Track unread messages in match chats and display count badge:
- Schema: Add user1LastReadAt/user2LastReadAt to Match model
- Backend: Calculate unreadCount in dashboard API
- Socket: Update lastReadAt when user joins match room
- Frontend: Display red badge with unread count on match avatar
Security improvements:
- Add random CUID slugs to Match model to prevent ID enumeration attacks
- Update all match URLs from /matches/:id to /matches/:slug
- Keep numeric IDs for internal Socket.IO operations only
Backend changes:
- Add slug field to matches table with unique index
- Update all match endpoints to use slug-based lookups (GET, PUT, DELETE)
- Add GET /api/matches/:slug/messages endpoint to fetch message history
- Include matchSlug in all Socket.IO notifications
Frontend changes:
- Update all match routes to use slug parameter
- Update MatchesPage to use slug for accept/reject/navigate operations
- Update MatchChatPage to fetch match data by slug and load message history
- Update RatePartnerPage to use slug parameter
- Add matchesAPI.getMatchMessages() function
Bug fixes:
- Fix MatchChatPage not loading message history from database on mount
- Messages now persist and display correctly when users reconnect
Security improvements:
- Add @default(cuid()) to Event.slug for auto-generated random slugs
- Prevent ID enumeration attacks (no more predictable slugs like "warsaw-dance-festival-2025")
- Event slugs now generated as secure cuid strings (e.g., "cmhz3lcgb00018vbn34v4phoi")
Test accounts:
- Add 3 test users to seed (john_dancer, sarah_swings, mike_blues)
- All users checked in to Warsaw Dance Festival 2025
- Pre-configured heats for testing matchmaking system
- Full profiles with WSDC IDs, social media, and locations
Seed improvements:
- Add bcryptjs for password hashing
- Add Prisma seed configuration to package.json
- Use worldsdcId for event upsert (instead of slug)
- Auto-generate event slugs via Prisma default
Documentation:
- Add test account credentials to SESSION_CONTEXT.md
- Document event slug security model
- Include sample heats for each test user
- Add 3 new database tables: divisions, competition_types, event_user_heats
- Add seed data for 6 divisions (NEW, NOV, INT, ADV, ALL, CHA) and 2 competition types (J&J, STR)
- Add API endpoints for divisions and competition types
- Add heats management endpoints in events route (POST/GET/DELETE)
- Implement unique constraint: cannot have same role in same division+competition type
- Add participant verification before allowing heats management
- Support heat numbers 1-9 with optional Leader/Follower role
Backend:
- Add event_checkin_tokens table with unique tokens per event
- Implement GET /api/events/:slug/details endpoint (on-demand token generation)
- Implement POST /api/events/checkin/:token endpoint (date validation only in production)
- Implement DELETE /api/events/:slug/leave endpoint
- Add comprehensive test suite for check-in endpoints
Frontend:
- Add EventDetailsPage with QR code display, participant list, and stats
- Add EventCheckinPage with success/error screens
- Add "Leave Event" button with confirmation modal to EventChatPage
- Install qrcode.react library for QR code generation
- Update routing and API client with new endpoints
Features:
- QR codes valid from (startDate-1d) to (endDate+1d)
- Development mode bypasses date validation for testing
- Automatic participant count tracking
- Duplicate check-in prevention
- Token reuse for same event (generated once, cached)
Replace sequential event IDs in URLs with unique alphanumeric slugs to prevent enumeration attacks. Event URLs now use format /events/{slug}/chat instead of /events/{id}/chat.
Backend changes:
- Add slug field (VARCHAR 50, unique) to Event model
- Create migration with auto-generated 12-char MD5-based slugs for existing events
- Update GET /api/events/:slug endpoint (changed from :id)
- Update GET /api/events/:slug/messages endpoint (changed from :eventId)
- Modify Socket.IO join_event_room to accept slug parameter
- Update send_event_message to use stored event context instead of passing eventId
Frontend changes:
- Update eventsAPI.getBySlug() method (changed from getById)
- Update eventsAPI.getMessages() to use slug parameter
- Change route from /events/:eventId/chat to /events/:slug/chat
- Update EventsPage to navigate using event.slug
- Update EventChatPage to fetch event data via slug and use slug in socket events
Security impact: Prevents attackers from discovering all events by iterating sequential IDs.
