Commit Graph

8 Commits

Author SHA1 Message Date
bvandeusen fac5ae6ce5 feat(explore): reach dial to escape dense clusters + anti-revisit (#1476)
CI / lint (push) Successful in 2s
CI / frontend-build (push) Successful in 21s
CI / backend-lint-and-test (push) Successful in 30s
CI / integration (push) Successful in 3m50s
The Explore walk got stuck in dense signatures — neighbours all too similar, so
forward-arrow couldn't escape and Random was the only exit. Root cause: MMR only
diversifies WITHIN the nearest ~400 pool; in a dense cluster that whole pool is
near-identical, so there's no escape route in it.

- gallery_service.similar(reach=0.0, exclude_ids=None): reach>0 widens the pool
  (cap 400→1000) and _reach_sample strides across an outward-growing distance span
  so the set handed to MMR spans near→mid-far (guaranteed escape routes), not just
  the tight cluster. exclude_ids drops already-walked images. Gallery 'more like
  this' (reach=0) is unchanged.
- api/gallery similar: parse reach + exclude_ids.
- explore store: default reach 0.4 (auto-diversifies without touching the dial),
  pass the breadcrumb as exclude_ids, setReach action.
- ExploreView: a Near↔Far reach slider in the trail.
- tests: _reach_sample math (deeper ranks with higher reach, near kept); similar
  exclude_ids drops walked + reach path runs clean.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-13 12:06:22 -04:00
bvandeusen f8105046dc feat(explore): exclude WIP-tagged work from the Explore rabbit-hole
Explore's neighbour grid (/api/gallery/similar → gallery_service.similar) now
takes an Explore-only exclude_wip flag that drops `wip` system-tagged images
from the candidates, alongside the banner/editor presentation tags. The
gallery's own "similar" button is unchanged (keeps wip, #1274) — only the
Explore store passes exclude_wip=1. The anchor itself may still be a WIP; only
neighbours are filtered.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-07 22:10:05 -04:00
bvandeusen 6d314d662f feat(modal): applied tags drop from search instantly; manual add == accepting a suggestion
CI / lint (push) Successful in 4s
CI / frontend-build (push) Successful in 40s
CI / backend-lint-and-test (push) Successful in 2m50s
CI / integration (push) Successful in 3m53s
Three tag-flow gaps in the view modal (and the Explore workspace, which
shares TagPanel):
- the type-to-add dropdown now filters both its sections against the
  imageledger applied tags reactively, so a just-added tag disappears
  from search the moment the chip rail updates instead of after a
  modal refresh
- manually picking or creating a tag the model also suggested routes
  through the suggestion-accept flow: the acceptance is recorded for
  head training and the row leaves the panel, instead of the add
  silently bypassing the feedback loop
- removing a tag reloads the suggestion lists, so a model-suggested tag
  returns to the suggestions area (flagged rejected, one-click
  reversible) rather than vanishing until the next modal open

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CDgx8bQS5YrGRK76v8HUnM
2026-07-02 22:00:40 -04:00
bvandeusen 0f472b2f9e fix(explore): diversify "more like this" so it stops getting stuck (#1188)
CI / lint (push) Successful in 2s
CI / frontend-build (push) Successful in 20s
CI / backend-lint-and-test (push) Successful in 26s
CI / integration (push) Successful in 3m32s
Pure nearest-cosine piled near-identical images into the neighbour grid — a
reposted banner filled all 24 slots, and once you wandered into a B&W /
comic-panel cluster every neighbour was more of the same with no way back to
colour without the Random button (operator-reported, with screenshot).

similar() now over-fetches a wide candidate pool (5x the requested limit, cap
200), then diversifies down to `limit`:
- pHash near-duplicate collapse: drop candidates within 6 Hamming bits of the
  anchor or an already-kept candidate, so a repost (and the anchor's own clones)
  appears at most once.
- MMR re-rank: greedily pick for closeness-to-anchor minus similarity-to-already
  -picked (lambda 0.55), so the result SPANS clusters instead of returning 40
  variations of one image. Falls back to nearest-order on any failure / small
  pool, so existing nearest-first behaviour is unchanged when there's nothing to
  diversify.

