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

Update changeset to document totalItems semantic fix

Browse Source 
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
2026-05-20 16:29:17 +02:00
parent 4829e8961e
commit 0119806204
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.changeset/split-query-entities-count.md
+2
View File
@@ -3,3 +3,5 @@
---
Split the `queryEntities` list and count into separate queries instead of a multi-reference CTE. When the `filtered` CTE was referenced twice (once for the count, once for the data), PostgreSQL refused to inline it, forcing full materialization of the filtered set before applying `LIMIT`. By running the count as a standalone query, the list CTE is only referenced once, allowing the planner to short-circuit on `LIMIT` and return the first page in milliseconds instead of waiting for the full filtered set to materialize.
The standalone count query also fixes a pre-existing bug where `totalItems` was inflated for entities with multi-valued sort fields (e.g. tags). The old CTE-based count counted search rows, so an entity with 3 tags would be counted 3 times. The new count uses `EXISTS` to count distinct entities, aligning `totalItems` with the number of entities actually reachable through cursor pagination.