TODO.md

TODO

Index Existing Photos

• test tree walking producer test
• test filters
• write hash
• write db schema
• store files in db during index
  • path & hash
  • Detect conflicts / dupes
  • Convert to sqlx
  • Print list of duplicate files
  • File date
  • exif date
  • mp4 date
  • mov date
• Handle dupes (VERY careful deletion..) - which one to delete? There's one that's not in the DB. Should we assume that whatever we are indexing is somewhat reasonably laid out? - Should we add '--delete-dupes' and '--move-existing' config options for 'index' command? - [ ] Convert to e.g. Cobra
• TBD should we move existing files to new standard naming convention?

Import New Directories

• Walk dir, generate hashes
• Detect & eliminate duplicates
• Calculate import path
• Carefully move (handle existing files carefully, do not overwrite, make that atomic if possible)

Thumbnail generation

• Generate a thumb on demand
• Pre-generate thumbnails for any photos that don't have one.

Web view

• Web server with hierarchical year/month view - Based on date from DB or from filesystem?

Advanced Dupe Detection

• File date or Exif date is identical - build a web view with possible dupes and the ability to delete one?
• Use a bloom filter for detecting dupes (instead of hammering the DB index.)