Batching must not start while files are still being uploaded, and the MD5
phase must move a whole chunk at once instead of one resolve per file:
- the upload queue drains completely first (failed uploads included) before
any matching starts;
- phase 1 asks e621 for every md5 (75 per posts.json request), builds the
md5 -> post map from the response, and sends the matches to the new
POST /api/uploads/link-bulk/ action, so a whole 75-file chunk moves into
Indexed in a single board update;
- link-bulk indexes the staged file directly when the post's MD5 matches
(identical bytes), so there is no per-file download round trip;
- phase 2 runs local visual similarity for whatever stayed pending, phase 3
the IQDB queue.
Verified end to end with real e621 files: one md5 query for the batch, one
link-bulk call, both matching files flipped to Indexed together, then the
visual and IQDB phases. 23 library tests green (link-bulk, visual phase,
deferred visual matching).
Uploads were doing md5 + local visual matching inside the upload request
(backend create) while the frontend later ran its own e621 MD5 pass, so the
pipeline looked interleaved per file. Now every step is a phase applied to
the whole batch in order:
1. upload (fast: md5 + exact-duplicate check only),
2. e621 MD5 lookup, 75 md5: metatags per posts.json request,
3. local visual similarity, one file at a time via the new
POST /api/uploads/<id>/visual-match/ action,
4. IQDB through the existing serial queue.
The board shows the active phase with its own progress bar (e621 MD5 in
peach, visual in lavender, IQDB in teal) and every step updates the staged
list as it lands. Verified from a headless run: one batched posts.json
request for 10 files, then 10 visual-match calls, then IQDB.
e621 intermittently answers 429 to the IQDB endpoint; browsers hide that
status behind CORS ('Access-Control-Allow-Origin missing'), so the SPA
cannot read it. Treat every network-level failure as a possible rate limit
and pause all e621 traffic for a minute. The cooldown is shared through
localStorage so extra tabs respect it, requests are spaced 1.5s apart
instead of 1s, user-cancelled requests do not trigger a cooldown, and the
upload queue waits the cooldown out with a countdown instead of looking
stuck.
Server side: the per-process e621 gap goes from 0.5s to 1s so two gunicorn
workers cannot together exceed e621's 2/s hard limit.
The modal held a snapshot of the staged upload, so IQDB results that landed
from the background check queue never appeared until it was closed and
reopened — the only hint a check was running was the e621 request history.
It now follows the live uploads query, so candidates, progress and errors
show up in place.
Related gaps fixed along the way:
- files flagged by the local visual-similarity check were skipped by the
IQDB pass entirely (only 'pending' files were checked), so their modal
could only ever show 'already in your library'; unresolved files of both
statuses are now checked, and the check button shows on visual-match
cards too;
- a check with no candidates posted nothing, leaving 'never checked' and
'checked, no match' indistinguishable; results are stored even when
empty and the modal now says which one it is;
- per-file failures surface in the modal instead of being swallowed, the
modal shows a spinner while the query runs and a check now/re-check
button, and auto-runs skip files already checked (and videos, since IQDB
is image-only).
Backend production code unchanged; tests pin the empty-result recording
(18 library tests, full suite 56 green).
Upload board:
- the tile grid no longer re-sorts itself as files finish (that reshuffled
the list under the cursor); it keeps insertion order, uses auto-fill tiles
of ~150px so they hold a readable size, scrolls inside a 60vh area and no
longer chains the page scroll (overscroll-contain);
- the files currently in flight are pinned in a small live strip above the
grid (name, percent, bar) so progress stays visible while the grid is
scrolled with hundreds of tiles.
Bulk rating: a 'bulk rate' button in the Pending & Unmatched header opens a
large modal with Safe/Questionable/Explicit pills, a tickable thumbnail grid
(Select all / Clear) and one confirm that moves every selected upload into
the library with that rating. Backed by POST /api/uploads/resolve-bulk/
(temp_ids + rating, own rows only): each staged file is resolved as a custom
entry (keeps its staged tags/notes), and already-completed or foreign ids are
reported per entry instead of failing the whole batch. Built for the
358-file backlog.
Tests: 4 bulk-resolve tests (resolution with the rating, input validation,
foreign ids untouched, mixed completed+pending) — full backend suite 53
green. Verified live end to end: staged a file, bulk-resolved it as 'q', saw
J-96 created with that rating, then removed the item, temp row and test
token.
