test(supabase): the audit privacy check no longer reads a uuid as a date

014 #237 matched « 20[0-9]{2}- » inside a workspace uuid on a long-lived stack.
The pattern now reads the metadata without workspace_id, still over every
audit row. It would have stopped the DEV procedure at db:test. Found by the
V1-1.6 verifier.
This commit is contained in:
bchanot committed 2026-10-26 17:53:37 +01:00
1 parent 10413478fb
commit e7fd7e34e6
2 files changed
+9 -1

No files matched your search

@@ -94,3 +94,9 @@ D-R7 form checks; no artificial Rust/wasm call; the boundary choice written in
client page now reloads its animals). Fixed: `dropLateReplays` in `e2e/support/sandbox.ts`, called after each test
by the three shell specs; the client journey ends on the reloaded animal section; `e2e:shell` 82/82 three runs in a
row.
- Verifier round 2 (ECARTS C10): pgTAP 014 #237 « no name, role, breed or date in the audit rows » failed on the
long-lived sandbox: its date pattern `20[0-9]{2}-` matched a workspace uuid (`…2012-…`) in the metadata of a real
row (a false positive, latent since V1-1.4, that would also have stopped the owner's DEV procedure at `db:test`).
Authorised exception to C1 (test file only, no migration): the pattern now reads the metadata without
`workspace_id` (the test before it proves that key is the only one), still over every audit row; counter-checked
(uuid alone: false; a name or a date added: true); pgTAP 919/919 on the sandbox and on PGlite.