and hooks to handle the new stage, including UI changes to confirm migration completion.
- Adjusted messages and progress indicators to reflect the new stage in the migration workflow.
- Integrated migration v2 functionality by importing necessary modules and registering IPC handlers.
- Updated the migration process to check for data migration needs and handle the migration window.
- Added new dependencies for stream-json and its types in package.json.
- Updated electron.vite.config.ts to include the new migration window HTML file.
- Added a garbage collection mechanism in CacheService to automatically remove expired cache entries every 10 minutes.
- Introduced a cleanup interval in PreferenceService, adjusting the frequency to every 5 minutes for better resource management.
- Enhanced cleanup methods in both services to ensure proper resource release during shutdown.
- Added detailed file overview comments to CacheService and DataApiService, clarifying their roles as infrastructure components rather than business services.
- Updated README.md to reflect the new structure and naming conventions, emphasizing the distinction between infrastructure and business logic components.
- Introduced a new TestService for API testing scenarios, providing mock data and various test cases.
- Created IBaseService interface to standardize service operations across the codebase.
- Improved organization of API handlers and services for better clarity and maintainability.
* style(linter): enable consistent-type-imports rule in typescript
* chore: add biome to lint script for improved code formatting
* chore: add oxlint-specific lint script for faster linting
* refactor: use type-only imports for better type safety and clarity
- Renamed "State Management" to "Data Management" in CLAUDE.md for clarity.
- Enhanced data management section with detailed descriptions of Cache System, Preference System, and User Data API.
- Updated README.md in shared/data to reflect new directory structure and provide clearer organization of type definitions and schemas.
- Added guidelines for selecting appropriate data access patterns based on data characteristics.
- Added "packages/ui/scripts/**" to .oxlintrc.json for linting.
- Excluded ".claude/**" from biome.jsonc.
- Refactored API path types in apiPaths.ts for better clarity.
- Updated error handling in errorCodes.ts to ensure stack trace is always available.
- Modified preferenceSchemas.ts to include new features and updated generated timestamp.
- Cleaned up tsconfig.json for better organization.
- Adjusted CustomTag component to improve rendering logic.
- Enhanced CodeEditor utility functions for better type safety.
- Improved Scrollbar story for better readability.
- Refactored CacheService to streamline comments and improve documentation.
- Updated useCache and useSharedCache hooks for better clarity and functionality.
- Cleaned up selectionStore and settings.ts by commenting out deprecated actions.
- Updated DataApiHookTests for better optimistic update handling.
- Moved CacheService functionality to a new implementation in @data/CacheService, enhancing modularity.
- Updated all references across the codebase to utilize the new cacheService instance.
- Removed the old CacheService files from both main and renderer directories to streamline the codebase.
- Eliminated AbortSignal handling from ApiClient interface and DataApiService to streamline request processing.
- Updated IpcAdapter to remove event sender logic for responses, aligning with the new direct IPC approach.
- Adjusted tests to reflect the removal of cancellation capabilities, emphasizing that direct IPC requests cannot be cancelled.
- Cleaned up related code and comments to enhance clarity and maintainability.
- Introduced new IPC channels for Data API requests and responses in IpcChannel.
- Added swr library to package.json for improved data fetching capabilities.
- Updated preload API to include Data API related methods for handling requests and subscriptions.
- Removed deprecated pending_default_values.ts file as part of data refactor.
- Updated preference keys from 'app.theme.*' to 'ui.*' for better organization and clarity.
- Introduced new preferences for assistant tab sorting and visibility.
- Refactored components to utilize the updated preference keys, improving consistency across the codebase.
- Cleaned up unused code and comments related to previous settings for maintainability.
- Updated auto-generated preference mappings to reflect the new structure.
- Updated preferences to utilize the new SendMessageShortcut type for better type safety.
- Refactored components to use the usePreference hook instead of useSettings for improved preference handling.
- Added new preferences for target language and send message shortcut in various components.
- Cleaned up unused code and comments related to settings for better maintainability.
- Updated auto-generated preference mappings to reflect recent changes in preference structure.
