* wip
* added user metadata key
* wip
* restructure onboarding system and add initial locale
* update language card and fix translation updating
* remove prints
* new card formattings
* fix cursed unmount effect
* add OAuth route onboarding
* remove required admin auth for onboarding
* delete the hotwire button
* update open-api files
* delete import
* fix failing oauth onboarding fields
* fix e2e test
* fix web e2e test
* add onboarding to user registration e2e test
* remove todo
this was a holdover during dev and didn't get deleted
* fix server small tests
* use onDestroy to save settings rather than a bind:this
* change to false for isOnboarded
* fix other auth small test
* provide type annotation in user factory metadata field
* remove onboardingCompelted from UserDto
* move translations to onboarding steps array and mark as derived so they update
* break language selector out into its own component as per @danieldietzler suggestion
* remove hello header on card
* fix flixkering on server privacy card
* label/id fixes
* openapi
---------
Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
* add setting switch
this isnt bound to anything yet
* make google casting opt-in
* doc updates
* lint docs
* remove unneeded translation items
* update mobile openai defs
* fix failing test
we need to mock user prefs since CastButton uses it
* feat: api access control
* feat(web): granular api access controls
* fix test
* fix e2e test
* fix: lint
* pr feedback
* merge main + new design
* finalize styling
---------
Co-authored-by: Alex <alex.tran1502@gmail.com>
* more robust person thumbnail generation
* clamp bounding boxes
* update sql
* no need to process invalid images after decoding
* cursed knowledge
* new line
* feat: private view
* pr feedback
* sql generation
* feat: visibility column
* fix: set visibility value as the same as the still part after unlinked live photos
* fix: test
* pr feedback
* Transform imported face RegionInfo according to Exif Orientation
* Add unit tests for re-orienting metadata face regions
* Make code DRY using ImmichTagsWithFaces instead of NonNullable
* Add e2e test for importing metadata face regions when orientation is RotateCW90
* Disable new e2e test until its asset is added to the test-assets project
* Simplify unit tests by using the same face tag definition
* Combine similar e2e tests
* Disable new e2e test until portrait-orientation-6.jpg is added to test-assets
* Fix lint error: Expected property shorthand
* Update test-assets ref to latest
* Enable new e2e test after updating test-assets
We would like to move away from the concept of finding and removing pending
jobs. The only place this is used is for album update notifications, and this
is done so that users who initially uploaded assets to an album will also
receive a notification if someone else then adds assets to the same album. This
can also be achieved with a job for each recipient. Multiple jobs also has the
advantage that it will scale better for albums with many users, it's possible
to send notifications concurrently, retries are possible without sending
duplicate notifications, and it's clear what recipient a job failed for.
* chore(server): don't insert embeddings if the model has changed
We're moving away from the heuristic of waiting for queues to complete. The job
which inserts embeddings can simply check if the model has changed before
inserting, rather than attempting to lock the queue.
* more robust dim size update
* use check constraint
* index command cleanup
* add create statement
* update medium test, create appropriate extension
* new line
* set dimension size when running on all assets
* why does it want braces smh
* take 2
---------
Co-authored-by: mertalev <101130780+mertalev@users.noreply.github.com>
* bump sharp
* test linking
* link in prod image too
* force global
* keep unnecessary libraries
* override sharp version
* revert dockerfile changes
* add node-gyp and napi
* dev dependency
Job workers are currently started on app init, which means they are started
before the DB is initialised. This can be problematic if jobs which need to use
the DB start running before it's ready. It also means that swapping out the
queue implementation for something which uses the DB won't work.
* fix: ensure oauth state param matches before finishing oauth flow
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* chore: upgrade openid-client to v6
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* feat: use PKCE for oauth2 on supported clients
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* feat: use state and PKCE in mobile app
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: remove obsolete oauth repository init
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: rewrite callback url if mobile redirect url is enabled
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: propagate oidc client error cause when oauth callback fails
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: adapt auth service tests to required state and PKCE params
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: update sdk types
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: adapt oauth e2e test to work with PKCE
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
* fix: allow insecure (http) oauth clients
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
---------
Signed-off-by: Tin Pecirep <tin.pecirep@gmail.com>
Co-authored-by: Jason Rasmussen <jason@rasm.me>
It seems insta360 stores metadata in XMP GPano tags, with their own
non-standard and undocumented addition `SourceImageCreateTime`. For some
pictures this is the only EXIF tag containing a creation time.
* remove face, person and face search entities
update tests and mappers
check if face relation exists
update sql
unused imports
* pr feedback
generate sql, remove unused imports
* feat: sync pictureFile with oidc if it isn't set already
fix: move picture writer to get userId
fix: move await promise to the top of the setPicure function before checking its value and automatically create the user folder
chore: code cleanup
* fix: extension double dot
---------
Co-authored-by: Jason Rasmussen <jason@rasm.me>
* always patch package when running npm i, install immich CLI outside of directory so post install doesnt run
* handles case where query is an object and defined but origin is not.
