feat(series): assisted-continuation matcher + suggestion queue — backend (FC-6.3)
CI / backend-lint-and-test (push) Successful in 26s
CI / frontend-build (push) Successful in 27s
CI / integration (push) Successful in 3m8s
CI / lint (push) Successful in 3s

Confirm-only "this post may continue this series" matcher.

- series_suggestion table (post_id, series_tag_id, score, signals jsonb, status
  pending|added|dismissed, UNIQUE(post,series)); migration 0041 + two settings
  knobs (series_suggest_enabled, series_suggest_threshold).
- series_match_service: weighted additive score (title-stem / same-artist /
  page-continuity / shared-distinctive-tags), no single signal gating. The title
  "pattern" is derived on the fly from the post titles already in a series, so it
  sharpens as more are confirmed (no persisted state to drift). Candidates are
  bounded to the post's artist. match_post upserts pending suggestions (UNIQUE +
  on-conflict, respecting prior added/dismissed decisions).
- accept reuses add_post_as_chapter then marks 'added'; dismiss marks 'dismissed'.
- rescan_series_suggestions_task: settings-gated, time-boxed + self-resuming from
  a post-id cursor (maintenance_long lane), like normalize_tags_task.
- API: GET /series/suggestions, POST .../<id>/accept|dismiss, POST .../rescan.
- Settings: enabled + threshold exposed via /settings/import.
- Tests: pure scoring helpers + matcher/accept/dismiss/rescan lifecycle + UNIQUE
  dedup.

Frontend (Suggestions tab + settings card) lands next.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-06-07 18:58:18 -04:00
parent 9e262cc5f0
commit c0fd80e694
10 changed files with 785 additions and 0 deletions
+51
View File
@@ -147,3 +147,54 @@ def normalize_tags_task(self) -> dict:
)
normalize_tags_task.delay()
return summary
# Time-box one rescan chunk well under the soft limit and re-enqueue from the
# cursor — scoring every post against its artist's series is O(posts) and grows
# with the library (FC-6.3). Mirrors normalize_tags_task.
_SERIES_RESCAN_CHUNK_SECONDS = 600
@celery.task(
name="backend.app.tasks.admin.rescan_series_suggestions_task",
bind=True,
autoretry_for=(OperationalError, DBAPIError),
retry_backoff=15, retry_backoff_max=180, max_retries=1,
soft_time_limit=1800, time_limit=2400, # 30 min / 40 min
)
def rescan_series_suggestions_task(self, after_post_id: int = 0) -> dict:
"""Score posts against their artist's series and write pending suggestions
(FC-6.3). Settings-gated; time-boxed + self-resuming from a post-id cursor.
Per-task async engine (NullPool) under its own asyncio loop, like normalize."""
import asyncio
from ..models import ImportSettings
from ..services.series_match_service import SeriesMatchService
from ._async_session import async_session_factory
async def _run() -> dict:
async_factory, async_engine = async_session_factory()
try:
async with async_factory() as session:
settings = await ImportSettings.load(session)
if not settings.series_suggest_enabled:
return {"skipped": "series suggestions disabled"}
threshold = settings.series_suggest_threshold
return await SeriesMatchService(session).rescan(
threshold=threshold,
time_budget_seconds=_SERIES_RESCAN_CHUNK_SECONDS,
after_post_id=after_post_id,
)
finally:
await async_engine.dispose()
summary = asyncio.run(_run())
if summary.get("partial") and summary.get("resume_after_id", 0) > after_post_id:
log.info(
"rescan_series_suggestions chunk done (%d scanned, %d suggested, "
"resume after %s) — re-enqueuing",
summary.get("scanned", 0), summary.get("suggested", 0),
summary["resume_after_id"],
)
rescan_series_suggestions_task.delay(summary["resume_after_id"])
return summary