Add WebSocket ConnectionManager to Handler:
- Add connManager field to Handler struct
- Update NewHandler to accept ConnectionManager parameter
- Update SetupRoutes to pass ConnectionManager through
- Import sync package with alias to avoid conflicts
This enables progress handlers to broadcast updates via WebSocket.
Add device token validation method for WebSocket authentication:
- Validates device auth tokens against database
- Returns device information for valid tokens
- Used by WebSocket handler for device authentication
This enables devices to authenticate WebSocket connections
using their bearer tokens.
Add ConnectionManager for real-time WebSocket communication:
- Message types for progress updates, annotations, conflicts
- Broadcast message structure with source device tracking
- Device connection tracking with user and device metadata
- Automatic broadcast loop with concurrent message delivery
- Connection management (add, remove, get by ID/user)
- Stale connection cleanup (2-minute timeout)
- Connection statistics by device type
- Background cleanup task runs every minute
This implements the core WebSocket infrastructure needed for
Week 9 of the Universal Sync Implementation Guide.
- Update device auth middleware to set actual device object
- Change from DeviceContext to database.Devices
- Fix RequirePermission to use database.Devices
- Ensures handlers can access full device information
- Required for KOReader sync handlers to function properly
- Create SyncProgress for bidirectional progress synchronization
- Create GetMetadata for book progress and annotation retrieval
- Create GetLibrary for user library sync
- Create SyncBookmarks for annotation management
- Support device matching by UUID, file path, or title/author
- Implement immediate and checkpoint sync modes
- Handle bookmarks, highlights, and notes synchronization
- Part of Phase 3 KOReader Integration implementation
- Add bulk_update_progress_from_koreader() function for batch processing
- Handles progress, annotations, and conflict detection
- Returns success/failure status for each book
- Supports device matching by UUID, file path, or title/author
- Implements automatic conflict detection for concurrent syncs
- Part of Phase 3 KOReader Integration implementation
- Add POST /api/libraries/:id/scan endpoint for admin library scanning
- Add GET /api/libraries/:id/media-items endpoint for library media items
- Move /api/libraries/types to public endpoint (no auth required)
- Update ScanEbooks handler to support library_id parameter
- Add internal/sync package with format detection
- FormatGroup types: reflowable, fixed_layout, comic_archive
- DetectFormatGroup() function based on mimetype and file extension
- MimeType mappings for common ebook formats
- Progress conversion engine with:
- ConvertProgress() between format groups
- Extract percentage from various progress formats
- PageToPercentage / PercentageToPage helpers
- CharacterToPercentage / PercentageToCharacter helpers
- MergeProgress() with 'max progress wins' strategy
- FormatProgressForDisplay() for UI rendering
- Add sqlc queries for format detection and progress updates
- BulkUpdateFormatGroups query for auto-format detection
- GetUniversalProgress query with all location references
- UpdateUniversalProgress query with device sync metadata
- ReadingHistory queries for session tracking
- Remove API_TESTING_SUMMARY.md, IMPLEMENTATION_SUMMARY.md
- Remove MIGRATION_PROGRESS.md, SECURITY_*.md files
- Clean up temporary documentation files from previous sessions
- Repository now focused on current working code
- ebook_scanner.go:293 now compiles successfully
- GetLibraryByFolder method available after database code regeneration
- Scanner service can properly find libraries by folder path
Fixes primary compilation error blocking build
- Fix CreateMediaItem ISBN field to use pgtype.Text wrapper
- Fix UpdateMediaItem to use correct Isbn field name
- Resolve type mismatch between request and database params
Resolves compilation errors in media item handlers
- Use c.Get("user").(database.Users) instead of c.Get("user_id").(string)
- Extract userUUID from user.ID.Bytes ([16]byte)
- Properly convert to pgtype.UUID for service layer
- Remove unnecessary uuid.Parse call
This fixes 500 Internal Server Error when creating libraries via API.
The JWT middleware sets user as database.Users struct, not string.
Related to Bruno Create Library request testing.
