Repository navigation
Conversation
An or's gather cte was joined to the cte before it even when that cte belonged to something else: the patterns before an enclosing or, or the previous query. The first selected those columns twice, so a where like a AND (b OR ((c OR d) AND e)) failed with an ambiguous column. The second built the or from another query's rows, so it returned nothing when that query was empty. Treat an or at group-idx 0 as cte-for-pattern treats a single pattern there. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01HRKkuGrp63pNCgPK1hVdC7
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
📝 Walkthrough
Merge Risk: ⚪ Minimal · up to
|
| Check name | Status | Explanation |
|---|---|---|
| Title check |  |
The title clearly and concisely describes the main fix: preventing an OR at the start of a group from joining the previous CTE. |
| Description check | |
The description directly explains the affected cases, the implementation change, regression coverage, and test results. |
| Linked Issues check | |
Issue #2947 requires an or at group-idx 0 to avoid joining the preceding CTE, including nested and later top-level query cases. The supplied change summary states that accumulate-ctes now passes… |
| Out of Scope Changes check | |
The reported changes are limited to the accumulate-ctes implementation and a regression test in server/test/instant/db/instaql_test.clj. Both changes directly support issue #2947. No unrelated cha… |
| Docstring Coverage | |
No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0… |
Full details: Linked Issues check
-
Explanation
Issue
#2947requires anoratgroup-idx0 to avoid joining the preceding CTE, including nested and later top-level query cases. The supplied change summary states thataccumulate-ctesnow passes-1toor-gather-ctefor the start-of-group case. The addedor-at-the-start-of-a-grouptest covers both reported cases and checks the expected results. The reported test suites passed with 393 tests and 19,653 assertions.
Full details: Out of Scope Changes check
-
Explanation
The reported changes are limited to the
accumulate-ctesimplementation and a regression test inserver/test/instant/db/instaql_test.clj. Both changes directly support issue#2947. No unrelated change is identified.
Full details: Docstring Coverage
-
Explanation
No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0 files. (2 skipped: 2 unsupported.)
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
Comment @coderabbitai help to get the list of available commands.
Fixes #2947.
accumulate-ctesalready avoids joining anor's gather CTE to the previous CTE when that CTE is the last one of a siblingorbranch. It still joins it in two other cases:orstarts a branch of an outeror, where the previous CTE comes from before the outeror;orstarts a pattern group, where the previous CTE belongs to the previous query.The first selects the earlier CTE's columns twice, so Postgres rejects them as ambiguous. The second builds the
orfrom another query's result rows, so it returns nothing when that query is empty.This treats an
orat the start of a group (group-idx0) like a single pattern there.cte-for-patternalready does the same, throughstart-of-group?, and doesn't join to the previous CTE.or-at-the-start-of-a-groupcovers both shapes on the zeneca app. Without the fix, the first assertion fails withambiguous-columnand the second returns no users. I also ran theinstant.db.*,instant.reactive.*andinstant.admin.*suites; all 393 tests passed, with 19,653 assertions.🤖 Generated with Claude Code
https://claude.ai/code/session_01HRKkuGrp63pNCgPK1hVdC7