Overview
| Comment: | ingest: consumer pending count decides drained, not a fetch timeout |
|---|---|
| Timelines: | family | ancestors | descendants | both | trunk |
| Files: | files | file ages | folders |
| SHA3-256: |
29aa399eea2ca1b77223ddb3de68af68 |
| User & Date: | dmestas on 2026-08-15 19:36:10.000 |
| Other Links: | manifest | tags |
Context
|
2026-08-15
| ||
| 20:19 |
index: ticket_change(ticket) index; refresh ticket docs once per batch
tcl's ticket corpus crawled at ~20 changes/s: refreshTicketDoc scanned the whole table per change. check-in: 83fc12d336 user: dmestas tags: trunk | |
| 19:36 | ingest: consumer pending count decides drained, not a fetch timeout check-in: 29aa399eea user: dmestas tags: trunk | |
| 19:23 |
index: scale branch recompute to real stores
Indexed reverse parent edge + tag lookup; recursive walk lands in an indexed temp table before the UPDATE probe. sqlite.org's repo (36k check-ins) hung for 15+ min before; milliseconds path now. check-in: cd81fba74d user: dmestas tags: trunk | |
Changes
Modified internal/ingest/ingest.go
from [b442691cbc]
to [c9ea20afeb].
[diff]