Bug 1 & 2 (webdav_handler.rs handle_put() update branch):
- After a successful file update via WebDAV PUT, if the content type is a supported image:
a. delete_thumbnails(file_id) — evicts the stale moka cache entry
b. Spawns a background task to read the new blob bytes and call generate_all_sizes_background_from_bytes
Bug 3 & 4 (dedup_service.rs):
- Added thumbnail_service: Option<Arc<ThumbnailService>> field with a with_thumbnail_service() builder
- In remove_legacy_reference(): calls delete_blob_thumbnails(hash) when ref_count hits 0
- In remove_manifest_reference(): calls delete_blob_thumbnails(file_hash) when manifest's last ref is dropped
- Wired in di.rs — the thumbnail service is created before dedup service so the ordering works cleanly
Security: session hardening
Refresh token rotation with theft detection (family_id)
- Added family_id column to auth.sessions (migration 20260507000000_session_family.sql) grouping all tokens issued from the same login into a family
- On refresh, the new session inherits the parent's family_id
- If a revoked token is replayed (indicates the token was stolen after rotation), the entire family is immediately invalidated and a warning is logged — forcing re-authentication on all devices
SameSite=Strict on refresh cookie
- Access cookie stays SameSite=Lax (needed for top-level navigation)
- Refresh cookie upgraded to SameSite=Strict — it is only ever used for explicit POST to /api/auth/refresh, never via cross-site navigation
Refresh token TTL: 30 days → 7 days
- With rotation, active sessions auto-renew and effectively never expire
- Inactive sessions expire after 7 days instead of 30, reducing the theft window
Five new public endpoints under /api/s/{token}/...:
GET /contents
GET /contents/{folder_id}
GET /file/{file_id}
GET /zip
GET /zip/{folder_id}
All honour the unlock cookie from /verify, so password-protected
folder shares work end-to-end.
Folder/file IDs are validated against the share subtree via a single
ltree containment query (O(log N) on the existing GiST index).
Out-of-scope IDs return 404.
download_shared_file refactored to a Range/304/206/416-aware
serve_share_file helper, shared with the new /file/{file_id}
endpoint. content_disposition extracted from FileHandler so RFC 5987
formatting is identical across auth and share download paths.
After successful POST /api/s/{token}/verify, the server issues a
short-lived signed JWT cookie (oxi_share_unlock_<token>; HttpOnly;
SameSite=Lax; 1h TTL). Subsequent /api/s/{token} and
/api/s/{token}/download requests honour the cookie to bypass the
password gate, so password-protected file downloads work end-to-end.
- New module src/infrastructure/services/share_unlock_cookie.rs:
issue_jwt, verify_jwt, extract_from_cookie_header, build_set_cookie
+ 10 unit tests.
- New ShareService methods issue_unlock_jwt and
get_shared_link_with_unlock; trait method get_shared_link_by_token
delegates to a private fetch_share_resolved with an
allow_password_protected flag.
- access_shared_item, verify_shared_item_password, and
download_shared_file honour the unlock cookie.
Reuses OXICLOUD_JWT_SECRET, no new env var. Auth-token JWTs and
unlock-cookie JWTs cannot be confused: auth requires
username/email/role/jti claims, unlock has only sub/exp/iat.
Cross-share replay rejected via claims.sub == requested_token check.
Decode XML-escaped WOPI discovery action attributes before storing them
so generated editor URLs use real query separators instead of literal
& entities. Add regression coverage for parsed actions and generated
edit URLs.
After the blob_hash dedup change, get_cached_thumbnail with
blob_hash=None (cache-first fast path) only checked moka. External
video thumbnails stored as ext-{file_id}.jpg on disk were not found
after a server restart when moka is empty.
Now checks ext-{file_id}.jpg on disk before falling through to the
blob-hash disk lookup, so video thumbnails survive restarts.
Thumbnails are now keyed by blob_hash on disk so identical files share
a single set of thumbnails (icon/preview/large). For 4000 duplicate
files with the same content, this reduces thumbnail storage from 12,000
files to just 3.
