fix(plan): handle correlated scalar outer projections by aunjgr · Pull Request #25920 · matrixorigin/matrixone

aunjgr · 2026-07-20T16:49:43Z

What type of PR is this?

Which issue(s) this PR fixes:

What this PR does / why we need it:

Direct outer-column references projected by correlated scalar subqueries could remain in the executable right subtree, causing internal errors and incorrect behavior through DISTINCT and ORDER BY/LIMIT wrappers.

This change:

normalizes the direct correlated projection before predicate pull-up;
preserves SINGLE JOIN cardinality checks for direct and ORDER BY-only cases;
uses MARK JOIN existence semantics for DISTINCT and LIMIT 1;
seeds predicate-free scalar MARK joins with TRUE so empty and nonempty inner inputs preserve all outer rows correctly;
adds planner and distributed regressions for empty, one-row, and multi-row inner inputs.

Testing

make build
go vet ./pkg/sql/plan
.agents/skills/mo-dev/scripts/mo-cgo-test -v -count=1 -timeout=180s ./pkg/sql/plan
focused mo-tester BVT: 21/21 passed

qodo-code-review · 2026-07-20T16:52:17Z

Qodo reviews are paused for this user.

Troubleshooting steps vary by plan Learn more →

On a Teams plan?
Reviews resume once this user has a paid seat and their Git account is linked in Qodo.
Link Git account →

Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center?
These require an Enterprise plan - Contact us
Contact us →

…d-scalar-projection

aunjgr · 2026-07-22T04:49:38Z

@gouhongshen Addressed in commit 20ecd5903. The rewrite now preflights the exact synthesized CASE signature through the function resolver and proceeds only when the resolved return type is semantically equal to the source type via types.Type.Eq. ENUM coercion and unsupported ROWID/vector cases therefore stay on the pre-existing path, as do metadata-changing cases such as BIT width changes; ordinary INT metadata is accepted using the codebase semantic type-equivalence rule. Added focused type regressions for INT, ENUM, ROWID, vector, and BIT. Full pkg/sql/plan tests, build, vet, and the targeted BVT pass.

XuPeng-SH

Reviewed latest head 20ecd59. No blocking findings.

The prior requested changes are resolved: OFFSET/rank/dynamic or zero LIMIT shapes now fall back before mutation; literal LIMIT 1 uses MARK existence semantics; LIMIT >= 2 removes the misplaced limit while SINGLE join preserves per-outer-row cardinality; synthesized CASE is gated by semantic type equality.

Verified with full pkg/sql/plan tests, focused regressions 5x, go build, go vet, and an exact-head mo-service execution matrix covering empty/one/multi-row inputs, NULLs, duplicate outer rows, DISTINCT, LIMIT 1, LIMIT 2, and proper 1242/21000 multi-row errors. Current CI has no failures (three jobs are still running).

gouhongshen

Codex automated review

No blocking or non-blocking findings remain. The latest discussion is supported by the current code: prior LIMIT/OFFSET and CASE type regressions are resolved, while unsupported CASE types conservatively retain pre-existing behavior without introducing a regression. Current-head CI is green; local focused tests could not run because libmo.dylib is absent.

mergify · 2026-07-22T06:43:32Z

Merge Queue Status

✅ Entered queue — 2026-07-22 06:43 UTC · Rule: main · triggered by rule Automatic queue on approval for main
🟠 Checks running · in-place
🚫 Left the queue — 2026-07-22 08:30 UTC · at 090ebf69767a58bf4381635cbeb2e0434868c474

This pull request spent 1 hour 47 minutes 16 seconds in the queue, with no time running CI.

Waiting for

All conditions

Reason

Pull request #25920 has been dequeued

Pull request from fork cannot be queued. This pull request comes from a fork, and Mergify needs the author's permission to update its branch.

The author needs to enable "Allow edits from maintainers" on this pull request.

Failing checks:

Hint

You should look at the reason for the failure and decide if the pull request needs to be fixed or if you want to requeue it.
If you do update this pull request, it will automatically be requeued once the queue conditions match again.
If you think this was a flaky issue instead, you can requeue the pull request, without updating it, by posting a @mergifyio queue comment.

Tick the box to put this pull request back in the merge queue (same as @mergifyio queue).