* move patch-package from a dev dependency to a normal dependency. Also copy the patches folder for the docker build to use and patch with
* fix Dockerfile
* use query.reject instead of throw for queryError
* package-lock to reflect the dev dependency change
* dont throw the error, just provide an empty string for query.origin if it does not exist
* remove npm link and demote patch-package back to a dev dependency
* modify patch to add defensive check to catch queries that will fail to parse and reject
* chore: add geodata indexes to table definitions
* chore: rename incorrectly name geodata index
* fix: import into geodata places with correct index names
* feat(server): Avoid face matching with people born after file creation date (#4743)
* lint
* add medium tests for facial recognition
---------
Co-authored-by: Alex <alex.tran1502@gmail.com>
* feat(server): extract full-size previews from RAW images
* feat(web): load fullsize preview for RAW images when zoomed in
* refactor: tweaks for code review
* refactor: rename "converted" preview/assets to "fullsize"
* feat(web/server): fullsize preview for non-web-friendly images
* feat: tweaks for code review
* feat(server): require ASSET_DOWNLOAD premission for fullsize previews
* test: fix types and interfaces
* chore: gen open-api
* feat(server): keep only essential exif in fullsize preview
* chore: regen openapi
* test: revert unnecessary timeout
* feat: move full-size preview config to standalone entry
* feat(i18n): update en texts
* fix: don't return fullsizePath when disabled
* test: full-size previews
* test(web): full-size previews
* chore: make open-api
* feat(server): redirect to preview/original URL when fullsize thumbnail not available
* fix(server): delete fullsize preview image on thumbnail regen after fullsize preview turned off
* refactor(server): AssetRepository.deleteFiles with Kysely
* fix(server): type of MediaRepository.writeExif
* minor simplification
* minor styling changes and condensed wording
* simplify
* chore: reuild open-api
* test(server): fix media.service tests
* test(web): fix photo-viewer test
* fix(server): use fullsize image when requested
* fix file path extension
* formatting
* use fullsize when zooming back out or when "display original photos" is enabled
* simplify condition
---------
Co-authored-by: mertalev <101130780+mertalev@users.noreply.github.com>
* add patch-package to dev dependencies
this allows us to patch upstream packages without waiting for PRs to be merged (or not!). Patch-package does a pretty good job of notifying if upstream does a change to invalidate the patch (its a git patch under the hood).
* Patch implementation of https://github.com/porsager/postgres/pull/944
This PR has not been merged by upstream and helps produce verbose error messages when postgres fails to connect (usually incorrect credentials). This is in contrast to error messages such as
`TypeError: Cannot read properties of undefined (reading 'replace'), stack: TypeError: Cannot read properties of undefined (reading 'replace')`
* have postinstall only run when not installing a global package (such as immich-cli in the Docker build)
we can't run specifically the handleUserDeleteCheck tests concurrently due to one of the tests modifying the config in the shared database
if run concurrently you can get race conditions where the other tests pick up the change, even with resetting the config in the beforeEach
therefore the test that checks a delete actually happens, fails
there are many ways to solve this, disabling concurrency for the suite, forcing sequential tests for just handleUserDeleteCheck, increasing the delete test deletedAt to more than the custom duration tests deleteDelay
I applied all three of these. You could also force all the user tests to run in their own databases, but that feels overkill
* fix(server) optimize number of sql calls for GET /api/albums
remove unnecessary join for getMetadataForIds
remove separate call to getLastUpdatedAssetForAlbumId
* fix(server) remove unnecessary getLastUpdatedAssetForAlbumId call for GET /api/album/:id
also remove getLastUpdatedAssetForAlbumId query as it is no longer referenced
* fix(server): correct lastModifiedAssetTimestamp return type + formatting and typing
* chore(server): address type issue with tests found via npm:check
tests & lint still pass before this commit.
* feat: use my.immich.app for externalDomain fallback
This is probably more useful than localhost.
* chore: remove port param
* fix: update expected value in tests
* fix: update expected value in e2e
* 16712: Proper intialisation of the memory store to avoid loading up duplicate object refs of the same asset.
* 16712: Add auth to memory mapping so isFavorite is actually return correctly from the server.
* 16712: Move logic that belongs in the store into the store.
* 16712: Cleanup.
* 16712: Fix init behaviour.
* 16712: Add comment.
* 16712: Make method private.
* 16712: Fix import.
* 16712: Fix format.
* 16712: Cleaner if/else and fix typo.
* fix: icon size mismatch
* 16712: Fixed up state machine managing memory playback:
* Updated to `Tween` (`tweened` was deprecated)
* Removed `resetPromise`. Setting progressController to 0 had the same effect, so not really sure why it was there?
* Removed the many duplicate places the `handleAction` method was called. Now we just called it on `afterNavigate` as well as when `galleryInView` or `$isViewing` state changes.
* 16712: Add aria tag.
* 16712: Fix memory player duplicate invocation bugs. Now we should only call 'reset' and 'play' once, after navigate/page load. This should hopefully fix all the various bugs around playback.
* 16712: Cleanup
* 16712: Cleanup
* 16712: Cleanup
* 16712: Cleanup
---------
Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
* feat(web): exposed a new job to create a manual database backup
* chore(server): added a new test case
* chore(server): moved job to backup db into the create job popup
* remove irrelevant change
* openapi
* chore: formatting
* docs: trigger backup documentation
---------
Co-authored-by: Lorenzo Montanari <13736036+l0ll098@users.noreply.github.com>
Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
Co-authored-by: Zack Pollard <zack@futo.org>
The API currently does not respect the documentation when returning a
person's birthDate. The doc/swagger says it will be of "YYYY-MM-DD"
format but the string is a full ISO8601-with-tz string. This causes
issue #16216 because the <input> tag is strict about supported value
formats.
I believe this was introduced by #15242 which switched some queries from
TypeORM to Kysely for the person repository. TypeORM does not parse
date, but our Kysely configuration does (explicitely).
This commits updates the types to represent both possibilities and ensure
the API always returns the correct format.
