Closed
Conversation
* feat : add compute_event_embedding to FaceEmbeddingService * chore: add .venv to gitignore --------- Co-authored-by: bentafat wail <150479778+wailbentafat@users.noreply.github.com>
* feat: return all detected faces with embeddings * feat: add batch face embedding service * feat: add batch face embeddings endpoint * chore: register batch face embeddings router * feat: wire batch face embedding service in container * feat: add photo_faces upsert query * chore: add generated photo_faces querier * feat: add batch face embeddings request schema * feat: add batch face embeddings response schema * fix: match photo_faces upsert signature * fix: commit/rollback per face and serialize bbox floats * feat: wire batch queue service into container * chore: add face-embedding stream settings * feat: add batch embeddings subject and ensure stream * feat: switch batch endpoint to enqueue jobs (202) * feat: add batch face embedding request DTO * feat: add batch face embedding enqueue response * feat: add batch face embedding job DTO * feat: enqueue batch face embedding jobs to JetStream * feat: add JetStream worker for batch face embeddings * chore: migrate photo_faces embedding to 512 dims * chore: merge alembic heads * chore: add sql up/down for photo_faces embedding dim change * chore: centralize shared content-type and url constants * chore: use default content type constant in MinIO * chore: use default content type constant in enrollment * chore: use constants for content type and source parsing * fix: refactor batch face embedding flow for clarity * chore: update generated db queriers * Add token blacklist and blocked checks in auth * Add admin user CRUD and block/unblock endpoints * Use settings and consistent DB error handling in user service * chore: remove token blacklist helpers * chore: remove legacy batch face worker * feat: add single-face match worker * refactor: move single-face worker into folder * refactor: move single-face worker into folder * feat: harden single-face match processing * chore: improve worker shutdown behavior * refactor: move constants into core * refactor: move settings into config * refactor: use core constants for MinIO buckets * refactor: use core constants for Google URLs * refactor: load face model settings from config * refactor: use config for MinIO retries * WIP: save work before rebase * Bon.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds a new grouped upload flow for staff users.
Until now, staff uploads only supported manually selected Google Drive files. With this change, the backend can also accept a Google Drive folder_id, read the folder contents, filter valid images, split them into batches, and create grouped upload requests under one parent import group.
The existing request-based upload flow is still supported and now shares a single unified entry point.
What was implemented
Added support for creating uploads from either:
photos
or folder_id
Added a new upload_request_group concept for folder imports
Added Google Drive folder listing support
Added automatic filtering of supported image files
Added deterministic sorting before batching
Added batching of folder images into upload requests of up to 20 photos each
Added group-level endpoints for:
listing groups
retrieving one group
listing group photos
approving a group
rejecting a group
Added NATS events for upload groups
Added database support for:
upload_request_groups
group_id on upload_requests
Flow
If the frontend sends photos, the backend creates a normal upload request
If the frontend sends folder_id, the backend:
lists the Google Drive folder
keeps only valid images
sorts them deterministically
splits them into batches of 20
creates one parent upload group
creates one child upload request per batch
stages all photos in MinIO
returns the full group with child requests and staged photo