* Bunch of new warns due to a stronger clippy config
* Moving old processing stuff to a new namespace
Also fixing a bunch of typos through the entire codebase.
* Rustfmt
* First draft on image labeling
* Fixing execution providers for other OSs
* Better error handling and shutdown
* Working with shallow media processor
* bruh
* Fix warnings
* Now hooked to media processor job
* Link desktop app with libonnxruntime to avoid TLS error during startup
* Be able to change models on runtime
Revert to use labels table instead of tags
* A bug on a model-less inference
* Show AI labels on Inspector
- Change yolo inference to use half precision
- Add labels api to core
* Remove LD_PRELOAD
* Fix race condition on model executor shutdown
* Don't load all images in memory moron
* Embeed yolo model in prod build
- Change yolo model path to new one relative to executable
* Disable volume watcher on linux, it was crashing the app
- Invalidate labels when they are updated
* Rust fmt
* Minor changes
* Gate onnxruntime linking to the ai-models feature
* Add build script to sd-server to handle onnxruntime linking workaround
* Move AI stuff to its own crate and normalize deps
* Rust fmt
* Don't regenerate labels unless asked to
* Now blazingly fast
* Bad merge
* Fix
* Fix
* Add backend logic to download extra yolo models
* Add models api route
- Add api call to get available model version
- Add api call to change the model version
* Improve new model download logic
- Add frontend to change image labeler model
* Fix new model downloader
* Fix model select width
* invalidate labels count after media_processor generates a new output
* Rename AI crate and first draft on download notifications
* fix types
---------
Co-authored-by: Vítor Vasconcellos <vasconcellos.dev@gmail.com>
Co-authored-by: Brendan Allan <brendonovich@outlook.com>
* basic FDA start
* FDA checks and prompt
* lockfile
* misc clippy
* Some small warnings
---------
Co-authored-by: Ericson Fogo Soares <ericson.ds999@gmail.com>
* clippy -_-
* migrations
* alter the migration so it just renames the `dimensions` field (no db re-creation required)
* remove confusing semver addition for `libheif-sys`
* remove warning on the migration as it's just a rename
* add sort opts for resolution and date image was truly taken
* major serde ckeanup & add epoch_time and pixel_count
* rename, cleanup et optimise
* clippy
* ignore this mess
* bindings
* add explanation to schema
* comment out dt test
* better comment and WIP time
* cleanup rust
* failed timezone attempt
* remove image resolution as a sort by option
* update schema (and rename the dimensions table instead of dropping it)
* just show raw date
* add comments and update bindings
* fix migration hopefully
* fix broken migration
---------
Co-authored-by: Jamie Pine <32987599+jamiepine@users.noreply.github.com>
Co-authored-by: Jamie Pine <ijamespine@me.com>
Co-authored-by: Vítor Vasconcellos <vasconcellos.dev@gmail.com>
* derive `Type` on `*Extension`
* update `image`, `webp` and add `rspc::Error` support to the `sd-images` crate
* rename function for verbosity
* cleanup clippy allows and use scaling fn for svg conversions
* remove dead comment
* specify `generic` image extensions and add fn to enumerate all compatible extensions
* re-exports and inline hot provided method
* another inline on a provided method
* `formatter` -> `handler` and return an explicit error if the extension isn't supported, and add `convert_image()`
* `get_ext()` inlined helper fn
* add mutation for conversion and query for getting all supported extensions
* fix `svg` scaling and generation
* update bindings
* rm old const
* temporary test for all exts
* verbosity and no manual drop
* remove dbg test
* big enum and remove `cur` extension as `image` doesn't support it
* add huge extension enum
* hopefully get path from file path id and location id
* main merge fix
* chore: update toml and lockfile
* security: update webp
* comment entire crypto crate for build times
* add bincode, huge enum and custom ser/de
* Revert "comment entire crypto crate for build times"
This reverts commit 458addaad6.
* lockfile
* theoretically working ser/de
* add svg scaling
* fix pdf matching bug and use options
* move image scaling fn
* major consts overhaul with good ser/de
* disable heif checking for now due to new trait api
* make the generic handler truly generic
* fix pdf types
* fix pdf scaling so that it's 120dpi A4 paper size by default - height/aspect res are maintained
* heavy cleanup
* note to myself
* bindings & update core to use the scaling fn from the images crate (reducing code dupes)
* move vulnerable `aovec` to be a dev dep so it stays out of release builds (it's only used in tests)
* remoev exif crate from the core as it's handled in a sep. crate
* silence a clippy warning and update lockfile
* lower the maximum file size a slight amount
* bindings and minor dep bumps
* Some improvements
---------
Co-authored-by: Ericson Fogo Soares <ericson.ds999@gmail.com>
* bring in rand deps
* SUPER crude pluscode attempt
* working pluscodes!
