Store confidence thresholds per tagger model so JoyTag no longer inherits WD tuning. Refresh the active threshold when switching models, guard stale threshold saves, and keep UI Lab mocks in sync.
Also tightens the onboarding model selector so the segmented control no longer stretches across the row.
Adds a fullscreen image-only slideshow from the current lightbox collection, with pause, keyboard navigation, hidden idle controls, and polished image transitions.
Adds slideshow duration and playback order settings, including random order support.
Add a reset_ai_tags backend command that removes AI-generated tag rows, clears AI tagging metadata, cancels active tagging jobs, and drops queued or failed tagging jobs in the selected scope.
Expose reset actions in AI Workspace and the Explore tag manager, refresh gallery/progress/tag state after reset, and add subtle AI source indicators to tag manager rows.
Pass the active colour filter through tag searches and apply the existing palette match inside the tag query and count paths.
Update the UI Lab mock backend so colour filtering behaves the same way when testing tag search results.
The visual k-means cluster feature was confusingly named tag_cloud / tagCloud /
TagCloud across the whole stack, while the actual tag list is explore_tags — the
two were trivially easy to mix up (and did cause confusion). Rename the cluster
side to visual_cluster / visualCluster / VisualCluster everywhere: command
get_tag_cloud -> get_visual_clusters (+ lib.rs registration and the invoke
string), VisualClusterEntry, the store fields/actions/tokens, and the mock
backend. Old names are retired rather than reused, so any missed reference fails
loudly instead of silently resolving to the wrong concept.
The tags side keeps its accurate explore_tags naming, and the user-facing
"Tag Cloud" UI label is unchanged.
Also rename the SQLite tag_cloud_cache table -> visual_cluster_cache (the old
table is dropped during schema setup — it is a disposable cache already
invalidated by the clustering version bump) and the TagCloud.tsx component file
-> ExploreView.tsx, since it is the Explore container hosting both the cluster
and tag views.
Computing visual clusters was O(n·k·dim) per Lloyd iteration over the whole
library, single-threaded — several seconds on an 80k-image library on first
view. Find centroids on a deterministic, evenly-strided sample (<=3000
embeddings) and then assign every image to its nearest centroid in one parallel
rayon pass. Libraries at or below the sample cap are unchanged.
Replace the greedy farthest-point seeding (which seeds outliers, leaving the
dense core under-represented on a sample so one centroid absorbed tens of
thousands of generic images) with proper density-aware k-means++ D² seeding,
made deterministic via a small fixed-seed SplitMix64 PRNG. This keeps clusters
balanced on large libraries.
A CLUSTER_CACHE_VERSION is folded into the tag-cloud cache key so existing
caches computed by the old algorithm are invalidated and recomputed. The
clustering timing line and the cache-write failure now go through the `log`
facade (debug/warn) instead of eprintln.
The get_tag_cloud cache key was built by loading and hashing every embedding
blob for the scope *before* checking the cache, so even a cache hit re-read
hundreds of MB on large libraries and stalled Explore for several seconds.
Validate the cache from a lightweight image-ID-set signature plus the embedding
revision instead, so a hit never loads embeddings. The ID-set hash keeps the key
membership-sensitive (add/remove/move between folders) and the revision covers
an image being re-embedded in place. Cache write failures are now logged rather
than silently ignored.
On the frontend, switching folders (or re-entering Explore) no longer leaves the
previous folder's clusters/tags on screen with no loading indicator:
loadTagCloud/loadExploreTags clear stale entries on a real folder switch. The
displayed folder is tracked separately (exploreTagsShownFolderId) from the
cache-dirty marker so a same-folder invalidation (tag edits, new AI tags) does
not masquerade as a switch and wipe the visible list mid-refresh.
Worker pause states can optionally survive app restarts. A new toggle in
Settings saves the current pause map to settings/worker_pauses.json; on
startup lib.rs restores it before workers are spawned. Backend: new
snapshot/replace helpers in indexer.rs, persist functions in commands.rs
(get/set_worker_pauses_persist). Frontend: workerPausesPersist store
field, load/setWorkerPausesPersist actions, toggle in SettingsModal.