The upload board partitions /api/uploads/ into Pending / Visual similarity /
Auto-uploaded, but the endpoint was paginated at 48 — a 69-file batch
silently lost 21 entries, and the similarity sweep (which reads the same
list back after uploading) only ever saw the first page. The staged-upload
list is now unpaginated: it is a transient per-user set, still limited to
the caller's rows and the uploader role. The page takes a plain array.
Watching progress with dozens of files was also poor:
- the queue uploads three files at a time instead of strictly one at a time;
- the Uploads section now shows a batch bar and 'n/m uploaded · x%' next to
the count, so the overall progress never scrolls out of sight;
- entries are ordered active-first (uploading, queued, failed, done) so the
file being uploaded is always at the top of the grid;
- tiles are larger (4 columns at lg instead of 5);
- the header reads 'Uploading n/m…' and 'Checking n file(s) against IQDB…'
instead of a bare spinner.
Tests: staged-upload list unpaginated past 48, per-user, uploader-only
(3 new; full suite 49 green). Live-checked the bare-array response.
Backend: a GreetingToken model stores only a SHA-256 hash of a j621r_…
key (shown once at creation) plus label, prefix, created/last-used. A
dedicated GreetingTokenAuthentication understands the usual
'Authorization: Token …' header but is registered only on RandomItemView
(alongside the normal token auth), so a greeting token authenticates
/api/random/ and is rejected with 401 everywhere else — exactly the scope
shell greetings need. Endpoints: GET/POST /api/auth/greeting-tokens/ and
DELETE /api/auth/greeting-tokens/{id}/ (own tokens only; the list never
returns keys or hashes).
Frontend: /tokens page (Account → Shell tokens card, command palette entry)
lists tokens with label, prefix, created/last-used and revoke (shared
confirm dialog). Creating one shows the key with Copy and 'Copy for fish'
buttons plus a pointer to extras/fish_greeting.
Tests: apps/accounts/tests/test_greeting_tokens.py — 9 tests covering
create-once semantics and hashing, hidden keys in listings, the scope
guarantee (random 200 with a signed URL; 401 on files, storage, me, tags
cloud, delete and the token list itself), unknown/revoked keys, cross-user
revocation, last-used tracking and label limits.
Verified live: created a token, rolled /random (signed URL), got 401 from
four other endpoints, saw the list omit secrets, revoked it (204) and the
same key then 401'd on /random. Full suite: 39 tests green.
Backend: GET /api/random/ (aliases /random and /random/) returns a random
library image with:
- rating=s,q,e filtering (comma separated, default any);
- fastfetch mode (?fastfetch=1 or any User-Agent containing "fastfetch")
that only considers png/jpg/gif - what terminal viewers can show;
- JSON with j_id, filename, extension, rating, size, e621 id plus absolute
url/download_url/thumbnail_url. Authenticated callers get signed URLs so
fastfetch and image viewers can load them without headers; guests get
unsigned URLs and never receive hidden_from_guests items.
Tests: apps/library/tests/test_random.py (8 tests) covering the response
contract, guest signatures, image-only default, the fastfetch format
restriction (flag and User-Agent), rating filters, guest visibility and the
short alias.
Frontend: /random page with rating pills, R to roll, Open/Download and a
library link, plus navigation and command palette entries; needs a backend,
hidden in local mode.
nginx: /random negotiates on Accept so browsers keep getting the SPA while
scripts get the JSON (verified with the proxy and frontend containers).
Also fixes a regression from the SSRF change: the guest download proxy
still referenced the removed 'parsed' variable on its success path, so
every proxied download would have 500'd. Redirect hops are now covered by
tests with a mocked requests.get.
Findings from the audit (50-check harness across guest/user/uploader/staff/
admin) and their fixes:
- SSRF: 'Download to Library' and the staged-upload resolve path fetched
any http(s) URL. services.validate_remote_url now enforces the e621
media allowlist and open_remote re-validates every redirect hop; the
download-task create endpoint and the guest proxy use them, so internal
addresses (127.0.0.1, LAN, metadata) are rejected with 400.
- Privilege escalation: staff could promote users to staff and demote
other staff. Role changes across the staff boundary now require an
admin, matching the account-deletion rules; the Users page hides what
the backend would refuse.
- SPA-only gating: /api/storage/ and /api/duplicates/* were readable by
any authenticated account (absolute paths, duplicate groups) while the
SPA only shows them to uploaders. They now require CanUpload.
