The f1-stream verifier's in-process headless Chromium kept tripping
hmembeds' disable-devtool.js Performance detector (CDP latency on
console.log vs console.table) and getting redirected to google.com.
This adds a single-replica chrome-service stack running Playwright
launch-server under Xvfb so callers can connect via WS+token to a
shared headed browser. f1-stream's _ensure_browser now prefers
chromium.connect(CHROME_WS_URL/CHROME_WS_TOKEN) and adds a vendored
stealth init script (webdriver/plugins/languages/Permissions/WebGL
spoofs + querySelector hijack to disarm disable-devtool-auto) on
every new context. Falls back to in-process headless if the env
vars aren't set.
Encrypted PVC for profile + npm cache, NetworkPolicy to TCP/3000
gated by client-namespace label, 6h tar.gz backup CronJob to NFS,
Authentik-gated nginx sidecar at chrome.viktorbarzin.me for human
liveness checks. Image pinned to playwright:v1.48.0-noble in
lockstep with the Python client's playwright==1.48.0.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Per user feedback: the demo Big Buck Bunny / Apple test streams aren't
useful in an F1-streams app. Removed DemoExtractor entirely. Tightened
the discord-extractor path filter from "any stream-shaped path" to
"direct embed/player path only" — the previous filter still let
sportsurge `/event/...` landing pages through, which the verifier
mistook for playable because they render player-class divs without a
real player.
Embed proxy now also rewrites window.fetch + XMLHttpRequest.open inside
the upstream HTML so that cross-origin XHRs (e.g. the hmembeds
`/sec/<JWT>` token-binding endpoint) go through our /embed-asset relay.
This avoids the CORS reject that fired when the player JS tried to call
hghndasw.gbgdhdffhf.shop/sec/... from an `f1.viktorbarzin.me` origin.
The verifier now requires a `<video>` element to mark embed streams
playable (not just a player-class div). Curated streams bypass the
verifier — hmembeds aggressively detects headless Chromium (devtool
trap, console-clear timing, automation flags) and won't progress past
JW Player init in our pod, but the user's real browser should clear
those checks. We can't honestly headless-verify hmembeds, so we trust
the curator instead of falsely rejecting them.
Image: viktorbarzin/f1-stream:v6.1.1
Cuts the stream list from 23 mostly-broken entries to ~6 confirmed-playable
ones, and adds an iframe-stripping proxy so embed sources (hmembeds, etc.)
load through our origin without X-Frame-Options / CSP / JS frame-buster
blocks.
Why: the previous list was dominated by Discord-shared news article URLs,
hardcoded aggregator landing pages, and other non-stream URLs that all sat
at is_live=true because embed streams skipped the health check entirely.
Users could not tell which links would actually play.
What:
- backend/playback_verifier.py: new headless-Chromium verifier (Playwright)
that polls each candidate stream for a codec-independent "playable" signal
(hls.js MANIFEST_PARSED for m3u8; <video>/player div for embed). Replaces
the unconditional is_live=True for embed streams in service.py.
- backend/embed_proxy.py: new /embed and /embed-asset routes that fetch
upstream embed pages, strip X-Frame-Options/CSP/Set-Cookie, and inject a
<base href> + frame-buster-defeat <script> that locks down window.top,
document.referrer, console.clear/table, and window.location so the
hmembeds disable-devtool.js redirect-to-google trap can't fire.
- extractors/curated.py: new always-on extractor with two known-good 24/7
hmembeds embeds (Sky Sports F1, DAZN F1) so the list isn't empty between
race weekends.
- extractors/__init__.py: register CuratedExtractor first; drop
FallbackExtractor (its 10 aggregator landing-pages can't iframe-play).
- extractors/discord_source.py: positive-match path filter (must look like
/embed/, /stream, /watch, /live, /player, *.m3u8, *.php) plus expanded
domain blocklist for news sites — was 10 noise URLs, now ~1.
- extractors/service.py: run_extraction now health-checks AND verifier-
checks both stream types; only verified-playable streams reach is_live.
