sulthan 78234f3c19 Browser-backed Sites join the Cover pipeline (#62) (#72)
Fixes #62

Browser-backed Sites join the Cover pipeline: kagane and novelfull Series now get their Covers at creation, through the same acquisition path as every other Site, instead of waiting for a poll pass.

## What changed

`latest.Acquirer` (creation-time acquisition, fired by the first Bookmark of a Series) previously skipped kagane and novelfull entirely — their pages only yield a Cloudflare challenge to the TLS client, so the request was spent for nothing. It now routes them like the poller does, with the two Sites split exactly as the issue demands:

- **kagane** — page fetched through the browser sidecar, cover URL extracted from the API JSON, bytes fetched through the browser sidecar (the only path that clears the challenge) into the content-addressed store. With no `BROWSER_WS_URL` configured, acquisition is skipped entirely and nothing falls back to a plain fetch.
- **novelfull** — page fetched through the browser sidecar, cover URL extracted from the HTML, bytes fetched over plain TLS through the ordinary gated fetcher (its image paths answer 200 with `access-control-allow-origin: *`, measured 2026-08-09). With no browser configured, the page fetch falls back to the TLS client — novelfull's challenge is a live time-varying fact (AGENTS.md), so when the page body answers, the Cover still lands; when it is challenged, nothing happens.

The byte-routing rule (kagane → browser, every other Site → TLS) is now one shared function (`latest.fetchCoverBytes`) used by both the Poller and the Acquirer, so the two cannot drift apart.

## Acceptance criteria

- [x] kagane cover bytes are fetched through the browser sidecar and stored in the content-addressed store — `TestAcquireKaganeCoverThroughBrowser`
- [x] novelfull cover URLs are extracted from the browser-fetched HTML, and its bytes are fetched over plain TLS — `TestAcquireNovelfullCoverOverPlainTLS`
- [x] With no browser sidecar configured, kagane Covers are absent and nothing falls back to a plain fetch — `TestAcquireKaganeSkippedWithoutBrowser`
- [x] With no browser sidecar configured, novelfull Covers still work if its page body is available — `TestAcquireNovelfullCoverWithoutBrowser`
- [x] Manually verified on-device: a kagane Series shows its Cover in the panel, not a broken-image glyph — being run by a separate manual-verification agent against a mocked scenario (no prod data); not part of this PR
- [x] `go test ./...` is green, with live-network checks gated behind `SMOKE_BROWSER_WS_URL` like the existing kagane image smoke test — new `TestSmokeAcquireKaganeCover` proves the end-to-end acquire path against the real browser when the env var is set

## Verification

- `go test ./...` green across all packages
- New unit tests exercise every routing decision with fakes — no network in the default suite
- Smoke test gated behind `SMOKE_BROWSER_WS_URL`, skipped by default

## Post-review changes (a66491a)

- **One routing rule for pages too** — `fetcherFor` is now a shared function used by both the Poller and the Acquirer; novelfull falls back to the plain-TLS fetcher in *both* when no browser is configured, so pre-existing (client-scraped) novelfull rows get healed by the poll as well, not just Series created after this change (`TestNovelfullUsesTLSWhenNoBrowserFetcher`).
- **Byte-level no-fallback proof** — `TestAcquireKaganeBytesNeverFallBackToPlainTLS` pins that kagane cover bytes never route to the TLS fetcher even when the page came through a browser.
- **Acquirer wired independent of the TLS client** — if `NewTLSFetcher` fails, kagane/novelfull acquisition still works via the sidecar (`main.go`).
- AGENTS.md (root + backend) updated for the novelfull plain-TLS fallback.

Reviewed-on: #72
Co-authored-by: Sulthan Zaki <sultankiki05@gmail.com>
Co-committed-by: Sulthan Zaki <sultankiki05@gmail.com>
2026-08-10 11:06:36 +07:00
2026-07-24 16:23:24 +07:00
2026-07-24 16:23:24 +07:00

Manga Bookmark

Track manga read-progress on asurascans.com (a.k.a. asuracomic.net), demonicscans.org, comix.to, and kagane.to from a phone (Bromite / mobile Chromium), synced to a self-hosted Go backend so bookmarks unify across all four sites and all devices.

