Milestone #127 step 2 (#1249). For You deliberately blends two
populations — a head of top-scored taste picks and a tail sampled from
deeper ranking (the freshness injection) — but the metrics judged it as
one blob, so its skip rate couldn't distinguish "the taste engine is
missing" from "the freshness tax is too high". That number decides the
exploration share before we tune it.
- Migration 0038: nullable pick_kind ('taste'|'fresh') on both
playlist_tracks (stamped at snapshot build) and play_events (frozen at
play-ingestion — the snapshot rebuilds daily, so attribution cannot be
reconstructed at read time).
- Builder: pickHeadAndTail marks head=taste / tail=fresh; the small-pool
fallback is all taste (top-N-by-score IS the taste mechanism). Other
variants persist NULL.
- Ingestion: for_you plays (live + offline replay) look the track up in
the user's current snapshot; not found → unattributed, never guessed.
- Metrics: For You's row gains a breakdown (taste / fresh / earlier
unattributed plays), parent row stays the sum; web card renders the
sub-rows indented with the same baseline deltas + low-data dimming.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
The recommendation metrics table was observable but not actionable: raw
source strings (album:<uuid> one-offs) drowned the stable surfaces, and
manual plays were excluded so skip rates had no control group.
- SQL: include NULL-source rows (the baseline) and carry completion_n
so family merges can weight avg_completion correctly.
- Handler buckets raw sources into stable families (radio:<uuid> →
Radio, album:/artist: → direct plays, etc.) grouped by surface
intent: go-to / discovery / direct — each band judged against its
job, since discovery mixes are expected to skip hotter. Families
under 20 plays are flagged low-confidence, not hidden.
- Settings card renders the baseline row and per-surface deltas in
percentage points vs baseline (worse-than-baseline deltas in danger
color), intent hint copy per group, low-data rows dimmed.
- Pure-unit test for the bucketing/merge; DB test updated to the new
contract (baseline included, radio:<uuid> collapse).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TsF3cNoKrqCYsU78cXC8U6
Per-source play outcomes so the operator can see whether each recommendation
surface is landing and tune the now-operator-tunable taste weights.
Server:
- query RecommendationSourceMetricsForUser: groups the user's play_events by
source (system-playlist surface), reporting plays / skips / avg completion
over a window; NULL-source (library/radio) plays excluded.
- GET /api/me/recommendation-metrics?days=30 (default 30, capped 365) →
{window_days, sources:[{source, plays, skips, skip_rate, avg_completion}]}.
- handler test: 401 unauth; per-source aggregation + NULL-source exclusion +
skip_rate / avg_completion math.
Web:
- lib/api/metrics.ts: query + friendly source labels.
- settings page gains a "Recommendation metrics" card (table of surface / plays
/ skip rate / avg completion), with loading/error/empty states.
- settings tests mock the new query (manual subscribe-store, hoisting-safe).
Note: You-might-like plays aren't source-tagged (it's a Home row, not a system
playlist), so this covers For-You / Discover / the mixes. Tagging YML plays
would be a client follow-up.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Three Go lint hits + two web test failures, all from the U3 push.
- internal/mailer/mailer.go: SentEmail.HtmlBody → HTMLBody (Go's
initialism convention; revive flagged); FakeSender.Send unused
ctx parameter renamed to _.
- internal/audit/audit_test.go: removed dead validUUID helper. It
was added speculatively in U1-T1 and never called by any test.
pgtype import stays — other tests use it.
- web/src/routes/settings/settings.test.ts: existing ListenBrainz
tests queried `getByRole('button', { name: /save/i })` which
worked when the page had only one Save button. The new Profile
card adds a "Save profile" button that also matches the regex,
triggering "found multiple elements". Anchored to /^save$/i for
exact-match. The newer Save profile tests still use
/save profile/i which is unique.
This is the same shape of test-fixture-lag I owe an answer for: I
landed new content that broke an existing test, and the existing
test had a too-loose selector. The right fix is to tighten the old
selector now (this commit) and to flag this pattern — selectors
that regex-match by partial words — as a candidate for the DRY
pass / test-utils consolidation.
Five fixes in one commit, all from the U1+U2+U3 push:
1. internal/api/library_test.go's Mount call wasn't updated when
U3-T4 added the mailer.Sender param. 17 args, sig wants 18.
Adds a trailing nil for the mailer. Same shape of test-fixture
lag the project has hit before — flagged as a recurring pattern
for the DRY-pass slice.
2. web/src/routes/settings/settings.test.ts mocked $lib/api/me with
`getAPIToken: vi.fn()` (no return value). The new /settings API
Token card's $effect calls `getAPIToken().then(...)` which threw
"Cannot read properties of undefined (reading 'then')" on every
ListenBrainz test. Default the mock to mockResolvedValue so any
test that doesn't override gets a valid resolution.
3. web/src/routes/admin/users/users.test.ts queried row buttons by
/^Delete$/ but T3 added per-user aria-labels ("Delete alice"),
so the row buttons' accessible name is no longer "Delete". The
modal's confirm button has no aria-label so its name IS "Delete"
exactly. Tests now click by per-user aria-label first, then by
/^Delete$/ for the modal confirm.
4. Same file: /Set password/i regex matched both the dialog's
submit button AND the row's "Reset password for ..." buttons
(because regex `Set` matches "ReSet" case-insensitively). Switched
to /^Set password$/ exact-match.
5. web/src/routes/reset-password/[token]/reset-password.test.ts had
/New password/i which matched both the "New password" and
"Confirm new password" labels. Switched to exact-match string
selectors.
6. web/src/routes/forgot-password/+page.svelte's onSubmit handler
had no catch — a rejected forgotPassword() bubbled as an
unhandled rejection in vitest. Added a silent catch since the
page intentionally shows the same success message regardless of
outcome (no-enumeration posture).
Completes the U3 frontend that T5a's crashed dispatch left half-done.
- /reset-password/[token] — public; new password + confirm. Calls
POST /api/auth/reset-password with the URL's token. invalid_token,
password_too_short, and mismatched-passwords surface as inline
errors. Success redirects to /login?reset=ok.
- /admin/integrations gains an SMTP card (alongside Lidarr) with
enabled toggle + all fields + Save and Send-test-email buttons.
Password input is a separate state from the loaded form, so empty
submission preserves the stored server-side value (matches the
backend's empty-password-preserves contract). Test button error
codes (no_email_on_file / not_configured / send_failed) surface
as actionable toasts.
Tests cover both new pages plus extensions to settings + integrations
test files for the cards landed in T5a.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds api.put helper, ListenBrainz API module with query/mutation
helpers, a /settings route with token + enable/disable UI, and 5
tests. Adds Settings to the Shell nav.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
bvandeusen