Skip to content

Releases: LibrePhotos/librephotos-docker

2021 Week 35

06 Sep 14:03
ec02d47
Compare
Choose a tag to compare

LibrePhotos Development: 2021 - Week 35

Upstream:

  • 🚀 Tagging docker images based on week
  • This week's docker image will be called "2021w35"
  • ✨ Fix issues found by flake8 linting (Implemented by @jmalm)
  • 🔨 Duplicated faces on rescan faces (Implemented by @rootkie)
  • 🔨 Fix pytorch dependency issue / Fix crash of backend
  • 🔨 Fix viewing autocreated albums
  • 🔨 Fix adding new places on a rescan
  • 🔨 Limit the number of similar images

New Issues:

  • 💡 Create AlbumThingView (Link)
  • 💡 Install dev tools to find memory leaks / performance issues (Link)
  • 💡 Implement faiss with low memory footprint (Link)
  • This was addressed with a commit, but it is not a good solution as only a part of all images can now be shown as a similar image.

New discussions:

  • Read/write metadata from/to disk (Link):

  • This PR will add the ability to write the favorite / rating data to a sidecar or the media file. This is currently in development and testing. If you want to help out or if you have input for this feature, hop onto our discord into dev channel and join the conversation.

2021 Week 34

31 Aug 21:28
ec02d47
Compare
Choose a tag to compare

Upstream:

🚀 When scanning, only scan new and updated photos
🚀 Provide 'full-scan' flag for photo scan cli (Implemented by @jayesh100)
🚀 Play videos only on hover
🚀 Add ARM images to reallibrephotos dockerhub
🚀 Build ARM images automatically with GitHub actions
✨ Add code formatting and linting (black, isort, flake8) (Implemented by @reneraab)
✨ Refactor photos in redux (Implemented by @jmalm)
✨ Removed dead code
🔨 Let users get photo details of their hidden photos
🔨 Fixed a frontend crash after adding a favorite and switching to the favorite view
🔨 Fixed scan starting from CLI
🔨 Fixed broken thumbnails in ModalPhotosShare view

Local data models

08 Jun 21:58
Compare
Choose a tag to compare

This version switches from s3 to GitHub to save the data_models