Hovering a tag in Explore now loads and displays co-occurring tags as a
weighted cloud. New `get_related_tags` SQL self-join (db.rs/commands.rs),
`loadRelatedTags` store action with per-folder keyed cache, and TagCloud
atlas UI with ResizeObserver-driven layout and RAF animation. Explore tag
limit raised to 180; tag cloud auto-refreshes 700ms after new AI tagging
completes.
Introduces a selectable tagging model so Phokus isn't locked to the
anime-leaning WD tagger. JoyTag uses the Danbooru schema but generalizes to
photographic content and is strong on NSFW concepts — a better fit for a photo
manager while keeping the explicitness range.
- `TaggerModel` enum (wd | joytag) persisted as a `tagger_model` setting;
`model_dir`, status, and download now follow the active model (per-model
repo/dir/file list). WD stays the default.
- `Tagger` trait implemented by `WdTagger` and the new `JoyTagger`;
`create_active_tagger` builds the selected one. The worker holds
`Box<dyn Tagger>` and rebuilds on model change (TAGGER_SESSION_DIRTY).
- Shared `assemble_batch` skeleton (pack -> one forward pass -> per-image
fallback, results in input order); both providers and the shared
decode/pad/resize are de-duped onto common helpers.
- JoyTag specifics: NCHW + RGB + CLIP-normalized input (vs WD's NHWC/BGR/raw),
flat top_tags.txt labels, logits -> sigmoid -> threshold (default 0.4). It has
no native rating, so the explicitness rating is derived from its NSFW tags.
- Tags are attributed to the model that produced them (ai_tagger_model),
via Tagger::model_name, instead of a hardcoded WD constant.
- New get/set_tagger_model commands.
Backend only; the Settings model picker and end-to-end testing against the
JoyTag model files come next.
Security / robustness / a11y polish on top of the color-search + tooltips work:
- URL opening: route through validated backend commands (open_map_location
with lat/lon bounds-checking, open_changelog_url with a fixed URL) instead
of the frontend opener:allow-open-url capability, which is now removed.
- EXIF GPS parsing: validate coordinate ranges and require the correct
N/S/E/W hemisphere ref byte, then clamp.
- Guard double-submit on album create / add-to-album (Lightbox, BulkActionBar,
Sidebar) and discard stale autocomplete responses in the bulk tag editor.
- Gallery tile: stop nesting <button>s — non-interactive tile div with an
overlay button for open/toggle; checkbox and Similar promoted with z-index
+ focus rings.
- Accessibility: keyboard handlers, focus-visible rings, and aria on album
rows, tag-manage actions, and tile controls; Tooltip uses a block <div>
wrapper in block mode and aria-hidden when hidden.
- Show the destination URL in a tooltip on the "Full changelog" link so the
user can see where it goes before clicking.
- Toolbar "All" filter also clears the color filter; color-filtered views no
longer get unfiltered newly-indexed images injected.
- Sidebar reorder: bail if the album set changed mid-drag.
- Tooling: add cargo fmt scripts; alphabetize package.json scripts.
Add dominant-color palette extraction, storage, filtering, and startup backfill so the gallery and Timeline can be filtered from the toolbar color picker.
Introduce a reusable tooltip component and migrate the color filter, update indicator, and gallery filename hover affordances to it.
Similar search scoping:
- Add current_album as a similar-scope option and remember the source album when similar or region searches are launched from an album.
- Route gallery and lightbox similar actions through scope-aware store helpers so Album/Folder/All choices are applied consistently.
- Keep pagination and scope toggles working for both whole-image and region-search result sets.
Backend filtering:
- Extend find_similar_images and find_similar_by_region params with album_id, giving album scope precedence over folder scope.
- Add album_membership filtering for HNSW whole-image search and brute-force crop embedding search.
