paulononato/backstage
1
0
Fork 0
Code Issues Pull Requests Actions 43 Packages Projects Releases Wiki Activity

incremental-ingestion: use ANY(array) instead of IN(...) on Postgres

Browse Source 
The whereIn('ref', refs) calls on ingestion_mark_entities generated
a unique prepared statement for every distinct array length, bloating
the Postgres query plan cache. On Postgres, use = ANY($1) with a
single array parameter instead. Falls back to regular whereIn on
SQLite/MySQL.

Signed-off-by: Fredrik Adelöw <freben@gmail.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Signed-off-by: Fredrik Adelöw <freben@gmail.com>
This commit is contained in:
Fredrik Adelöw
2026-05-19 11:39:28 +02:00
parent 8f867a2078
commit 32f0dfe7f8
3 changed files with 168 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.changeset/incremental-ingestion-any-array.md
+5
View File
@@ -0,0 +1,5 @@
---
'@backstage/plugin-catalog-backend-module-incremental-ingestion': patch
---
On PostgreSQL, `WHERE ref IN ($1, $2, ..., $N)` queries on the `ingestion_mark_entities` table now use `= ANY($1)` with a single array parameter instead. This reduces prepared statement bloat in the query plan cache when the number of entity refs varies between calls.