- Throttling (REST_FRAMEWORK, env-overridable, counted in Redis):
anon 120/min, user 600/min, login 5/min, register 20/hour, guest e621
proxy 60/hour. Login now goes through a throttled view.
- e621 API keys are encrypted at rest with a Fernet key derived from
SECRET_KEY (apps/accounts/crypto.py); a data migration encrypts existing
rows and the column widens first. Reads decrypt transparently, legacy
plaintext still works, and a changed SECRET_KEY reads as 'not
configured' instead of leaking. Rotating SECRET_KEY now invalidates
stored keys as well as signed media URLs.
- Hardening: the server refuses to start with DEBUG=False while SECRET_KEY
is still the development default.
Verified: corrected harness 50/50 (guest visibility, IDOR, signed-URL
tamper/expiry, staged-upload/similarity privacy, role matrix, SSRF),
login throttles at the 6th attempt with 429, anon polling unaffected, the
guest proxy still reaches allowlisted hosts, live e621 auth works with the
decrypted key, and DB rows hold only ciphertext.
Footer:
- Left is now 'Backend Storage:' with a capacity bar (blue, peach at 80%,
red at 95% per DESIGN.md) and a used/total/free tooltip; the watched
folder path is no longer printed. /api/status/ returns a compact storage
summary instead of the path (the full storage page still shows paths to
authenticated users).
- Centre shows the backend API origin (empty = same origin). Staff get a
link to /setup to point the browser elsewhere; everyone else sees it as
plain text. The Account 'Backend connection' card is gone — this is
installation plumbing, not a per-user setting.
- Design spec updated to match.
Staff role:
- The custom role did nothing on several endpoints that only accepted
Django's is_staff/is_superuser. One canonical check now exists:
User.is_app_staff (superuser, Django staff, or the staff role), used by
the stats/users APIs, item object permissions, can_delete, upload/
similarity/download/match querysets, and the management commands
(which also pick staff-role accounts for e621 sync/match and file
ownership).
Verified with a role-only staff account (is_staff/is_superuser false):
stats/users 200, all 32 downloads + 2 scans visible, others' items
editable; the same account as role=user gets 403 for all of those.
- django-cors-headers with env-driven CORS_ALLOWED_ORIGINS,
CORS_ALLOW_ALL_ORIGINS, CORS_ALLOW_CREDENTIALS and CSRF_TRUSTED_ORIGINS;
same-origin traffic is unaffected and a disallowed origin gets no CORS
headers. Token auth needs no cookies, so credentials stay off by default.
- TRUST_PROXY_HEADERS=true lets a TLS-terminating proxy supply
X-Forwarded-Proto/Host for correct absolute URLs.
- API media URLs (raw/thumbnail/upload/similarity/staged previews) are now
absolute, built from the request host, so <img>/<video>/fetch() keep
working when the SPA is served from another origin. Signed URLs are still
per-user; nothing is stored in the DB.
- The SPA gains VITE_API_BASE (build-time, empty = same-origin) applied by
a small apiUrl() helper used for XHR/fetch and the few URL fallbacks.
Verified with a throwaway instance: preflight and GET responses carry the
allowed origin, foreign origins get nothing, media GETs include CORS for
cross-origin fetch(), and payload URLs use the request host (dev :8000
unchanged).
Jobs already run on the server — leaving the page or closing the tab does
not stop them — but the SPA lost its link to them because the task id
lived in component state. The online detail page now looks up the newest
task for the post: an active one resumes the progress bar and cancel
button, and a finished one shows "your last download for this post
finished — J-xx". The downloads list accepts a post_id filter for that
lookup.
The footer's Active Workers count is also a link to the staff stats
dashboard, which is the global view of running jobs.
- Active jobs on /stats get a cancel button wired to the existing
download/match cancel endpoints, showing "cancelling..." and an inline
error when the task already finished.
- Cancelling now sets the status immediately, so a task whose runner died
in a restart stops showing as "downloading".
- Download streams use a bounded read timeout (10 s connect / 60 s read):
a stalled socket fails within a minute (previously it could block
forever), and a task cancelled while stalled is marked cancelled rather
than error.
- The stats job list reaps stale download/match tasks, so phantom jobs
never appear on the dashboard.