EXIF info panel:
- New on-demand get_image_exif command (kamadak-exif) returning camera/lens/
aperture/shutter/ISO/focal-length and decimal GPS; read from the file when the
lightbox opens (no DB schema change, works on already-indexed images).
- Lightbox shows a Camera panel; GPS opens the location in the browser via
OpenStreetMap (adds opener:allow-open-url capability).
Tag management:
- Backend rename_tag (rename, or merge when the target exists) and delete_tag
(library-wide), both clearing the tag-cloud cache; store actions invalidate
tag caches, refresh Explore, and re-point/refresh an active tag-search.
- Explore -> Tag Cloud gains a Manage mode: a flat list with per-tag rename/
merge/delete.
Albums:
- Drag-to-reorder in the sidebar via framer-motion Reorder with a hover handle;
order persists through the existing reorder_albums command.
Reads live store order on drag end and robustly derives GPS hemisphere from raw
EXIF ref bytes (review follow-ups). CHANGELOG updated.
Albums (manual collections):
- New albums/album_images tables with FK cascades; DB functions and Tauri
commands for create/rename/delete/delete-many/reorder/list, add/remove
images, and paginated get_album_images.
- Distinct sidebar "ALBUMS" section with cover thumbnails, create/rename/
delete, and a Manage multi-select mode for bulk album deletion.
- Album view reuses the gallery grid (activeView "album" + selectedAlbumId);
spans folders; add from the bulk bar or the lightbox, remove from within.
Gallery multi-select + bulk actions:
- gallerySelectedIds selection model with a top-left corner checkbox that
reveals on corner hover; click-to-toggle in selection mode, double-click
to open.
- Floating BulkActionBar: tag (inline autocomplete popover), rating,
favorite, add-to-album, and a delete with an explicit "from disk"
confirmation. Batch commands bulk_update_details/bulk_add_tags/
bulk_remove_tag.
Also:
- Duplicate Finder delete now requires confirmation with clear "from disk"
wording (was single-click fire-and-forget).
- CPU/CUDA build-variant badge in Settings (get_build_variant).
- Rating/favorite no longer re-sorts derived collections (similar/region/
semantic/tag/album results); single and bulk paths replace in place there.
- Album-aware indexed-images/media-updated handlers so thumbnails paint and
newly-indexed files don't leak into an album view.
- CHANGELOG updated.
Post-update UX for the next release (0.1.2):
- What's New: after a version change, greet the user with a toast that opens
an in-app release-notes screen (collapsible Added/Changed/Fixed sections)
sourced from the bundled CHANGELOG.md, reopenable from Settings -> Updates.
Backed by a last_seen_version settings file (get/set_last_seen_version
commands) that distinguishes upgrades from fresh installs.
- Updater: the download/install progress toast now reappears when an update is
started from the title-bar indicator or Settings after the prompt was
dismissed; Settings -> Updates gains a real progress bar with a percentage.
- Light theme: fix the recurring subtle-light breakage. Neutral surfaces now
rely on the CSS-variable remap instead of light-theme:bg-white (which forced
surfaces dark because --color-white is remapped dark); the green action
buttons drop the broken light-theme:hover:bg-emerald-200 (remapped dark,
unreadable on hover) for an override-free emerald-500 tint that auto-themes,
across the updater, What's New, and onboarding.
- Debug panel: add What's New triggers (toast / modal / reset).
Derive default implementations for captioner and tagger option enums, simplify sorting and progress multiple checks, and remove redundant iterator conversions.
Replace native add-folder dialogs with an in-app folder picker that supports collecting folders from multiple locations before adding them together.
Add backend directory listing and batch add commands with duplicate skipping, plus store actions and a themed picker UI with a dedicated folders-to-add panel.
Drops and recreates the sqlite-vec tables at the current model dimension, then
re-queues every image for embedding. Fixes "dimension mismatch" search errors
that occur when the vector table was built for a different model/dimension (e.g.
after experimenting with 768-dim models against this 512-dim build).