Changes:
- get_thumbnail_path() keys by blob_hash instead of file_id
- get_thumbnail(), get_cached_thumbnail(), generate_all_sizes_background()
accept blob_hash parameter for disk dedup
- generate_all_sizes_background() fast path: if blob-hash thumbnails
already exist on disk, skip image processing entirely and just
populate moka cache for the new file_id
- delete_thumbnails() only invalidates moka cache (shared disk
thumbnails must not be deleted when one file is removed)
- delete_blob_thumbnails() added for GC; garbage_collect() now cleans
up orphaned thumbnail files alongside blob files
- External thumbnails (video frames) stored as ext-{file_id}.jpg
since they are client-generated and not dedup-able
- ThumbnailPort trait updated with blob_hash parameters
- All handler call sites updated (file_handler, preview_handler)
- Tests updated for new signatures
- Move all sync I/O (id3::Tag, mp3_duration) into spawn_blocking via
extract_metadata_blocking() to avoid stalling Tokio worker threads
- Replace fetch_all with streaming .fetch() in reextract_all_audio_metadata
for O(1) memory usage regardless of audio file count
- Consolidate get_duration_secs into extract_metadata_blocking, eliminating
redundant file open (tag was read twice before)
- Replace blocking blake3::hash() with DedupService::hash_file() in
file_upload_service (create_file, update_file). Uses spawn_blocking +
mmap_rayon instead of blocking the Tokio async worker.
- Parallelize thumbnail resize+encode with rayon par_iter() for the 3
sizes (Icon, Preview, Large) inside spawn_blocking.
Replace hardcoded DEFAULT_MAX_CONCURRENT_DECODES=4 with adaptive
max_concurrent_decodes() that uses available_parallelism()/2 (min 2).
Matches the pattern already used in image_transcode_service.
Improves thumbnail throughput on 16+ core servers by 2-4x.
Resolves conflict between main and PR #229 by applying timeout support
on top of main's drop(data) memory optimization. Changes:
- Add thumbnail_generation_ms to TimeoutConfig (default 30s)
- Add generation_timeout field to ThumbnailService
- Wrap spawn_blocking in tokio::time::timeout to prevent hanging
- Update DI to pass timeout from config
- Update tests to provide timeout parameter
https://claude.ai/code/session_015FD62aAoMYx1XBZbzPvUu8
Explicitly drop the encoded image buffer after decoding and extracting
EXIF orientation data. This reduces peak memory consumption during
thumbnail generation by the size of the original file.
The encoded data is no longer needed once the image is decoded into
a DynamicImage, but it was being held in memory until the end of the
spawn_blocking scope.
Fixes excessive memory consumption in thumbnail generation.
The database schema defines owner_id as UUID, but the Calendar entity
stored it as String. This caused a type mismatch error when creating
calendars via CalDAV clients like DAVx:
column "owner_id" is of type uuid but expression is of type text
Changes:
- Change Calendar.owner_id from String to Uuid
- Update new() and with_id() to accept Uuid
- Update owner_id() getter to return &Uuid
- Update belongs_to() to accept &Uuid
- Update calendar_storage_adapter to pass Uuid directly
- Update tests to use Uuid::new_v4()
Fixes#200
The Nextcloud Android client compares ETags before and after upload to
verify its write landed. OxiCloud was returning the stable file UUID as
the ETag, which never changed on content updates, causing false
SYNC_CONFLICT errors on every upload.
Five fixes applied:
1. Thread blob_hash (SHA-256) through File entity, FileDto, all read/write
queries, and all WebDAV/PROPFIND responses as the ETag — changes on
every content update, no DB migration needed.
2. Honor X-OC-Mtime header: parse the client-supplied mtime and use it
for updated_at via COALESCE(to_timestamp($n), NOW()).
3. Disable phantom checksum capability (preferredUploadType/supportedTypes)
that the server never actually implemented, stopping retry loops.
4. Add nc:creation_time and nc:upload_time to PROPFIND responses.
5. Return oc-etag header in chunked upload MOVE (assemble) responses.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace manual schema.sql application with sqlx's built-in migration
system. Migrations are embedded at compile time and tracked in the
_sqlx_migrations table. Pending migrations run automatically on startup.