Backend: GET /api/stats/ (staff only) gathers psutil CPU/memory counters,
nvidia-smi GPU stats, the cached disk numbers and the running/finished
download + match jobs. Root logging now also writes a rotating file
(backend/logs/j621.log) so the dashboard can tail it, and psutil joins the
requirements. The storage payload computation is shared with the existing
storage endpoint.
Frontend: a /stats route + Stats nav entry for staff, polling every 2 s —
per-core CPU bars, memory and swap, GPUs (utilization, VRAM, temperature),
disk with the media/temp breakdown, active jobs with progress bars,
recently finished jobs with summaries, and the log tail with level colours
and an auto-scroll toggle. Section 4 of the roadmap is complete.
Backend:
- POST /api/files/J-x/optimize/ applies a browser-processed file: replaces
every copy (renaming when the extension changes), recomputes MD5, size
and perceptual hashes, seeds guest visibility; 400 when identical,
409 when the result matches another item, owner/staff only.
Frontend (no server-side processing by design):
- optimize.worker.ts + pipelines: Mediabunny/WebCodecs for video with a
prefer-hardware hint and per-browser codec detection; MozJPEG/OxiPNG/
libwebp (jSquash) for images; gifuct-js+gifenc and UPNG for GIF/APNG.
- OptimizeModal: per-file-type options, original vs processed previews
with sizes/savings, progress bar with ETA, then Apply (overwrite).
- Optimize button on the library detail for the uploader/staff.
run_download_task sets e621_match_status from MediaItem, but the module
only imported DownloadTask — every download that carried e621 metadata
failed right after indexing, leaving the file in the library unlinked.
Verified the runner end-to-end with a stubbed fetch.
- /similar (nav: Similar): drop a file to get the exact MD5 match, the
perceptual matches against the library, and e621 IQDB candidates
(auto-run for images when credentials are configured). Read-only —
nothing enters the library.
- SimilarityCheck model + /api/similarity/ (create/list/retrieve/delete)
with signed preview URLs and an expires_at timestamp.
- Temp files are wiped on startup (AppConfig.ready, file-only so no
database access during initialization), lazily past
SIMILARITY_TTL_MINUTES (default 30, env-overridable), on delete, and
by manage.py cleanup_similarity.
- uploadFile() takes a target path; .env.example documents the TTL.
Backend (new apps.follows):
- FollowedTag/FollowedPool/FollowedPost models; per-user follows with
unseen tracking, plus FollowCloud for the cached blacklist cloud.
- Two periodic commands sharing one fetch path: sync_followed_tags and
sync_followed_pools fetch each followed tag/pool's newest posts (one
e621 search per unique follow), store unseen feed rows, refresh covers
and pool metadata; both fall back to anonymous e621 access.
- API: /api/follows/tags|pools (follow, unfollow, mark seen), a merged
feed with per-follow filtering, and /api/follows/cloud/ which rebuilds
the blacklisted-tag cloud in a daemon thread when its 10 min cache is
stale (polling returns building/ready).
- e621 client now supports anonymous reads; trimmed posts carry preview
URLs for covers and feed tiles.
Frontend:
- /followed page: follow forms, cover cards with unseen badges and
Mark seen, merged feed with filter/unseen toggle, and a blacklist
cloud panel that polls while building. Followed nav entry added.
- MediaItem gains e621_match_status (unknown/matched/not_found/deleted)
and e621_checked_at, backfilled for existing matched items.
- Server-side e621 client (apps/library/e621.py) using the user's stored
credentials, throttled to 2 req/s, with typed errors.
- Matching service: MD5 lookup, manual post linking (flags MD5
mismatches), unlink, metadata refresh, deleted-post detection.
- Detail actions POST /api/files/J-x/match/ and /unlink/ (uploader or
staff only).
- Background library scans: MatchTask + /api/matches/ with missing/all
scopes, progress polling, cancel and stale-task reaping; the scan
counts toward the footer's Active Workers. Same pass available as
manage.py match_e621 for cron.
- Library gains not_found/deleted status filters; the detail page adds
an e621 match card (check / link by post ID / unlink) and the metadata
card warns when a post was deleted on e621.
Items flagged hidden_from_guests (blacklisted tags) returned 404 for
<img> requests since tags cannot send the auth header. The API now
exposes signed raw_url/thumbnail_url fields (mirroring upload previews
and avatars), and the SPA uses them in the gallery, detail view,
duplicates and delete screens, and upload visual matches.
