Each row of the webhook page's target list now shows two more lines: Delivered: and Failed:, each as a total and as the number in the last 24 hours. Deliveries still pending or retrying count in neither.
The totals are the target's delivered and failed running totals from #368, which retention does not reduce. The 24-hour figures reuse that issue's count over the deliveries' status index, one query grouped by target for all of them. No schema change, so no database needs recreating.
If the event database cannot be read, each row says the delivery counts could not be read rather than showing zeros. A webhook with no event database yet shows zeros, and the page does not create one.
The archive file details and the Download button are unchanged; the new lines follow them.
The browser test's seeded failed delivery now has a finish time and is counted in its target's totals, as the delivery engine leaves one, so the browser test can check the row.
Judgement call: the target list reads the totals and the 24-hour count itself rather than sharing the statistics pane's reads, so the page runs the 24-hour count twice.
Model: opus-5-5
Each row of the webhook page's target list now shows two more lines: `Delivered:` and `Failed:`, each as a total and as the number in the last 24 hours. Deliveries still pending or retrying count in neither.
The totals are the target's `delivered` and `failed` running totals from https://git.eeqj.de/sneak/webhooker/issues/368, which retention does not reduce. The 24-hour figures reuse that issue's count over the deliveries' status index, one query grouped by target for all of them. No schema change, so no database needs recreating.
If the event database cannot be read, each row says the delivery counts could not be read rather than showing zeros. A webhook with no event database yet shows zeros, and the page does not create one.
The archive file details and the Download button are unchanged; the new lines follow them.
The browser test's seeded failed delivery now has a finish time and is counted in its target's totals, as the delivery engine leaves one, so the browser test can check the row.
- Judgement call: the target list reads the totals and the 24-hour count itself rather than sharing the statistics pane's reads, so the page runs the 24-hour count twice.
Model: opus-5-5
Each row of the webhook page's target list now shows the target's
delivered and failed deliveries, in total and in the last 24 hours.
The totals are the target's running totals, so retention does not
reduce them; the 24-hour figures are counted over that window of the
deliveries' status index, in one query for all targets. Deliveries
still pending or retrying count in neither.
Model: opus-5-5
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Each row of the webhook page's target list now shows two more lines:
Delivered:andFailed:, each as a total and as the number in the last 24 hours. Deliveries still pending or retrying count in neither.The totals are the target's
deliveredandfailedrunning totals from #368, which retention does not reduce. The 24-hour figures reuse that issue's count over the deliveries' status index, one query grouped by target for all of them. No schema change, so no database needs recreating.If the event database cannot be read, each row says the delivery counts could not be read rather than showing zeros. A webhook with no event database yet shows zeros, and the page does not create one.
The archive file details and the Download button are unchanged; the new lines follow them.
The browser test's seeded failed delivery now has a finish time and is counted in its target's totals, as the delivery engine leaves one, so the browser test can check the row.
Model: opus-5-5
Review passed.
Model: opus-5-5