The rebuild runs under the embedding worker's DB write lock and resets the job
queue inside a single transaction, so it can't interleave with an in-flight
embedding batch (per code review).
Add a failed-tag discovery flow for background worker failures.
Changes:
- Add a Failed Tags toolbar filter that appears when tag failures exist.
- Add Locate buttons for failed tag tasks in the background worker prompt.
- Route Locate to the affected folder and filter the gallery to images with tagger errors.
- Fetch and display failed tag filenames/errors in the expanded worker details.
- Add a backend query and gallery filter flag for images with failed AI tagging.
- Improve subtle-light contrast for failed worker chips, filenames, and Locate/Retry buttons.
- Also slightly increases the title bar update indicator pulse size for better visibility.
Timeline:
- Add a right-edge scrubber (year labels + month dots) that jumps to any
period; runs in the same direction as the scrolled content
- Load the full filtered set in Timeline view so the scrubber spans the whole
library instead of just the first page
Gallery & UI:
- Virtualise the gallery grid
- Folder reordering in the sidebar (drag + persisted sort_order) and a themed
dropdown component
- Subtle Light theme contrast fixes for toggles, secondary buttons and the
update toast
Embedding workers now defer video jobs without a thumbnail at claim time and
requeue any previously-failed deferred jobs on startup, so videos no longer
churn through failed embeddings.
QoL polish across BackgroundTasks, DuplicateFinder, Lightbox and VideoPlayer.
Phase 5 of the 0.1.0 release prep:
- FFmpeg no longer blocks startup (or crashes the app offline): provisioning
runs in a background thread emitting throttled ffmpeg-progress events,
with installed/downloading/failed state queryable via get_ffmpeg_status
and a retry command
- video jobs are invisible to claiming and tier-priority checks until FFmpeg
is ready, so they wait as pending without failing — and without starving
image embeddings/tagging (include_videos gating in db.rs)
- 7-step show-don't-tell onboarding wizard: welcome + live FFmpeg progress,
real first-folder picker, faked animating pipeline bar, placeholder
gallery tiles, cycling search-syntax demo, views overview, and an AI
features step with a real opt-in tagger download
- skippable at every point (Escape included), persisted via
settings/onboarding_completed.txt, re-runnable from Settings > General,
which also gains an FFmpeg status/retry row
Auto-fix pass (uninlined format args, needless borrows, to_vec,
size_of_val, map_err->inspect_err) plus hand fixes: fully annotated the
sqlite-vec transmute (c_char for cross-platform correctness), type alias
for the duplicate-scan tuple, slice signatures over &mut Vec/&PathBuf,
and #[allow(dead_code)] documenting the intentionally dormant caption
worker. The CI clippy step now fails on any new warning.
Phase 4 of the 0.1.0 release prep:
- tauri-plugin-log: rotating file log (5 MB, Info) in the app log dir plus
stdout; all 54 backend println!/eprintln! sites migrated to log macros so
release builds finally produce diagnostics
- tauri-plugin-single-instance (registered first): second launches focus the
existing window instead of racing the same SQLite DB with a second worker
fleet; tauri-plugin-window-state persists window geometry
- real CSP replacing null (scripts self-only, media/images via the asset
protocol, IPC endpoints whitelisted, object-src none) with a devCsp
variant for Vite HMR
- asset protocol scope narrowed from '**' to thumbnails statically plus
per-folder runtime grants (setup, add_folder, update_folder_path)
Panic audit (plan item) concluded with no changes: worker loops already
catch and log all Result errors; remaining unwraps are startup fail-fast,
poisoned-lock convention, infallible-by-construction, or test code.
Replace the [scanned, total] tuple progress event with a structured
{phase, processed, total, skipped} payload covering all three scan
phases (checking, hashing, confirming). find_duplicates now returns a
DuplicateScanResult with scanned/candidate/skipped counts, and the UI
surfaces phase labels plus a warning when unreadable files were
skipped. Also adds a clean:app script for cargo clean.