Backend:
- MediaItem gains search_tags (custom + e621 tags, lowercase) and
has_custom_data, maintained on save with a data migration backfill
- File list search accepts search_type=filename|tags|both (tag search is
word-AND across the flattened tag text) and status=matched|custom|
unknown filters
- New /api/tags/cloud/ endpoint (cached 2 min per guest/auth, invalidated
on item changes and deletions) returning the most-used tags, honouring
guest visibility
Frontend:
- Library sidebar: Filename/Tags/Both selector, status pill toggles
(persisted), and a clickable tag cloud that runs a tag search
- Roadmap updated
Backend:
- Perceptual hashes (aHash/dHash/pHash/wHash via imagehash, no imgdd)
stored on items, computed on upload/download and by the new
compute_visual_hashes command
- Duplicates API: exact duplicates (multi-location items), visual matches
for one item, union-find similarity groups with pagination
- Delete API with ownership/staff checks, per-item and per-copy deletion,
watched-folder path validation; storage overview and temp cleanup;
file list accepts j_ids batches
- Staged uploads are flagged visual_match with their library matches
(threshold via VISUAL_MATCH_THRESHOLD)
- Staff users API: list with upload counts, set role and avatar by J-ID;
User.avatar FK with signed avatar URLs
- Download threads close their DB connection and stale tasks are reaped,
keeping behaviour Gunicorn-friendly
Frontend:
- /duplicates: exact duplicate groups with per-copy delete, visual
similarity controls, search similar to a J-ID, paginated groups with
selection, bulk delete and dismiss
- /delete: storage cards, delete by J-ID with preview grid, temp cleanup
- /users: staff directory with role selects and avatar J-ID inputs
- Nav + command palette entries; top-bar avatar; upload cards and the
metadata modal show library visual matches
Backend:
- DownloadTask model + background thread runner: streams the file with
progress (%, bytes, speed) and a cancel flag, then indexes it, names it
J-<id>.<ext> and applies the e621 metadata
- DownloadTaskViewSet (create/retrieve/cancel) replaces the synchronous
endpoint; the status footer's worker counts now reflect download jobs
- Client download proxy (/api/online/file/) streams an e621 original to
the browser with Content-Disposition: attachment, restricted to the
configured e621 CDN hosts so it cannot be used as an open proxy
Frontend:
- Online detail: progress bar with percentage, transferred size, speed
and cancel while downloading; success links to the new J-ID
- New 'Download to client' button available to everyone (guests too)
Files added through the upload pipeline and Download to Library are
renamed to their J-ID right after indexing, so every new library file is
traceable by its identifier (scanned files keep their existing names).
index_file now returns the created location so callers can rename it;
the location record is updated to the new path.
- 'dismiss all' clears every indexed record at once
- Indexed cards show the actual file preview: completed records now get a
signed library media URL (raw for images, thumbnail for videos) so
<img>/<video> tags can load it, including items hidden from guests
- Media raw/thumbnail endpoints accept the signature for anonymous
requests and fall back to the normal guest-filtered path otherwise
- Guest blacklist keeps a persistent Redis mirror: an expired TTL or an
unreachable e621 keeps the last successful list instead of falling
back to the small local list
- IQDB responses carry no preview/file data, so candidates only showed an
ID; the SPA now enriches them with one batched posts lookup (preview,
rating, score, favourites, dimensions, tag preview)
- Candidate tiles are selectable instead of instantly resolving: picking
one shows its info and an explicit 'Link selected post' button
- Linking a post now fetches the e621 original into the library and
drops the staged upload; when the staged file's MD5 already equals the
post's file, the staged copy is moved instead (identical bytes)
- Keep the file URL in stored e621 metadata; sanitize the new candidate
fields server-side
- Staged files are now served through a signed URL (Django signing, 24h)
so <img>/<video> tags can load previews without an Authorization
header; the file endpoint accepts header auth or a valid signature,
rejects tampered signatures, and still scopes access to the owner
- Serializer responses now carry the request context so URLs are signed
per user
- Add .webp to the allowed extensions (backend + upload hint)
Backend:
- TempUpload model: staged files (pending / visual_match / completed /
error) with resolution, e621 payload, custom metadata and IQDB data
- Files land in a temp folder and only move into the watched library
folder once resolved; duplicates resolve immediately without a copy
- Endpoints: stage (multipart), list, retrieve, temp file, IQDB save,
resolve (link to post or custom metadata), discard/dismiss
- cleanup_temp_uploads command for old staged files
- Replaces the old direct-to-library upload endpoint
Frontend:
- Upload page is now a three-column board (Pending & Unmatched /
Visual Similarity Detected / Auto-uploaded & Indexed)
- After upload: MD5s are batch-checked against e621 and matches
auto-complete with full post metadata; remaining files run through
IQDB and move to the similarity column when candidates exist
- Metadata modal with IQDB candidates, post-ID linking and custom
tags/rating/notes; discard and dismiss actions
- e621 client gains fetchPostsByMd5 and iqdbSearch helpers
Roadmap updated with the completed upload items.
