Block a user
DeleteDB returns on the first failing suffix, so it can report a leftover file when the history is already destroyed
Try every event database file on delete and say which was left (closes #275)
Try every event database file on delete and say which was left (closes #275)
Review passed.
Model: opus-5-5
Clickable things must look clickable: the entrypoint URL "copy" control, "+ Add", and every other one
Make every clickable control look clickable, in two shared styles (closes #375)
Make every clickable control look clickable, in two shared styles (closes #375)
Review passed: all five findings from the review of 2 October are fixed.
Model: opus-5-5
Try every event database file on delete and say which was left (closes #275)
Added a handler test where the event database file itself cannot be removed (a non-empty directory at its path). It checks that the logged line says the event database file is still on disk and…
The main database's open errors do not name the file
The withRetry bookkeeping-failure branch is unpinned: removing its error check leaves the suite green
Opened sneak/webhooker#458. Tests only: withRetry's check on writing the delivery result, and six other error checks in internal/delivery/target_http.go that could be…
Pin the HTTP target's unpinned error checks (closes #285)
Follow-ups from the PR 251 review: two comments state the wrong mechanism, and the resubmit route's middleware is untested
Done in sneak/webhooker#457.
- The comment in
loadResubmitSourcenow says a reaped event is gone because the retention reaper deletes its row outright. - The comment…
Fix two resubmit comments and test the resubmit route's middleware (closes #252)