The page hotlinked ~100 AniList CDN covers per load and some failed at random.
The reader now downloads each catalog cover once (sequential, ~1/s, stops on
401/403/429, images up to 2 MB) into data/greatest-covers/ and serves it from
/api/greatest-cover with a year-long browser cache; until a copy exists the
route redirects to the catalog URL. The page also accepts the server's
relative cover URLs, so owned titles show their library cover again.
GREATEST_COVER_WARM=0 turns the downloads off (tests, offline installs).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Cache prune counts directory entries first and only reads files when over
the cap; tests for cross-kind prune, schema-mismatch and corrupt entries.
- Scoring tolerates malformed manga.json and cached records (non-list tags,
object or null relations, non-object items); a tag without a rank no longer
passes the top-tag filter; authors without an integer AniList id are dropped.
- discovery.json publishes only candidates some feed shows; examined/clipped
still describe the pool. Tests that asserted the published count equals the
pool size now check the pool via candidatePool() and the published set
against the feeds.
- The post-harvest rescore failure test now breaks discovery.json right before
the publish (via a scraper stand-in) instead of relying on tags: 5, and a
following successful harvest clears the stale lastError.
- Consent text names everything sent to AniList; paused label mentions admin
harvests; the UI test asserts the XSS seed instead of skipping it.
- Slot-mix test proves the fresh/outside reservations matter.
- Docs: privacy note, fresh fallback reason, per-period cap, dry-run wording,
published-candidate rule; plan/spec hygiene and spec status.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Chapter rows show when you first finished a chapter (setting: date + hour,
date, off) from a new durable progress doneAt; older reads fall back to the
reading clock's earliest remembered completion.
- chapter.json keeps added (arrival in this library, now kept on replace) plus
x-dates.source (source upload, bulk imports flagged), x-dates.release
(official, for a later metadata job) and x-dates.ecosystem (first entry into
any Manga Circus library; carried over peering, earliest wins, clamped).
- Rows show one release date by tier: official > source upload > ecosystem >
library, with all known dates in the tooltip (setting to hide).
- Fixes: WeebCentral dates came from the chapter page, which has none (every
chapter got the download time); they now come from the full chapter list.
MangaDex no longer writes a group's publishAt into added. Peering sends dates.
- tools/backfill-chapter-dates.js migrates an existing library (dry run default).
- Coverage dry run on the server: docs/chapter-dates-coverage-2026-10-24.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The stall watchdog now runs inside the scraper (ScraperService.checkStalls,
every 30 s from the work loop): same rule as before, one checkpoint-safe
restart, quarantine on the second stall, missing checkpoint is no evidence,
and paused time no longer counts. Stalls are recorded as scraper incidents
(failure.kind 'stall'). The private /api/recovery endpoint is gone.
Dropped with the service: reader/scraper-unreachable incidents (supervisor
and Docker handle crashes; the health page shows "Scraper: unreachable"),
duplicate job-failure incidents and the unused /metrics endpoint.
The stack is now reader + scraper (+ optional peer-gateway).
The supervisor test moved to runtime/test (npm run test:runtime stays).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
An uncached scan probed one image at a time and read 256 KB of each. On
a NAS that is bound by per-file latency (the server/Synology: ~20 ms per
image, ~100 min for 298k images). Now chapter folders of a book and
their images are probed SCAN_CONCURRENCY (default 8, 1-64) at a time
and assembled in folder order, and imageSize() reads 32 KB first, then
256 KB, then the whole file: a prefix gives either no size or the same
size, so results do not change. Tests: every format and read stage
equals the whole-file answer; concurrency 1 and 8 build the same index
and cache; the limiter keeps its bound and per-task errors. (BL-0043)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New POST /api/admin/user/role {name, admin}: admin-only (checked on the
server), keeps the password and sessions, takes effect on the target's
current session at once, logs who changed whom. Revoking your own
rights is refused, so at least one admin always remains. Settings ->
Users & access shows "Make admin" / "Remove admin" next to every other
user, with a confirmation. Test: normal users and anonymous callers are
refused, input checks, promotion and revocation apply immediately and
are saved.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On the server the backup folder is a Synology NFS share with "map all
users to admin": access(W_OK) returns EACCES for uid 1000 although
writes succeed, so the first scheduled run failed with "not usable".
Create and remove a probe file instead; test that it leaves nothing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The reader owns a scheduled backup into BACKUP (Compose: BACKUP_PATH mounted at
/backup, off when unset). A run pauses the scraper queue through its control
API, waits (bounded) until no job is active, writes a verified tar.gz of DATA
with db.json from memory and scraper/queue.json stored paused, excludes
runtime/lock/pid/temp files, resumes only a queue it paused, then applies
count- or day-based retention to its own archive names only. Schedule,
retention, "Back up now" and last result live in Settings > Service health.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaashkinoozheandClaude Opus 5.5