- Fix scoping issue with err variable in os.Stat check
- Properly check for non-existent vs inaccessible folders
- Use reassignment (=) instead of declaration (:=) since err already declared
- Add GET /api/media-items/search endpoint
- Try partial matching first (ILIKE with wildcards)
- Fallback to fuzzy search if no results found
- Return 404 with 'no results found' when no matches
- Limit results to 50 items by default
- Supports search across title, author, series, tags, contributors
- Respects library visibility settings per user
- Add SearchMediaItems query with ILIKE partial matching
- Add SearchMediaItemsFuzzy query with word_similarity()
- Use sqlc.narg() for named parameters (search_pattern, search_query)
- Rank results by relevance: title > author > series > tags
- Fuzzy threshold set to 0.3 for word_similarity
- Generated Go models with proper parameter types
- Add /bookshelf route as default page for logged-in users
- Update login and register handlers to redirect to /bookshelf
- Update homepage to auto-redirect to /bookshelf when logged in
- Preserve /dashboard route for backward compatibility
- Update test redirects to use /bookshelf
Changes:
- main.go: Add /bookshelf protected route
- auth.go: Change login/register redirects from /api/dashboard to /bookshelf (2 locations)
- edge_cases_test.go: Update test redirect to /bookshelf
- Maintains backward compatibility with existing /dashboard route
This makes the beautiful bookshelf the default landing page
for all authenticated users while keeping the old dashboard accessible.
- Fix scheduler.go log.Printf calls to convert pgtype.UUID to string before formatting
- Fix ebook.go fmt.Printf calls to convert pgtype.UUID to string before formatting
- Add missing Enabled field to rate limiter config in security test
- Prevents format string errors when logging library IDs
This resolves compilation errors where pgtype.UUID was being formatted
with %s which expects a string, not a UUID struct.
- Add TestMode, RateLimitEnabled, RequestsPerMinute to Config
- Add getEnvBool() and getEnvInt() helper functions
- Update rate limiter to support enabled/disabled state
- Pass test environment variables through docker-compose
- Configure rate limiter dynamically in main.go
This allows disabling rate limiting for integration testing while
maintaining security in production environments.
- ListLibraries now returns {"data": []} instead of []
- ListUsers now returns {"data": []} instead of []
- ListMediaItems now returns {"data": []} instead of []
This provides consistent response structure across all list endpoints
and aligns with integration test expectations.
- Fix type assertion panics in library.go (lines 58, 109, 237)
Changed from *database.Users to database.Users to match JWT middleware
- Fix ISBN type mismatch in ebook.go (lines 249, 308)
Changed from pgtype.Text to string to match database schema
- Fix ISBN type mismatch in ebook_scanner.go (line 421)
Changed from pgtype.Text to string to match database schema
These changes fix 500 errors in library creation and ebook operations.
- Increase ISBN column from VARCHAR(13) to VARCHAR(17) to support ISBN-13 with hyphens
- Add normalize_isbn() database function to automatically remove hyphens and spaces
- Create trigger to auto-normalize ISBNs on INSERT/UPDATE operations
- Update all Ebook and MediaItem queries to use ISBN normalization
- Add GetEbookLibraryID query to check for existing ebook libraries
- Add graceful error handling when no ebook library exists
- Return helpful error message: 'no ebook library found. Please create an ebook library first'
- Create comprehensive tests for ISBN normalization and library selection
- Add Bruno test files for various ISBN formats and error scenarios
- Update documentation with ISBN normalization details
- Update scanner to run asynchronously in background worker pool
- POST /api/scanner/scan now returns immediately with job ID (HTTP 202)
- Add GET /api/scanner/status/:jobId for checking scan job progress
- Integrate watch mode with library system for instant ebook detection
- Auto-start watch mode for all libraries on server startup
- Add endpoints for managing watch mode per library:
- POST /api/scanner/watch/start
- POST /api/scanner/watch/stop
- GET /api/scanner/watch/status
- Track which libraries are currently being watched
- Auto-start scheduler on server boot
- Add Worker service with configurable worker pool for async job processing
- Implement job queue with status tracking (pending, running, completed, failed, cancelled)
- Add Scheduler service for auto-scanning based on user scan settings
- Check scan settings every 5 minutes and schedule background scan jobs