Debounce embedding/tagging completion notifications per folder (6s window)
so rapid file additions from downloaders fire one notification instead of
one per file. Add per-folder mute via the sidebar context menu and a global
pause toggle in Settings > General, both persisted across restarts.
Delete thumbnails alongside DB rows in all three removal paths (watcher
detect, delete-from-disk command, remove-folder command) so orphans no
longer accumulate passively.
Intercept RenameMode::Both watcher events and handle them as in-place
path updates: renames the thumbnail file to the new hash and updates the
DB row without touching the embedding, avoiding a full rebuild on move.
Falls back to thumbnail regeneration if the file rename fails.
Scans the thumbnails directory and cross-references against thumbnail_path
values in the images table, deleting any files not linked to an indexed image.
Surfaces count and reclaimable MB in the General section alongside the
existing Compact database card.
Adds get_database_info and vacuum_database Tauri commands. The General
section now shows current DB size and reclaimable space on load, with a
Compact now button that runs PRAGMA wal_checkpoint(FULL) + VACUUM and
reports how many MB were freed. Button disables automatically when the
database is already compact (< 0.5 MB reclaimable).
- Remove Workers section (read-only status, belongs in background tasks panel)
- Remove Captioning coming-soon placeholder
- Add General section with Open data folder button (tauri-plugin-opener)
- Persist queue scope and folder selection across sessions via settings files
- Show inline validation errors on threshold/batch size inputs instead of silent revert
- Fix acceleration save errors appearing in wrong card
- Disable folder selection controls when target scope is All media
Monitors all registered folders using OS-native events (ReadDirectoryChangesW
on Windows) so new, modified, and deleted files are reflected in the gallery
automatically without manual reindexing.
Key design points:
- Adaptive blocking: recv() when no events pending (zero CPU), switches to
recv_timeout(earliest_deadline) only when debounce timers are running — wakes
exactly when the soonest event is ready, no busy-polling
- 500 ms per-path debounce coalesces rapid OS event bursts into one action
- Change detection preserved: build_record skips upsert if file_size + mtime
unchanged, preventing spurious thumbnail/embedding re-queues and avoiding
clobbering of existing metadata like thumbnail_path
- Access events (reads) filtered out; only Create/Modify/Remove trigger work
- Deletion path: emits watcher-deleted event with image IDs; frontend removes
those images from state and clears selectedImage if it was deleted
WatcherHandle stored in app state; add_folder / remove_folder / update_folder_path
commands keep the watched path set in sync with the DB.
Adds a full discovery and organisation layer to the gallery.
## AI Tagger
- WD tagger (ONNX via ort) with DirectML/CPU acceleration
- Per-image and per-folder tag queuing; configurable batch size and
confidence threshold; model downloaded on first use via ureq/zip
- Tags stored with source ('ai' | 'user'); user tags are never
overwritten by AI; AI tags protected from accidental promotion
- Tagger pause/resume per folder; in-flight batches discarded cleanly
on pause or cancellation without leaving jobs stuck in processing
## Semantic & Tag Search
- CLIP text-query embedding via candle (HuggingFace hub model)
- Progressive candidate doubling with filter post-processing so
folder/rating/media-kind filters do not silently under-return results
- Tag search with DB-backed pagination (offset + COUNT total)
- Filename search unchanged; prefix syntax: s:, t:, f: in search bar
## Similarity Search
- HNSW in-memory index (hnsw_rs) with monotonic embedding_revision
counter for cache invalidation; build_index retries if a concurrent
write advances the revision during construction
- Folder-scoped similar search uses brute-force cosine scan (no k limit)
- Region-based similarity: crop an arbitrary rectangle in the lightbox,
embed it with CLIP, find the nearest images globally or within folder
- Pagination for both similar and region results; scope toggle
(all media / current folder) re-runs the query without reopening image
## Duplicate Finder
- Three-phase detection: stat (size) → sample hash (4×16 KB windows)
