syntaqxA file's blame is fetched by its page, so a slow first one still arrives. 834f2998 streamed blame from the blob loader, but the site keeps a page's stream open for 5 seconds (entry.server.tsx streamTimeout), and a first blame can take up to about 6 (repos' walk is cut at 6 s): live on flagon-io/g1t's services/actions/src/plan.rs at 834f2998 the stream ended at 5.7 s with the blame cut off, the browser logged Unexpected Server Error, and the page said "Working out who changed each line" for good. The page now opens with the file and fetches the blame from /<owner>/<repo>/blame/<ref>/<path> (routes/repo/blame-data.ts, JSON, private and noindex), on a request of its own with no stream to outlast; it says the blame could not be worked out when the request fails. robots.txt keeps crawlers off that path too (Disallow: /*/*/blame/). Checked: the site's typecheck (tsc -b --force) and unit tests (950 passing, the robots test with the new rule).0c83797History
- ..
- about.rs
- activity.rs
- addresses.rs
- alerts.rs
- archives.rs
- audit.rs
- billing.rs
- blobs.rs
- broker.rs
- caches.rs
- checks.rs
- commits.rs
- contents.rs
- coordination.rs
- deploy_keys.rs
- deployments.rs
- folios.rs
- follow.rs
- lib.rs
- limits.rs
- logs.rs
- mcp.rs
- notifications.rs
- oauth.rs
- oidc.rs
- onboarding.rs
- openapi.rs
- operations.rs
- packages.rs
- people.rs
- pins.rs
- projects.rs
- protection.rs
- reference.json
- renamed.rs
- responses.rs
- rest.rs
- rules.rs
- run_artifacts.rs
- runners.rs
- security.rs
- telemetry.rs
- token_policy.rs
- toolkit.rs
- tools.rs
- why.rs