Frontend forwardTarget drops the now-redundant skip-nearest-third hack (the list
is already diversified server-side) — plain random-over-unvisited gives the
variance now.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
2026-06-30 09:01:01 -04:00
bvandeusen 301f2de989 fix(explore): variance + no loop-back on → navigation (#94)
CI / lint (push) Successful in 2s
CI / frontend-build (push) Successful in 21s
CI / backend-lint-and-test (push) Successful in 26s
CI / integration (push) Successful in 3m26s
Two reports: → sometimes "loops back", and the walk gets stuck on near-identical
images. Cause: forwardTarget picked a uniformly-random neighbour from the 24
NEAREST, so it (a) often landed on an image already in the trail — which snaps
the cursor back into history and makes → bounce between visited nodes — and (b)
only ever offered near-duplicates.

forwardTarget now: excludes already-visited neighbours (→ opens something new,
no snap-back), and skips the closest third of the (similarity-sorted) pool so the
jump favours the more-varied remainder instead of lookalikes. Neighbour pool
widened 24→40 for more variety to browse + jump into. The post-← browser-forward
walk through visited crumbs is unchanged.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
2026-06-29 20:44:16 -04:00
bvandeusen 353b5d8087 feat(explore): ← / → keyboard navigation through the walk
CI / lint (push) Successful in 3s
CI / frontend-build (push) Successful in 19s
CI / backend-lint-and-test (push) Successful in 27s
CI / integration (push) Successful in 3m25s
Arrow keys walk the Explore breadcrumb trail: ← steps back, → goes forward to
an already-visited item or — with no forward history — jumps to a RANDOM
neighbour to keep the rabbit-hole going (operator-asked).

The trail gains a cursor (browser back/forward semantics): stepping back no
longer trims the forward branch, so → can return to it; a genuinely new walk
off a back-step truncates the stale branch then appends. The crumb-bar "current"
highlight follows the cursor, not the tip.

Arrows are ignored while typing a tag, but still navigate when the tag input is
focused-but-empty (it auto-focuses after every walk, so otherwise arrow-nav
would dead-end after one step). Modifier-key combos pass through untouched.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
2026-06-28 21:41:23 -04:00
bvandeusen 2d1cddd9b7 feat(explore): 3-pane tagging workspace — gallery | viewer | tag rail
CI / lint (push) Successful in 3s
CI / frontend-build (push) Successful in 19s
CI / backend-lint-and-test (push) Successful in 30s
CI / integration (push) Successful in 3m18s
Reworks Explore from "anchor + neighbour grid + cluster tag-gap rail" into a
persistent 3-pane workspace that unfolds the image modal so you can tag while
rabbit-holing (operator concept 2026-06-26):

- LEFT  neighbour grid (larger thumbs), click = walk; breadcrumb retained.
- CENTER light viewer — reuses ImageCanvas + ImageMetaBar(:image) for the
  focused image; "Open full viewer" still launches the overlay modal.
- RIGHT  the modal's TagPanel, hosted on the anchor for modal-parity tagging
  (chips, autocomplete, suggestions + Accept, fandom-on-chip, T/"/" focus).

Reuse without destabilising the audited modal store: TagPanel and
SuggestionsPanel gain an optional `host` prop (default = modal store, so the
image modal is unchanged); the explore store implements the same small
tag-CRUD surface (current/currentImageId + reloadTags/addExistingTag/
removeTag/createAndAdd) over the anchor. ImageMetaBar gains an optional
`image` prop for the same reason.

Drops the mass/cluster tagger (TagGapPanel deleted; clusterIds/thumbById
removed) — per-image tagging feeds the per-tag reference-embedding centroid
better than bulk ops.

Nav: keep the Explore tab but bare /explore now SEEDS a random image
(GET /api/showcase?limit=1 → /explore/:id) so the tab kick-starts a rabbit
hole; explicit meta.navOrder pins nav order (Explore after Gallery) since
router.getRoutes() doesn't preserve declaration order.

Note: the backend cluster tag-gaps route/service (#94a) is now frontend-orphaned
— left in place; flag for a separate cleanup.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-26 01:15:11 -04:00
bvandeusen 7b712920a4 feat(explore): Explore view + tag-gap closing + modal meta/download (#94b–d, #4a/b)
CI / lint (push) Successful in 2s
CI / frontend-build (push) Successful in 20s
CI / backend-lint-and-test (push) Successful in 26s
CI / integration (push) Successful in 3m17s
Cluster C frontend, milestone #94.

#94b Explore walk: new /explore/:imageId route + ExploreView + explore store.
Anchor (reuse /api/gallery/image), neighbour grid (reuse /api/gallery/similar,
24), click a neighbour to re-anchor; in-memory breadcrumb that trims on
backtrack (route is the source of truth). Empty/loading/error + no-embedding
states.

#94c tag-gap closing: components/explore/TagGapPanel — fetches
/api/images/cluster/tag-gaps for the anchor+neighbours, a consensus-threshold
slider (default 60%), per gap shows present/total + the missing thumbnails +
'Apply to N missing' → /api/tags/images/bulk/tags (source manual) → re-fetch.

#94d entry points: 'Explore' button in the modal RelatedStrip; the TopNav entry
comes free from the route's meta.title.

#4a metadata HUD + #4b split Download: new modal ImageMetaBar (always-on, above
ProvenancePanel) shows dimensions/size/type and a split Download button
(default Download, chevron → Copy link via utils/clipboard — no clipboard-image,
rule 95).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XCUHUGQLrBrkgyk1t49kpX
2026-06-23 02:08:34 -04:00