→ full-file hash (xxh3, large files only) to eliminate false positives
- Filesystem-first deletion: only removes DB rows for files successfully
deleted from disk; failed files remain visible for retry
- Persisted scan cache (SQLite) with invalidation on reindex and deletion;
both global and per-folder scopes invalidated after any deletion
## Tag Cloud & Explore
- Visual cluster explore: k-means over CLIP embeddings, configurable k,
representative thumbnail per cluster; cache keyed on xxh3 of embedding set
- Tag explore: ranked tag list with image counts and representative
thumbnail per tag; invalidated when AI tagging completes or folder removed
- Tag autocomplete for search bar
## Folder Management
- Inline rename (display name only, not OS rename)
- Relocate: file-picker to update folder path; all image paths rewritten
using SUBSTR prefix replacement to avoid corrupting paths that contain
the folder name as a substring
- Missing-folder recovery banner: Locate or Remove, never silent deletion
- Right-click context menu on sidebar items: Reindex, Rename, Locate,
Remove; hover buttons preserved alongside context menu
## Background Tasks
- Per-folder progress panel with embedding, tagging, caption, and
thumbnail stage breakdown
- Retry button per task for failed embeddings and tagging jobs
- Completed-task notifications (system tray via tauri-plugin-notification)
- Scan errors shown inline; WalkDir permission errors protect existing
records from deletion rather than cascading data loss
## Backend correctness
- sqlite-vec DML performed outside transactions throughout (virtual-table
limitation); embedding revision incremented on delete as well as insert
- Tagging job discard checks status = 'processing' so paused jobs reset
to 'pending' are also skipped, not just cancelled ones
- Stale async responses in Lightbox guarded with cancellation flags and
currentImageIdRef for both tag-load and tag-add callbacks
- Duplicate cache cleared on reindex; folder-removal clears vector rows
outside transaction before deleting the folder row
- Fix root cause: embedding_source_path() returns Result<PathBuf>, returning
Err for videos without a thumbnail instead of silently falling back to the
raw .mp4 path that CLIP cannot decode
- indexer: split embedding batch into pre-failed (no source) and embeddable
jobs; pre-failed are marked immediately without hitting the CLIP model
- db: retry_failed_embedding_jobs skips videos still without a thumbnail so
they no longer re-fail immediately on retry
- Add get_failed_embedding_images command listing failed files + error per folder
- Gallery: amber warning badge on tiles with embedding_status = 'failed'
- BackgroundTasks: fetch and show failed filenames/errors in expanded panel
- Toolbar: conditional 'Failed Embeddings' amber filter pill shown when any
folder has embedding_failed > 0; filters at DB level via new
embedding_failed_only param on get_images / count_images
- TagCloud: replaced vocabulary/dictionary label system with representative
image thumbnails per cluster; results cached in SQLite by image-id hash
Introduces an Explore view with a tag cloud that clusters image embeddings
using cosine k-means and labels clusters via vocabulary-nearest-neighbour
CLIP matching. Vocabulary embeddings are disk-cached (FNV hash-keyed) to
avoid redundant inference. Enables CUDA for candle dependencies and adds a
build.rs check that surfaces a clear error when the toolkit is missing.
- add semantic text search with toolbar mode switching and sqlite-vec query support
- improve embedding progress visibility, failure recovery, and similar-image affordances
- add search clearing and keyboard controls for filename vs semantic search modes
- refine background task interactions and gallery/lightbox embedding states
Refs: #4
- add Candle + HF Hub CLIP image embedding pipeline with background embedding worker
- write image embeddings into sqlite-vec and expose similar-image lookup through a new backend command
- surface embedding progress and recovery in the UI, including retries for failed embeddings
- improve gallery/lightbox embedding UX and make similar-image actions directly accessible
Refs: #3, #4
Sets up Tauri v2 + React frontend with SQLite metadata store,
sqlite-vec for CLIP embedding infrastructure, and r2d2 connection
pool replacing the single Arc<Mutex<Connection>>. Indexer now uses
rayon for parallel file metadata collection.
LyAhn