* some cleanup
* `encodeURIComponent` for URL escaping
---------
Co-authored-by: Oscar Beaumont <oscar@otbeaumont.me>
* impl `Display` and a custom serde serializer
* remove the + as rust already shortens coords to 8SF
* remove manual serde impl and string conversion
* add `formatOrientation` to the `MediaData` part of the inspector
* make requested changes
---------
Co-authored-by: Jamie Pine <32987599+jamiepine@users.noreply.github.com>
* sd-images crate which will support raw/dng, bmp, etc
* more work on the image formatter
* re-work `sd-images`, add svg support, r/g/b and r/g/b/a HEIF image support (will all be async again soon)
* remove `ImageFormatter`, add note about r/g/b/(a) heif impl
* implement the image formatter
* rename the conversion trait and minor cleanups
* isolate heif feature and major cleanup
* very untested raw support
* change fn name to `from_path` (a lot more idiomatic)
* clean up orientation fixing
* heif is no longer forbidden (linux has good heif)
also all extensions are correctly matched in lowercase
* fix builds, ext matching, feature gating
* attempt to fix svg handling?
* raw attempt, quite a few errors
* add comment
* new (untested) attempt
* remove `raw` stuff for now
* replace `sd-svg` with a `ToImage` `SvgHandler` impl
* add some simple math to appropriately scale thumbnails (and bmp/ico support)
* add comments regarding how the math works for image thumbs
* rename the trait to `ImageHandler`
* basic layout
* lock
* add codec to image
* very messy wip
* rm that
* lock and toml
* working perfect exif extraction
* formatting
* migration and formatting
* mostly working
* fix
* set date created properly
* fix tsc
* working media data creation
* fix bad main merge? sorry brendan
* schema, migrations, bindings
* working exif data extraction
* why won't it work
* update migrations
* fix bad merge
* final cleanup
* cleanup migrations
* remove test (that was purely used for testing my code)
* working media data pulling, correct thumbnail orientation
* slightly optimise thumbnail rotation
* rename location to prevent specta clashes
* further improvements (location parsing is still broken)
* fix coordinate parsing i think
* rspc add some todos regarding final cleanup
* further thoughts
* major upgrades
* Some improved handling of errors and async calls
* accordion component
* heavily wip flash refactor
* fix builds all because of a (terrible) merge/move
* annoying missing newline
* i really hate exif
* remove dead code
* further flash progress :D
* docs(media-data): clarification
* minor cleanup
* cleanup and some async removal
* fix location parsing
* remove async (will do proper impl for async eventually) and major cleanup
* big W
* clippy and `FlashMode::Unknown` instead of `Invalid`
* add `NIKON` gps ref support
* comments about gps refs
* commit the submodule -_-
* major cleanup & composite image support
* remove old test image
* major cleanup and easier to use API
* remove old consts
* move `ExifReader` to dedicated module
* Media Data Extractor job and shallow job
* Extracting media data on watcher
* report no exif data on file gracefully
* cleanup errors and doctests
* Merging thumbnailer and media data extractor
* Job name and some strings in UI
* remove reliance on `sd-prisma` within the media data crate
* rename query to be more specific
* custom serializer for `MediaTime`
* tweak to format utc -> naive correctly
* generate migrations
* comment out duration in mobile
* delete test-assets folder
* all optional fields
* fix migrations again
* make requested name changes
* make further requested changes
* remove erroneous files from another wip branch
* updates procedures
* use strings where appropriate
* regen pnpm-lock
* add base layouts for video and audio media data
* use appropriate data types in schema and add audio media data framework
* make requested changes
* general cleanup and renaming of enum
* cleanup media data api
* rename media metadata type
* finishing touches
---------
Co-authored-by: Ericson Soares <ericson.ds999@gmail.com>
Co-authored-by: ameer2468 <33054370+ameer2468@users.noreply.github.com>
Co-authored-by: Jamie Pine <32987599+jamiepine@users.noreply.github.com>
Co-authored-by: Oscar Beaumont <oscar@otbeaumont.me>
Co-authored-by: Utku <74243531+utkubakir@users.noreply.github.com>
Co-authored-by: Brendan Allan <brendonovich@outlook.com>