- Move db/schema.sql → migrations/20260307000000_initial_schema.sql
- Remove apply_schema() and split_sql_statements() from db.rs
- Add run_migrations() using sqlx::migrate!() macro
- Remove docker-compose schema.sql mount (app handles it now)
- Enable sqlx "migrate" feature in Cargo.toml
Future schema changes: add a new timestamped .sql in migrations/.
Closes#190
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Fix all clippy lints (collapsible if, clone on Copy, needless borrow,
redundant bindings, unused params) and apply rustfmt across the codebase.
Update test mocks to match Uuid-based trait signatures.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace all 3 ImageFormat::WebP encode sites with JpegEncoder q=80.
Update fast-path to detect JPEG SOI instead of RIFF/WEBP magic.
Change file extension .webp -> .jpg, Content-Type headers, and
browser toBlob. Remove unused ImageFormat import and stale comments.
The webp feature stays for DECODING uploaded WebP images.
Schema (media_sort_date denormalization):
- Add media_sort_date column to storage.files with DEFAULT created_at
- Add trigger sync_media_sort_date: when file_metadata is upserted,
copies COALESCE(captured_at, created_at) into files.media_sort_date
- Add partial index idx_files_media_timeline on (user_id, media_sort_date DESC)
WHERE NOT is_trashed AND media type -- enables Index Scan + Limit (no Sort)
- copy_folder_tree now copies media_sort_date for copied files
- Remove dead idx_file_metadata_captured (no longer needed)
Query optimization (list_media_files):
- Rewrite to use fi.media_sort_date instead of COALESCE(fm.captured_at,...)
- Eliminates LEFT JOIN file_metadata -- one fewer table touch
- Plan: Limit to Index Scan O(LIMIT) instead of Sort O(N)
Video thumbnail pre-scaling (client + server):
- JS: pre-scale canvas to max 400px before toBlob -- 22x less RAM, 15x less BW
- Rust: fast-path in store_external_thumbnail -- if payload is already
WebP with dims within max_dim, store as-is (zero decode, zero encode)
Solution A — Cache-first path:
- get_thumbnail now tries moka (RAM) → disk BEFORE any DB query
- Ownership was verified at creation time; UUIDv4 prevents enumeration
- 146-photo timeline: 0 SQL queries instead of 146 per page load
Solution C — ETag / If-None-Match short-circuit:
- Deterministic ETag 'thumb-{id}-{size}' on every response
- If browser sends matching If-None-Match → 304 with zero I/O
- Cache-Control: immutable prevents even conditional requests
Additional improvements:
- Remove moka TTL (was 600s); thumbnails are immutable, weight-only eviction
- generate_all_sizes_background populates moka after disk write;
first GET after upload served from RAM with zero disk I/O
- DB path only taken on cache miss for images needing generation
Backend:
- get_thumbnail: for non-images, serve cached thumb or 204 (not 400)
- store_external_thumbnail: validate + re-encode to WebP + persist
- get_cached_thumbnail: memory → disk lookup without generation
- PUT /api/files/{id}/thumbnail/{size} endpoint for client uploads
- ThumbnailPort trait: add get_cached_thumbnail + store_external_thumbnail
Frontend (photos.js):
- On <img> error for video tiles, use <video> + <canvas> to extract
a frame at 25% duration using browser's native codec
- Show frame immediately via blob URL
- Fire-and-forget PUT to server for permanent caching
- Subsequent visits serve cached WebP instantly (no re-extraction)
Zero server-side dependencies — CPU distributed across clients.
- folder_db_repository: cast user_id to ::text in SELECT, ::uuid in INSERT/WHERE
- file_blob_write_repository: cast ::uuid in INSERT for user_id column
- file_blob_read_repository: bind Uuid natively instead of .to_string()
Fixes runtime errors:
- column user_id is of type uuid but expression is of type text
- operator does not exist: uuid = text
- Replace sql.chars().collect::<Vec<char>>() with direct byte-level
iteration over sql.as_bytes()
- Saves ~140 KB heap allocation for 35 KB schema (4× input size)
- Remove unused _start variable
- SQL is ASCII-safe so byte comparison is sufficient for all delimiters
Edouard Vanbelle