Roles:
- JakeBreathild is now staff + superuser (the real account); the 'jake'
smoke-test account was demoted to a regular user
- /me exposes is_superuser and the account page shows an admin badge
e621 metadata:
- MediaItem gains e621_post_id and e621_data (trimmed post payload:
tags by category, rating, score, favourites, comments, sources,
description, pools, relationships, file info, uploader)
- Download to Library accepts the post payload from the SPA and stores
it; the item's custom rating is seeded from the e621 rating when empty
- Library detail shows an e621 metadata card: link to the in-app post,
score/favourites/comments, taxonomy-coloured tags, DText description,
sources and pools; grid cards get an e621 badge and fall back to the
e621 rating for their colour (display_rating)
- Guest visibility now also considers e621 tags, so downloaded explicit
content is hidden from anonymous visitors
Backend:
- User.role (user/uploader/staff) with can_upload; uploads and downloads
gated to uploader+; owners and staff can edit their items
- MediaItem.uploaded_by plus J-<id> identity (serializer, admin,
scan_files --user, first superuser as default owner)
- API resolves J-<id>, bare numeric ids and MD5s; neighbors and lookup
return j_ids
- Guest safety: mirror e621's anonymous default blacklist into Redis
(parses comments, negations and wildcards), flag hidden_from_guests
and filter lists, details and lookups for anonymous users
- POST /api/online/downloads/ writes an e621 file into the watched
folder and indexes it for the uploader
- MariaDB + Redis via docker compose (host ports 3307/6380), PyMySQL
driver shim, Redis cache replacing the file cache; SQLite data
dumped and loaded into MariaDB
Frontend:
- Single /detail/:itemId route with an adaptive shell: J-<id> renders
the library item, bare numbers render the e621 post
- Legacy /view/<md5> and /online/view/<id> redirect to canonical URLs
- Cards expose J-IDs; library custom-data editor is read-only for
non-owners
- Role gating: Upload hidden/blocked for regular users, account shows
the role, guest hint on the library
Backend:
- POST /api/files/lookup/ reports which MD5s are already in the library
Frontend:
- e621 client extended with post/tag/favorite types and helpers: post
search, post detail, batch posts by id, tag autocomplete, toggle
favorite
- /online: tag search with autocomplete, post grid with rating colors
and in-library badges, numbered pagination, page tag cloud, blacklist
panel and filtered counts from /users/me.json, anonymous hint
- /online/view/🆔 media viewer (sample or original, video support),
taxonomy-colored tags by category, specs sheet, favorite/unfavorite,
description, sources, pools and parent/children thumbnails
- Shared CollapsibleSidebar extracted from the library page; Online
added to the nav, command palette and sidebar toggle
Backend:
- GET /api/files/{md5}/neighbors/ returns previous/next items in the
current ordering (name, size, created_at) for keyboard navigation
Frontend:
- / focuses the library search input
- D downloads the file on the detail page
- [ and ] navigate to the previous/next item; hint shown on the page
- Ctrl/Cmd+K opens a command palette (navigation, toggle filters,
focus search, log in/out); Escape closes it
- Sort order now persists alongside the other library filters so
prev/next stays consistent
Backend (Django 6.1 + DRF):
- Token auth with a custom User model (register/login/logout/me)
- Library models (MediaItem, MediaLocation) and REST endpoints
- File list/detail with search, rating filter, sorting, pagination
- Multipart upload with optional rating/tags/notes
- Range-aware media serving (video seeking) and ffmpeg thumbnails
- scan_files management command for the watched folder
Frontend (React 19 + Vite + TypeScript):
- Catppuccin Mocha design tokens from the design docs
- App shell, token persistence, protected routes
- Library grid with filters, file detail with custom data editor
- Upload page with per-file progress via XHR
- Dev proxy to the Django API
JakeBreath