- Fix static file serving paths to use correct storage directory
- Remove /api prefix from admin photo URLs to prevent double /api/api/ issue
- Fix thumbnail URL generation in gallery to use correct path format
- Update storage path resolution to support both relative and absolute paths
The issues were:
1. Admin images had URLs like /api/api/admin/events/2/thumbnail/90
2. Gallery thumbnails were looking for /thumbnails/thumb_*.jpg but paths were wrong
3. Static serving middleware was using incorrect storage paths
All images and thumbnails should now load correctly in both admin and gallery views.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Backend changes:
- Update multer configuration to accept up to 500 files per upload
- Implement batch processing (10 files per transaction) for better performance
- Add memory-efficient Sharp configuration for thumbnail generation
- Increase Express body parser limits to handle large payloads
- Add proper error handling and reporting for partial upload failures
Frontend changes:
- Update validation to allow 500 files maximum
- Implement chunked uploads (50 files per chunk) to prevent timeouts
- Add progress tracking with chunk information display
- Update error messages and translations (EN/DE)
Performance optimizations:
- Disable Sharp cache to prevent memory buildup
- Limit Sharp concurrency to 2 operations
- Use sequential read for large images
- Process files in database transaction batches
- Return detailed upload results including success/failure counts
This implementation ensures the application can handle large photo uploads
efficiently without running into memory or timeout issues.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add comprehensive tests for formatBoolean() function
- Test PostgreSQL vs SQLite boolean handling
- Test date formatting and manipulation functions
- Verify insertAndGetId handles both database result formats
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add formatBoolean() usage to all boolean database queries
- Fix 36 boolean comparisons across 18 backend files
- Ensure compatibility between SQLite (0/1) and PostgreSQL (true/false)
- Update all WHERE clauses and UPDATE operations with boolean values
This completes the database compatibility fixes for production deployment.
All boolean operations now work correctly with both database systems.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add 20-file limit validation to PhotoUpload component
- Prevent Multer "Unexpected field" errors by enforcing client-side limit
- Fix JSON parsing error in dateFormatter when value is already an object
- Add missing translation keys for upload error messages
- Handle both string and object values for date format settings
These fixes resolve the 400 error when uploading more than 20 files
and the "Unexpected token o in JSON" error during email queue creation.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixed boolean value handling differences between databases:
- SQLite stores booleans as 0/1, PostgreSQL as true/false
- Add formatBoolean() calls to critical queries that were failing
Files fixed:
- adminEvents.js: Fixed status filters and archive queries
- adminDashboard.js: Fixed active/archived event counts
- expirationChecker.js: Fixed expiration checking queries
- dbCompat.js: Updated to avoid circular dependency
Added migration 024 to:
- Enable foreign keys for SQLite (PRAGMA foreign_keys = ON)
- Document boolean compatibility requirements
This fixes queries returning 0 results in SQLite when checking
boolean columns like is_active, is_archived. Critical for
proper event management and expiration handling.
Note: 23 more boolean comparisons remain to be fixed in other files.
Critical fixes for database compatibility issues:
INSERT operations:
- Fix all INSERT queries to use .returning('id')
- Handle both PostgreSQL (returns objects) and SQLite (returns IDs)
- Fixed in: events.js, adminArchives.js, adminEvents-enhanced.js, create-test-event.js
Date operations:
- Replace SQLite-specific db.raw("datetime('now', '+30 days')")
- Use JavaScript Date objects for cross-database compatibility
- Fixed in: adminArchives.js
Database utilities:
- Add dbCompat.js utility for handling database differences
- Provides consistent API for inserts, dates, booleans, and DB-specific operations
- Centralized database compatibility logic
Migration:
- Add migration 023 documenting PostgreSQL compatibility requirements
- Ensures future developers are aware of compatibility needs
This resolves all 'not iterable' errors and ensures the application
works correctly with both PostgreSQL (production) and SQLite (development).
- Fix event creation failing with 'not iterable' error
- Add .returning('id') to insert queries for PostgreSQL
- Handle both PostgreSQL (returns objects) and SQLite (returns IDs)
- Add missing fields to email_queue insert
This fixes the 500 error when creating new events in production.
- Remove deploy/ folder with complex Docker Swarm configs
- Remove local development scripts referencing non-existent docker-compose.local.yml
- Remove setup scripts for files that already exist
- Remove docker-compose.dev.yml in favor of unified approach
- Keep certbot/, CHANGELOG.md, docker-compose.prod.yml, and production guide
This simplifies the repository structure and removes confusing
duplicate deployment approaches, making it clearer for new users.
- Add bug report template with environment details
- Add feature request template with use case section
- Add documentation improvement template
- Add security vulnerability template with warnings
- Add question template for support
- Add issue template config with helpful links
- Add comprehensive pull request template
- Update mirror workflow to exclude production guide
These templates will help maintain quality contributions
and streamline the issue/PR process on GitHub.
- Create comprehensive README.md optimized for GitHub/SEO
- Consolidate deployment instructions into single DEPLOYMENT.md
- Add all standard GitHub documentation files:
- CONTRIBUTING.md with development guidelines
- CODE_OF_CONDUCT.md for community standards
- SECURITY.md with vulnerability reporting
- CHANGELOG.md following Keep a Changelog format
- Simplify deployment with single docker-compose.yml
- Remove complex deployment configurations (Swarm, Traefik)
- Add backup script for easy maintenance
- Update .github-mirror-exclude to hide complex configs
- Remove redundant documentation files
This prepares PicPeak as a professional open-source alternative
to PicDrop and Scrapbook.de with clear, simple deployment.