- main.py: register /embed + /embed-asset routes; defer initial extraction
by 8s so the verifier can reach the local /embed proxy on 127.0.0.1:8000.
- frontend/lib/api.js + watch/+page.svelte: route embed iframes through
/embed proxy instead of the upstream URL, so X-Frame-Options/CSP can't
block them.
- Dockerfile: install Playwright chromium + system codec-runtime libs.
- main.tf: bump pod memory 256Mi → 1Gi for chromium.
Verified end-to-end with Playwright against
https://f1.viktorbarzin.me/watch — 6/6 streams reach a player UI; the 3
demo m3u8s actually play (codec-bearing browser); the 3 embeds (Sky
Sports F1, DAZN F1, sportsurge) render iframes through the proxy.
Image: viktorbarzin/f1-stream:v6.0.5
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Add three new extractors (Streamed.pk, DaddyLive, Aceztrims) for live
F1 streams. Extend ExtractedStream model with stream_type/embed_url
fields, skip health checks for embed streams, fix broken Akamai demo
stream, add variant playlist validation, and add iframe player support
in the frontend for embed-type streams.
- Phase 6: CDN token lifecycle with 3-strategy URL matching and periodic refresh
- Phase 7: SvelteKit 2/Svelte 5 frontend with schedule calendar and hls.js player
- Phase 8: Multi-stream layout supporting up to 4 simultaneous HLS streams
- Update Dockerfile to multi-stage build (Node.js frontend + Python backend)
- Switch deployment to :latest tag with Always pull policy for CI-driven deploys
- Update Woodpecker CI to use explicit latest tag
Phase 4 - Stream Health and Fallback:
- StreamHealthChecker with partial GET validation of m3u8 content
- Bitrate extraction from BANDWIDTH tags
- Response time measurement for quality ranking
- Fallback ordering: live first, fastest response time first
- GET /streams now only returns health-verified streams
Phase 5 - HLS Proxy Core:
- GET /proxy?url= - m3u8 playlist fetch with full URI rewriting
- GET /relay?url= - chunked segment relay (never buffers full segment)
- m3u8 rewriter handles master, variant, and segment URIs
- Base64url encoding for URL parameters
- CORS middleware for browser playback
- Range header forwarding for seeking support
- BaseExtractor ABC with health_check method
- ExtractorRegistry with concurrent fan-out extraction
- ExtractionService with in-memory cache and background polling
- DemoExtractor with 3 public HLS test streams
- Adaptive polling: 5min during live sessions, 30min otherwise
- GET /streams, GET /extractors, POST /extract endpoints
- Fetch 2026 F1 race calendar from jolpica API with all sessions
(FP1-3, Qualifying, Sprint, Race) and UTC timestamps
- Persist schedule to NFS as JSON, load on startup if fresh
- APScheduler daily refresh at 03:00 UTC
- GET /schedule endpoint with live/upcoming/past session status
- POST /schedule/refresh for manual refresh trigger
Replaces the existing Go-based f1-stream service with a new Python/FastAPI
backend as the foundation for the rebuilt F1 streaming aggregation service.
- New FastAPI backend with health and root endpoints
- Python 3.13 slim Dockerfile (replaces Go multi-stage build)
- Updated Terraform deployment (port 8000, reduced resources)
- Buildx-based redeploy.sh with --platform linux/amd64
- Added Woodpecker CI pipeline for automated builds
- Removed all old Go source, node_modules, static assets
Remove the module "xxx" { source = "./module" } indirection layer
from all 66 service stacks. Resources are now defined directly in
each stack's main.tf instead of through a wrapper module.
- Merge module/main.tf contents into stack main.tf
- Apply variable replacements (var.tier -> local.tiers.X, renamed vars)
- Fix shared module paths (one fewer ../ at each level)
- Move extra files/dirs (factory/, chart_values, subdirs) to stack root
- Update state files to strip module.<name>. prefix
- Update CLAUDE.md to reflect flat structure
Verified: terragrunt plan shows 0 add, 0 destroy across all stacks.