Two parts:

  • backend/ — tiny Go (net/http + Postgres via pure-Go pgx) sync service. 4 routes, static binary, distroless container.
  • userscript/manga-bookmark.user.js — single Bromite-compatible userscript (no GM_* APIs) that injects an on-page bookmark UI and syncs via fetch().
Bromite userscript (isolated world, Shadow DOM UI, localStorage cache)
    -- fetch() HTTPS -->  reverse proxy (TLS + CORS)  -->  Go net/http  -->  Postgres (volume)
                                                              |
                                                              | CDP over the tailnet
                                                              v
                                                    headless Chrome, on-demand,
                                                    on a separate machine
                                                    (chrome/, ADR-0006)

Kagane and novelfull sit behind a Cloudflare JavaScript challenge no TLS fingerprint clears, so the poller reaches those two through a real Chrome over CDP. That browser is not part of the API stack: it is its own compose unit on a second machine, spawned on the first connection and reaped when idle. The API needs it only to discover new chapters and to fetch a kagane cover once — covers are stored, so the library renders in full with the browser switched off.


1. Backend

Config (env)

Var Default Notes
TOKEN_KEY (required) Secret every Reader's userscript credential is derived from (issue #24); only SHA-256 hashes of credentials are stored.
OWNER_DISCORD_ID (required) Discord user ID of the owner: seeded as the first Reader, owns every pre-registration bookmark, and is the only Reader who can revoke another's sessions.
ALLOWED_ORIGINS Asura + Demonic + Comix + Kagane origins Comma-separated CORS allowlist.
DATABASE_URL (required) Postgres connection URL, e.g. postgres://bookmarks:…@postgres:5432/bookmarks?sslmode=disable. Compose builds it from POSTGRES_PASSWORD.
COVER_DIR (required) Filesystem volume for immutable, content-addressed Cover bytes. Compose builds the image and mounts cover-data at this path; standalone runs may choose another writable durable path.
PORT 8080 Plain HTTP; TLS terminated by the proxy.
BROWSER_WS_URL empty CDP endpoint of the remote browser (ws://<tailnet IP>:9222), used to poll Kagane/Novelfull past their JS challenge and to fetch uncached Kagane covers. Must be an IP or localhost — Chrome's DevTools handler 500s any other Host header, MagicDNS names included. Unset disables both; stored covers still serve.
DISCORD_CLIENT_ID (required) Discord application credentials for the browser sign-in (ADR-0002).
DISCORD_CLIENT_SECRET (required) As above. Never logged, never echoed in an error.
DISCORD_GUILD_ID (required) The one guild whose membership gates sign-in, checked at login only. Membership is registration: any member becomes a Reader on first login.
DISCORD_REDIRECT_URI (required) Exact callback URL; Discord matches it verbatim against the registered redirect.
DISCORD_REQUIRED_ROLE empty Role snowflake a member must additionally hold. Empty means guild membership alone suffices.
DISCORD_API_BASE https://discord.com/api/v10 Test seam — tests point it at a local stub so the real token exchange runs.
USERSCRIPT_PATH /userscript/manga-bookmark.user.js Bindmounted file served at /u/{token}/manga-bookmark.user.js.
NOVEL_USERSCRIPT_PATH /userscript/novel-bookmark.user.js Same, for the novel library.
LATEST_CHAPTER_POLL_ENABLED 1 0 turns the poller off entirely.
LATEST_CHAPTER_POLL_COOLDOWN 1h Rest between checks of one plain-TLS series; floor 15m.
LATEST_CHAPTER_POLL_BROWSER_COOLDOWN 6h Rest between checks of one browser-backed series; floor 15m.
LATEST_CHAPTER_POLL_INTERVAL 10m How often the poller wakes. Cannot shorten either cooldown.
LATEST_CHAPTER_POLL_BATCH 14 Series per wake. Keep BATCH × STAGGER under INTERVAL.
LATEST_CHAPTER_POLL_STAGGER 20s Delay between fetches in a batch — this is the outbound request rate.

Compose reads a few more from the same .env that the backend never sees: POSTGRES_PASSWORD (required — DATABASE_URL is built from it, and Postgres only applies it while postgres-data is empty), BOOKMARK_API_HOST and BOOKMARK_WEB_HOST (required by the prod override), and the optional PROXY_NETWORK / TRAEFIK_ENTRYPOINT / TRAEFIK_CERTRESOLVER. The browser unit has its own chrome/.env on its own machine — BROWSER_BIND_ADDR (required, the tailnet IP the CDP port is published on) and the optional BROWSER_TZ. Full commentary is in .env.example and chrome/.env.example; deployment order is DEPLOY.md.