Requeue this pull request

fix(plan): handle correlated scalar outer projections

cb6f856

aunjgr temporarily deployed to ci July 20, 2026 16:49 — with GitHub Actions Inactive

mergify Bot added the kind/bug Something isn't working label Jul 20, 2026

aunjgr had a problem deploying to ci July 20, 2026 16:50 — with GitHub Actions Failure

aunjgr temporarily deployed to ci July 20, 2026 16:50 — with GitHub Actions Inactive

matrix-meow added the size/M Denotes a PR that changes [100,499] lines label Jul 20, 2026

aunjgr marked this pull request as ready for review July 20, 2026 16:51

aunjgr requested review from XuPeng-SH and heni02 as code owners July 20, 2026 16:52

aunjgr added 2 commits July 21, 2026 22:13

test(plan): cover scalar projection limit fallback

46306e4

Merge remote-tracking branch 'upstream/main' into fix/23062-correlate…

7bd9467

…d-scalar-projection

aunjgr had a problem deploying to ci July 21, 2026 14:15 — with GitHub Actions Error

aunjgr temporarily deployed to ci July 21, 2026 14:15 — with GitHub Actions Inactive

aunjgr had a problem deploying to ci July 21, 2026 14:15 — with GitHub Actions Error

Merge branch 'main' into fix/23062-correlated-scalar-projection

65f264c

aunjgr temporarily deployed to ci July 21, 2026 14:18 — with GitHub Actions Inactive

aunjgr temporarily deployed to ci July 21, 2026 14:19 — with GitHub Actions Inactive

aunjgr had a problem deploying to ci July 21, 2026 14:19 — with GitHub Actions Failure

aunjgr added 2 commits July 22, 2026 00:08

Merge branch 'main' into fix/23062-correlated-scalar-projection

0f2a02b

test(plan): cover scalar projection fallbacks

e8c56ff

aunjgr temporarily deployed to ci July 21, 2026 16:14 — with GitHub Actions Inactive

aunjgr requested review from XuPeng-SH and gouhongshen July 22, 2026 04:51

XuPeng-SH approved these changes Jul 22, 2026

View reviewed changes

Merge branch 'main' into fix/23062-correlated-scalar-projection

e65e3fa

XuPeng-SH temporarily deployed to ci July 22, 2026 05:13 — with GitHub Actions Inactive

XuPeng-SH temporarily deployed to ci July 22, 2026 05:14 — with GitHub Actions Inactive

heni02 approved these changes Jul 22, 2026

View reviewed changes

gouhongshen approved these changes Jul 22, 2026

View reviewed changes

mergify Bot added the queued label Jul 22, 2026

Merge branch 'main' into fix/23062-correlated-scalar-projection

090ebf6

mergify Bot temporarily deployed to ci July 22, 2026 08:17 Inactive

mergify Bot had a problem deploying to ci July 22, 2026 08:17 Failure

mergify Bot temporarily deployed to ci July 22, 2026 08:17 Inactive

mergify Bot requested a deployment to ci July 22, 2026 08:17 In progress

XuPeng-SH merged commit 2d2a11b into matrixorigin:main Jul 22, 2026
22 of 25 checks passed

XuPeng-SH had a problem deploying to ci July 22, 2026 08:30 — with GitHub Actions Failure

XuPeng-SH temporarily deployed to ci July 22, 2026 08:30 — with GitHub Actions Inactive

mergify Bot added dequeued and removed queued labels Jul 22, 2026

XuPeng-SH temporarily deployed to ci July 22, 2026 08:40 — with GitHub Actions Inactive

XuPeng-SH requested a deployment to ci July 22, 2026 08:46 — with GitHub Actions In progress

Ariznawlll mentioned this pull request Jul 22, 2026

Internal error when scalar subquery returns more than one row #23062

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(plan): handle correlated scalar outer projections#25920

fix(plan): handle correlated scalar outer projections#25920
XuPeng-SH merged 10 commits into
matrixorigin:mainfrom
aunjgr:fix/23062-correlated-scalar-projection

aunjgr commented Jul 20, 2026

Uh oh!

qodo-code-review Bot commented Jul 20, 2026

Uh oh!

aunjgr commented Jul 22, 2026

Uh oh!

XuPeng-SH left a comment

Uh oh!

gouhongshen left a comment

Uh oh!

mergify Bot commented Jul 22, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

Uh oh!

Conversation

aunjgr commented Jul 20, 2026

What type of PR is this?

Which issue(s) this PR fixes:

What this PR does / why we need it:

Testing

Uh oh!

qodo-code-review Bot commented Jul 20, 2026

Qodo reviews are paused for this user.

Uh oh!

aunjgr commented Jul 22, 2026

Uh oh!

XuPeng-SH left a comment

Choose a reason for hiding this comment

Uh oh!

gouhongshen left a comment

Choose a reason for hiding this comment

Codex automated review

Uh oh!

mergify Bot commented Jul 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merge Queue Status

Reason

Hint

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

mergify Bot commented Jul 22, 2026 •

edited

Loading