- Add three different approaches for mirroring to GitHub
- Approach 1: Filter out sensitive files on a separate branch
- Approach 2: Use git archive with .gitattributes exclusions
- Approach 3: Use rsync for flexible file filtering
- Add exclusion lists for sensitive files and directories
- Protect production configs, environment files, and private data
This allows maintaining a public GitHub mirror while keeping
sensitive configuration and data private on the Gitea instance.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Reduce minimum password length from 12 to 8 characters
- Make special characters optional for gallery passwords
- Lower strength requirement from score 3 to 1 for galleries
- Add eye icon toggle for password visibility on each field
- Remove redundant 'Show passwords' checkbox
- Add translation for password security requirements error
- Update both English and German translations
This allows users to use simpler passwords like 'Sommer2025\!' for events
while maintaining security through other measures like expiration dates.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fix trust proxy to use specific values instead of permissive 'true'
- Fix clear old notifications to use database-agnostic date calculation
- Fix database size check to support both PostgreSQL and SQLite
- Add caching and better error handling for session timeout queries
- Add query timeout to prevent hanging connections
- Improve JSON parsing error handling for setting values
These fixes address:
- ERR_ERL_PERMISSIVE_TRUST_PROXY warning
- PostgreSQL datetime function errors
- ENOENT errors looking for SQLite file in PostgreSQL deployment
- Connection terminated errors for session timeout checks
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add detailed validation for SMTP configuration
- Add specific error messages for common SMTP issues (connection, auth, network)
- Add logging to help debug email configuration issues
- Check for masked passwords that shouldn't be used for sending
- Parse smtp_port as integer and handle boolean smtp_secure properly
This helps identify the exact cause of email sending failures instead of
generic 500 errors.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Set Express to trust proxy headers for proper IP detection with Traefik
- Add safe JSON parsing for email template variables and activity log metadata
- Create migration to fix invalid JSON data in database
- Add error handling to prevent JSON.parse crashes
This fixes the 500 errors caused by invalid JSON data and the trust proxy
warning from express-rate-limit when running behind Traefik.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fix email templates list endpoint to handle both schema variations
- Add migration for default CMS pages with language-specific columns
- Create debug script to identify 500 error causes
- Update CMS migration to use correct column names (title_en/de, content_en/de)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update migration to detect and handle both old and new email template schemas
- Fix migration to insert into correct columns based on existing schema
- Update adminEmail routes to handle both schema formats gracefully
- Add proper fallbacks for German language columns
This ensures the application works whether the language migration has been
applied or not, preventing null constraint violations.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add migration to fix email_templates column structure after language migration
- Add migration to ensure default email templates exist
- Create diagnostic script to check database issues
- Fix docker-compose configuration for proper routing without path stripping
The backend expects routes with /api prefix, so removing the stripprefix
middleware allows proper routing to work.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add robust connection pool configuration with timeouts and retry settings
- Implement retry logic in maintenance middleware for connection errors
- Increase connection stability with keepAlive and proper timeout values
- Handle "Connection terminated unexpectedly" errors gracefully
This prevents 503 errors when the database connection is temporarily interrupted
and ensures the application can recover from transient connection issues.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add missing created_at column to email_queue table
- Fix 502 Bad Gateway errors with proper Traefik routing configuration
- Create docker-compose.traefik.yml for external Traefik deployment
- Fix health check endpoint path for API path stripping
- Add PostgreSQL init script for Umami database creation
- Add comprehensive deployment guide for Traefik setup
The backend now properly handles /api prefix stripping by Traefik and
migrations run safely in production environments with existing schemas.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Major fixes for production deployment with Traefik:
1. API Path Fixes:
- Remove double /api prefix from all frontend service calls
- Fix auth.service.ts to use correct paths (/auth/admin/login)
- Update all services to use single /api prefix from base URL
- Fix template literal paths in photo services
2. Docker Configuration:
- Add build args for VITE_API_URL in docker-compose.prod.yml
- Create Dockerfile.prod with proper API URL configuration
- Ensure frontend is built with correct API base path
3. Documentation:
- Add comprehensive TRAEFIK_DEPLOYMENT.md guide
- Document proper Traefik labels and routing configuration
- Include troubleshooting steps for common issues
- Explain network configuration and SSL handling
This resolves:
- 502 Bad Gateway errors
- Double /api/api paths in requests
- Frontend unable to communicate with backend
- Login functionality not working
The frontend now correctly calls the backend API through Traefik's
routing, with all requests going to /api/* being forwarded to the
backend service on port 3000.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Change from bcryptjs to bcrypt to match installed dependency
- Fixes "Cannot find module 'bcryptjs'" error when creating admin user
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Major fixes for production deployment issues:
1. Migration System:
- Add safe migration runner that handles existing schema
- Create migration helper functions for idempotent operations
- Auto-detect existing tables and mark migrations as applied
- Handle "relation already exists" errors gracefully
2. Production Initialization:
- Create init-production.sh script for proper startup sequence
- Fix directory creation and permissions
- Add admin user creation from environment variables
- Ensure proper service initialization order
3. Documentation:
- Add comprehensive PRODUCTION_DEPLOYMENT_GUIDE.md
- Add MIGRATION_ERROR_FIX.md for immediate issue resolution
- Document all known production issues and solutions
- Include backup/restore procedures
4. Safety Improvements:
- Add migrate:safe npm script for production use
- Update wait-for-db.sh to use safe migrations in production
- Add proper error handling and logging
This resolves the "relation already exists" error and prevents similar
issues in future deployments. The safe migration system can handle both
fresh installations and existing databases.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Gitea Actions Bot
paul