Endpoints

Method Path Auth Description
GET /bookmarks Bearer All bookmarks of the acting Reader.
PUT /bookmarks/{key} Bearer Upsert one series; returns the row as stored.
DELETE /bookmarks/{key} Bearer Remove one.
GET /healthz none 200 ok.
GET /u/{token}/manga-bookmark.user.js credential in path Serves the userscript with the requesting Reader's credential substituted in and an mtime-derived @version.

key is <site>:<series_id> — e.g. asura:trash-of-the-counts-family-f886a8af, demonic:Infinite-Level-Up-in-Murim, comix:12345, or kagane:3fa85f64-5717-4562-b3fc-2c963f66afa6. Sync is last-write-wins.

updated_at orders the bookmark list, so it moves only on real reading progress: the server applies its timestamp when the row is new or last_chapter_num changes, and otherwise keeps the stored one. Favouriting a series or recording a newly published chapter therefore leaves the order alone. Because the timestamp a client sends is only a candidate, PUT echoes the row as stored and clients adopt that rather than their own payload.

Develop / test

cd backend
go test ./...                          # unit + handler tests
CGO_ENABLED=0 go build                 # static binary

go test ./... requires Docker. The store talks to a real Postgres, so each test package starts a throwaway postgres:17-alpine container and gives every test its own database inside it (internal/pgtest). Nothing is stubbed and nothing reaches the network beyond the local Docker daemon.

Run the stack

cp .env.example .env
# edit .env: set TOKEN_KEY (openssl rand -hex 32) and
#            POSTGRES_PASSWORD (openssl rand -hex 24)

docker compose up -d --build           # binds 127.0.0.1:8080

That brings up two services — the API and Postgres. The browser is deliberately not one of them; without BROWSER_WS_URL the poller logs and skips kagane and novelfull, and everything else works. To run one locally, publish it on the Docker bridge gateway so the API container can name it by IP:

cd chrome
echo 'BROWSER_BIND_ADDR=172.17.0.1' > .env
docker compose up -d --build
# then in the repo's own .env:  BROWSER_WS_URL=ws://172.17.0.1:9222

Bind it to 127.0.0.1 instead if you only want to drive it from the host, e.g. SMOKE_BROWSER_WS_URL=ws://127.0.0.1:9222 go test -run TestSmokeKagane ./internal/latest. In production that address is the home machine's tailnet IP and nothing else — see DEPLOY.md §7 and ADR-0006.

Smoke test:

# The credential is per Reader and derived, so there is no token in .env to
# grep. Take yours from the Userscripts panel's install link after signing in,
# or read it out of an installed script's API_TOKEN constant.
TOKEN=<your Reader credential>
curl -s localhost:8080/healthz                                    # ok
curl -s localhost:8080/bookmarks                                  # 401
curl -s -H "Authorization: Bearer $TOKEN" localhost:8080/bookmarks # []
curl -s -X PUT -H "Authorization: Bearer $TOKEN" -H 'Content-Type: application/json' \
  -d '{"title":"Test","last_chapter":"Chapter 1","last_chapter_num":1}' \
  localhost:8080/bookmarks/asura:test-1
curl -s -i -X OPTIONS -H 'Origin: https://asurascans.com' \
  -H 'Access-Control-Request-Method: PUT' \
  localhost:8080/bookmarks/asura:test-1 | grep -i access-control   # 204 + CORS headers

Deploy behind your reverse proxy

Route https://bookmark-api.<domain> → the service on :8080 (TLS at the proxy).

  • Host proxy (nginx/Caddy on the host): the base compose already binds 127.0.0.1:8080; point the proxy proxy_pass http://127.0.0.1:8080;.

  • Docker proxy (Traefik/nginx in a container on its own network): use the override, which drops the published port and joins the shared network:

    docker network create proxy            # once, if it doesn't exist
    docker compose -f docker-compose.yml -f docker-compose.prod.yml up -d --build
    

    Set PROXY_NETWORK in .env if your network isn't named proxy.

Verify: https://bookmark-api.<domain>/healthz returns ok over valid TLS (no mixed-content), and an OPTIONS preflight from a real site origin returns the CORS headers.


2. Userscript

Install

Sign in to the web UI and open the Userscripts panel: it offers one install link per library. Each link serves a script rendered with your own credential already inside it — you never see, type or copy a credential. The served script carries @downloadURL/@updateURL pointing at its credential-bearing path, so Violentmonkey keeps auto-updating it.

The bindmounted files carry __API_TOKEN__ placeholders; the backend substitutes the requesting Reader's credential at serve time, so no real credential is ever committed. Rotating the credential (same panel) invalidates every installed copy immediately — reinstall on all devices.

Install on Bromite (mobile)

Bromite runs Chromium's native userscript engine (no Tampermonkey needed):

  1. Bromite → Settings → User scripts → enable user scripts (allow the permission prompt).
  2. Open the install link from the web UI — Bromite detects the .user.js and offers to install it.
  3. Confirm the install; the @match list covers both sites.
  4. Open a series on either site — a 📑 button appears bottom-right.

