mirror of
https://github.com/hanzoai/index.git
synced 2026-10-30 18:34:27 +00:00
zeekayandHanzo Dev
fd38e23eca
Two bugs, one symptom. The docs site serves pages under /docs/, so `docs.hanzo.ai/errors#<code>` was wrong by a path segment — and no error page existed to land on anyway. Every error this server returns carried that link. The prefix is fixed in the ONE place that builds it (search-types error.rs; the v4/v5 dump readers keep their own copies for the same reason they keep their own error enums). The page it now points at is generated from the same `make_error_codes!` table, so all 281 codes have an anchor, with the type and HTTP status the table already declares. Anchors are the crate's own `name()` output, taken from convert_case rather than derived by hand: three identifiers contain a digit, and `InvalidS3SnapshotRequest` snake-cases to `invalid_s_3_snapshot_request`, not the `invalid_s3_...` a regex produces. Guessing there would have shipped a page whose deep links still missed. The 46 remaining links pointed at upstream doc topics that have no equivalent here, so they now point at the product reference. Less specific than a dead link, and it resolves. Co-authored-by: Hanzo Dev <[email protected]>