- Support multiple libraries with individual scan frequencies (15-1440 minutes)
- Regenerate queries.sql.go with refresh token queries
- Update models.go with RefreshTokens type
- Update querier.go with new query methods
- Update db.go with generated code
- Test password complexity requirements
- Test account lockout mechanism
- Test rate limiting functionality
- Test JWT expiration (1 hour)
- Test refresh token expiration (7 days)
- Test password requirements list
- Verify transaction manager and error handler types
- All tests passing
- Implement strict password requirements:
- Minimum 8 characters
- At least one uppercase letter
- At least one lowercase letter
- At least one number
- At least one special character
- Add custom validator for Echo integration
- Add GetPasswordRequirements helper function
- Add ValidatePassword function for manual validation
- Add os package import for file system checks
- Validate that folder paths exist before adding to library
- Check folder accessibility to prevent invalid paths
- Return clear error messages for invalid folders
Improves user experience by catching path errors early
- Enforce maximum pagination limit of 1000 items per request
- Prevent negative offset values in pagination
- Apply limits to both /api/ebooks and /api/media-items endpoints
- Protect against DoS attacks from large limit values
Fixes security issue: No maximum pagination limit
- Trim whitespace from usernames and validate non-empty
- Normalize role values to lowercase for case-insensitive comparison
- Prevent registration with whitespace-only usernames
- Maintain backward compatibility with existing functionality
Fixes validation gap: Username whitespace handling
- Add rate limiter middleware (10 requests/minute per IP)
- Apply rate limiting to POST /api/auth/register and /api/auth/login
- Prevents brute force attacks and registration spam
- Automatic cleanup of old request records
Closes security issue: No rate limiting on auth endpoints
- Fix database authentication error by exposing actual database error messages
- Update error handling to follow pgx v5 standards with detailed error reporting
- Restore token environment variable management in Register User.bru for subsequent requests
- Enable proper debugging of database connection issues during user registration
The registration API now provides detailed error messages instead of generic 'failed to check existing users'
when database connection or authentication fails, making debugging easier.
This major update implements a complete user annotation system:
## 🎯 New Features
- User notes with position tracking for media items
- Text highlighting with customizable colors
- Highlight-note associations for detailed annotations
- Full CRUD API for both notes and highlights
- Backward compatibility with existing ebook endpoints
## 📊 Database Changes
- Add media_notes table (id, media_item_id, user_id, content, position, timestamps)
- Add media_highlights table (id, media_item_id, user_id, selection_text, start/end_position, color, optional note_id)
- Add foreign key relationships with CASCADE deletes
- Add proper indexes for performance
- Add database schema views for ebook backward compatibility
## 🔧 API Implementation
- Complete REST API endpoints for notes and highlights
- JWT authentication with proper middleware bypass
- Request validation with meaningful error responses
- UUID validation and type safety
- Support for hex color codes in highlights
## 🧪 Testing & Documentation
- Comprehensive test suite covering authentication scenarios
- Bruno API collection for manual testing
- Detailed testing guide with troubleshooting
- Updated documentation in README and TESTING.md
## 📁 Backward Compatibility
- Existing ebook endpoints continue working
- Database views maintain API contracts
- No breaking changes for existing integrations
The annotation system is now fully functional and ready for production use.
- Add complete CRUD API for media items notes (/api/media-items/:id/notes/*)
- Add complete CRUD API for media highlights (/api/media-items/:id/highlights/*)
- Add backward compatibility endpoints for ebooks (/api/ebooks/:id/notes/*, /api/ebooks/:id/highlights/*)
- Implement proper validation for request payloads and UUIDs
- Support hex color codes for highlights with default yellow (#ffff00)
- Support position tracking (page:offset or CFI formats)
- Support optional note association with highlights
- Add MediaNotes and MediaHighlights model structs with pgx v5 types
- Add EbookNotes and EbookHighlights for backward compatibility
- Add complete CRUD SQL queries for notes and highlights
- Add database connection pool function using pgx v5
- Generate sqlc code for new annotation functionality
john-okeefe