- `_SOULSYNC_BASE_VERSION` 2.4.2 → 2.4.3
- helper.js — flip 2.4.3 WHATS_NEW header to "May 8, 2026 — 2.4.3
release"; bump fallback default from 2.4.2 → 2.4.3
- docker-publish.yml — manual-trigger default tag 2.4.2 → 2.4.3
Drive-by — make sidebar version + version-modal subtitle dynamic.
The sidebar version button (`v2.4.1`) and version-modal subtitle
(`Version 2.4.1 — Latest Changes`) were hardcoded text in the HTML.
2.4.2 shipped without these getting bumped — silent drift, easy to
miss at every release.
Added a Flask context_processor that injects `soulsync_version` and
`soulsync_base_version` into every template, then templated the two
hardcoded values:
v{{ soulsync_base_version }}
Version {{ soulsync_base_version }} — Latest Changes
Now bumping `_SOULSYNC_BASE_VERSION` updates the UI everywhere it's
rendered. No more "I forgot to bump the sidebar" at release.
2232/2232 full suite green. Ruff clean. JS parses clean.
@ -3414,8 +3414,8 @@ function closeHelperSearch() {
// release time and add a real `date:` line at the top of the version block.
constWHATS_NEW={
'2.4.3':[
// --- post-2.4.2 dev work — entries hidden by _getLatestWhatsNewVersion until the build version bumps ---
{date:'Unreleased — 2.4.3 dev cycle' },
// --- May 8, 2026 — patch release ---
{date:'May 8, 2026 — 2.4.3 release' },
{title:'Discover: Sharper Track Selection (Diversity, Source-Aware Popularity, Library Dedup, SQL Genre Filter)',desc:'four selection-quality fixes on the soulsync-made discover playlists. (1) hidden gems and discovery shuffle had no diversity caps — they could return 50 tracks from the same artist or 20 from one album. now both apply the existing `_apply_diversity_filter` (over-fetch 3x then enforce per-album/per-artist caps; shuffle uses tighter caps because it should feel maximally varied). (2) `popularity` thresholds were spotify-shaped (0-100 scale, popular >= 60 / hidden < 40), but deezer writes its rank value into that column (often six-digit integers) and itunes writes nothing meaningful. for deezer-primary users this meant popular picks pulled essentially everything and hidden gems pulled nothing. new `_get_popularity_thresholds(source)` returns per-source values: spotify (60, 40), deezer (500_000, 100_000) ballpark, itunes/other (None, None) which skips the popularity filter entirely and falls back to random + diversity. (3) `get_genre_playlist` used to load up to 1M discovery_pool rows into python and run a substring keyword filter on the json column. now the keyword OR chain pushes down into sql via `(artist_genres LIKE ? OR ...)` placeholders, fetch_limit drops to limit*10. parent-genre expansion via GENRE_MAPPING preserved. (4) discovery selectors now exclude tracks the user already owns — `_select_discovery_tracks` gained `exclude_owned: bool = True` (default on) which adds a `NOT EXISTS (SELECT FROM tracks WHERE source_id matches)` correlated subquery covering the spotify/itunes/deezer-id columns (with the deezer-column-name asymmetry handled inline: discovery_pool.deezer_track_id vs tracks.deezer_id). hidden gems / shuffle / popular picks / decade / genre browser all benefit automatically. 12 new tests (27 total in the file): diversity caps, source-aware threshold values, threshold-skip behavior, sql-pushed genre filter, parent-genre expansion, owned-track exclusion, opt-out flag, and the deezer-column-name asymmetry trap. 2232/2232 full suite green.',page:'discover'},
{title:'Discover: Stop Showing Undownloadable Tracks (+Lift +Cleanup)',desc:'audit found multiple discover-page sections (hidden gems / discovery shuffle / popular picks / decade browser / genre browser) had no `WHERE (spotify_track_id IS NOT NULL OR itunes_track_id IS NOT NULL OR ...)` gate on their selection sql. tracks with no source ids in the discovery pool were getting displayed, the user would click download, and the download would silently fail because there was nothing to look up. fix: lifted all five discovery_pool selection methods into shared private helpers (`_select_discovery_tracks`, `_apply_diversity_filter`, `_compute_adaptive_diversity_limits`) on `PersonalizedPlaylistsService`. mandatory id-validity gate is hard-coded into the selector — no opt-out flag, every public method inherits it for free. behavior preserved: same diversity tiers, same over-fetch multipliers, same popularity thresholds, same blacklist filter. ~314 lines of repeated select/diversity boilerplate collapsed across the 5 methods (-55% on those methods\' business logic). also deleted four sections that had been stubbed returning [] for ages (recently added / top tracks / forgotten favorites / familiar favorites) — frontend, backend endpoints, html blocks, helper docs, all gone. on the frontend, lifted the duplicated decade-browser + genre-browser tab management (~314 lines of identical fetch-tabs / render-tabstrip / fetch-content / render-tracklist / wire-sync-button code) into one shared `createTabbedBrowserSection(config)` helper. each browser is now a thin wrapper: ~3 lines per public function. 14 new tests pin the gate (every selector filters null-id rows), the diversity caps, the adaptive limit tiers, the source filter, and the blacklist filter.',page:'discover'},
{title:'Internal: Discover Controller — Cin Pre-Review Polish',desc:'tightened the controller before opening the PR. (1) dropped the magic `extractItems` defaults — controller used to auto-pull `data.items` / `data.albums` / `data.artists` / `data.tracks` / `data.results` if no extractor was provided. removed the fallback chain. each section now MUST supply its own `extractItems(data) => array` callback. cin standard: explicit > implicit; the auto-fallback could silently grab the wrong key on endpoints that return multiple arrays. validated at register-time so misuse fails immediately. all 10 existing call sites already had explicit extractors so no migration churn. (2) replaced the `renderItems` returning null convention (used by Your Albums + manualDom-style sections) with an explicit `manualDom: true` config flag. clearer intent at the call site, less likely to be confused with a renderer error. (3) added a minimal node `--test` JS test file at `tests/static/test_discover_section_controller.mjs` — 32 tests pin the lifecycle contract: config validation (every required field), happy-path fetch+render, empty/stale/error states, no-fetch `data:` mode, manualDom mode, callable `fetchUrl`, load coalescing, refresh bypass, hook error containment, error toasts. runs via `node --test tests/static/` directly, OR via the regular pytest sweep (`tests/test_discover_section_controller_js.py` shells out to node and asserts a clean exit). skipped gracefully when node isn\'t available or is < 22. closes the "controller is a contract, pin it at the test boundary" gap that cin would have flagged. 2205/2205 full suite green (was 2204 + 1 new pytest wrapper); 32/32 node --test pass; ruff clean; js parses clean.',page:'discover'},
@ -4292,7 +4292,7 @@ function _getLatestWhatsNewVersion() {
Broque Thomas
1 month ago