Backend:
- Add EventParticipant model to track user-event participation
- Create database migration for event_participants table
- Record participation when user joins event chat via Socket.IO
- Update GET /api/events to include isJoined flag for current user
- Sort events: joined events first, then by start date
- Add authenticate middleware to GET /api/events
Frontend:
- Replace mock events with real API data from backend
- Add loading and error states to EventsPage
- Display "Joined" badge on events user has joined
- Highlight joined events with colored border
- Show "Open chat" vs "Join chat" button text
- Auto-refresh events list when navigating back
When users join an event chat, this is now recorded in the database.
Joined events appear at the top of the list with visual indicators.
- Add country and city fields to User model
- Create database migration for location fields
- Add validation for country and city (max 100 characters)
- Create countries.js with complete list of 195 countries
- Add country dropdown select and city text input to profile page
- Include country and city in GET /api/users/me response
- Update profile form to support location data
Users can now select their country from a dropdown list of all
countries and enter their city name.
- Add YouTube, Instagram, Facebook, and TikTok URL fields to User model
- Create database migration for social media link columns
- Add custom validators to ensure URLs contain correct domains
- Update profile page with social media input fields
- Include social media URLs in GET /api/users/me response
- Add icons for each social platform in the UI
Users can now add links to their social media profiles. Each field
validates that the URL contains the appropriate domain (e.g.,
instagram.com for Instagram, youtube.com/youtu.be for YouTube).
Enhanced WSDC registration flow with auto-lookup and account validation:
Backend changes:
- Add accountExists flag to WSDC lookup endpoint
- Check database for existing users with WSDC ID
- Fix Prisma binary target for Alpine Linux Docker containers
Frontend changes:
- Auto-lookup WSDC data after entering 4+ digits (500ms debounce)
- Show live preview dropdown with dancer information
- Display warning if account with WSDC ID already exists
- Block registration and suggest login for existing accounts
- Improve UX with real-time validation feedback
- Add CheckCircle, XCircle, AlertCircle icons for visual feedback
This prevents duplicate WSDC ID registrations and provides immediate
feedback to users, improving the registration experience.
Tested with:
- ID 26111 (Vince Yap) - new account allowed
- ID 26997 (Radoslaw) - existing account blocked
Backend features:
- AWS SES email service with HTML templates
- Email verification with dual method (link + 6-digit PIN code)
- Password reset workflow with secure tokens
- WSDC API proxy for dancer lookup and auto-fill registration
- Extended User model with verification and WSDC fields
- Email verification middleware for protected routes
Frontend features:
- Two-step registration with WSDC ID lookup
- Password strength indicator component
- Email verification page with code input
- Password reset flow (request + reset pages)
- Verification banner for unverified users
- Updated authentication context and API service
Testing:
- 65 unit tests with 100% coverage of new features
- Tests for auth utils, email service, WSDC controller, and middleware
- Integration tests for full authentication flows
- Comprehensive mocking of AWS SES and external APIs
Database:
- Migration: add WSDC fields (firstName, lastName, wsdcId)
- Migration: add email verification fields (token, code, expiry)
- Migration: add password reset fields (token, expiry)
Documentation:
- Complete Phase 1.5 documentation
- Test suite documentation and best practices
- Updated session context with new features
Phase 1 - Step 2: PostgreSQL Setup
**Infrastructure:**
- Add PostgreSQL 15 Alpine container to docker-compose.yml
- Configure persistent volume for database data
- Update backend Dockerfile with OpenSSL for Prisma compatibility
**Database Schema (Prisma):**
- 6 tables: users, events, chat_rooms, messages, matches, ratings
- Foreign key relationships and cascading deletes
- Performance indexes on frequently queried columns
- Unique constraints for data integrity
**Prisma Setup:**
- Prisma Client for database queries
- Migration system with initial migration
- Seed script with 4 test events and chat rooms
- Database connection utility with singleton pattern
**API Implementation:**
- GET /api/events - List all events (with filtering and sorting)
- GET /api/events/:id - Get single event with relations
- Database connection test on server startup
- Graceful database disconnect on shutdown
**Seed Data:**
- Warsaw Dance Festival 2025
- Swing Camp Barcelona 2025
- Blues Week Herräng 2025
- Krakow Swing Connection 2025
**Testing:**
- Database connection verified ✅
- API endpoints returning data from PostgreSQL ✅
- Migrations applied successfully ✅
All systems operational 🚀