Exact menu wording varies by Bromite build; if "User scripts" is absent, update Bromite or use a build with userscript support.

Desktop iteration (optional)

The script is GM_*-free, so it also runs in Tampermonkey/Violentmonkey on desktop for faster testing — install the same file unchanged.

Use

  • Bookmark: on a series or chapter page, open the panel → + Bookmark this.
  • Auto-progress: opening a chapter of a bookmarked series records it when the chapter number ≥ the stored one (re-reading older chapters never regresses progress; unparseable numbers set the current chapter).
  • Manual override: panel → Edit on any row forces a specific chapter.
  • Continue: jumps to the last-read chapter (or the series page).
  • Latest chapter: rows read Read: … · Latest: … once the newest published chapter is known and it is ahead of your progress. See below for how that is found.
  • Favourites: the ☆ on any row toggles it; the ★ Favourites tab narrows the list. Favourited series still appear under All. The flag syncs, so it follows you across devices; the chosen tab does not persist.
  • Archive: the Archive button on any row parks a series — it drops out of All and ★ Favourites and moves to the Archived tab. The server keeps checking it for new chapters, so it is worth coming back to. Archiving does not touch read progress, and reading an archived series leaves it archived.
  • Finished: series you have completed live in a Finished tab in the web UI only. It is set there and nowhere else — the API rejects the value — and finished series are hidden from every userscript tab and are no longer polled for new chapters.
  • Bookmarks made on Asura appear when the panel is opened on Demonic, and vice versa — the backend is the shared store.

Neither favouriting nor learning a new chapter reorders the list — only reading progress does.

Offline / backend down: changes are cached in localStorage and retried on the next successful load (last-write-wins).

How "latest chapter" is found

Only a series page lists every chapter (a reader page links just its neighbours), and the backend cannot fetch either site — Cloudflare blocks server-side requests, and neither site offers an API or feed to poll. So the userscript does the looking, from your own browser session:

  • Opening a bookmarked series page records its newest chapter directly.
  • Otherwise it fetches series pages in the background — same-origin only, so browsing Asura refreshes Asura bookmarks and Demonic refreshes Demonic. One series per navigation, and at most one check per series every 4 hours (LATEST_CHECK_BATCH / LATEST_CHECK_THROTTLE_MS). Failures are silent and simply retried after the window.

Freshness is tracked per device in localStorage under bmgr:manga:lastchecked and is deliberately not synced, since each device checks on its own.

This means a bookmark is as current as its last check — not the moment a chapter drops. Nothing can be instant here: neither site offers push, feeds, or an API.


Adapter reference (verified live 2026-07-24)

The site adapters key everything off URL regex, with title from og:title (or the page's own heading where a site ships none). No adapter reads a cover: the backend acquires, stores and serves every Cover from its own origin (ADR-0007). Confirmed against live pages via Playwright:

Site Series URL Chapter URL series_id
Asura (asurascans.com) /comics/<slug-hash> /comics/<slug-hash>/chapter/<n> <slug-hash>
Demonic (demonicscans.org) /manga/<slug> /title/<slug>/chapter/<n>/<page> (chaptered.php?manga=<id>&chapter=<n> 301s here) <slug>
Comix (comix.to) /title/<id>-<slug> /title/<id>-<slug>/<uploadId>-chapter-<n> <id>
Kagane (kagane.to) /series/<uuid> /series/<uuid>/reader/<bookUuid> <uuid>

Notes:

  • asuracomic.net deep links are dead (re-checked 2026-07-25). They 301 to the asurascans.com root, discarding the path, at the edge — before the userscript gets a document — so nothing client-side can rescue them. Reach series through asurascans.com. The host stays matched in case the redirect starts preserving paths again.
  • Asura og:title carries a Chapter N - Read Online \| Asura Scans suffix that the adapter strips; Demonic chapter og:title is <Title> Chapter N.
  • Demonic's <slug> is identical on /manga/… and the canonical /title/… reader, so a bookmark set from the series page and the auto-update from the reader resolve to the same key.
  • Asura showed no Next.js markers on the live site, so navigation uses a framework-agnostic watcher (history patch + polling) rather than a Next-only hook — works for client-routed and full-reload sites alike.

If either site changes its URL shape, update the regex in the matching adapter in userscript/manga-bookmark.user.js and re-verify.

S
Description
No description provided
Readme 13 MiB
Languages
Go 75.9%
JavaScript 15.1%
CSS 4.6%
HTML 3.6%
Shell 0.5%
Other 0.3%