Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Enhancement] support constant folding in translating scalarOperator phase (backport #50734) #50848

Merged
merged 6 commits into from
Sep 19, 2024

Conversation

mergify[bot]
Copy link
Contributor

@mergify mergify bot commented Sep 9, 2024

Why I'm doing:

Failed to execute query like

with cc as (select 1 as a) select percentile_approx(1, cc.a) from cc;

Because of

if (!functionCallExpr.getChild(1).getType().isNumericType() ||
                    !functionCallExpr.getChild(1).isConstant()) {
                throw new SemanticException(
                        "percentile_approx requires the second parameter's type is numeric constant type");
            }

What I'm doing:

After #49724, we will replace columnRef to constant if it's possible, but sometimes the translator may directly invoke visit() method instead of visitSlot() method.
So return constant if it's possible in visit() method.

Actually, this pr still can't resolve the constant folding totally. The below method can still return a columnRef even when it can be replaced by a constant.

public static ScalarOperator translate(Expr expression, ExpressionMapping expressionMapping,
                                           List<ColumnRefOperator> correlation, ColumnRefFactory columnRefFactory,
                                           ConnectContext session, CTETransformerContext cteContext,
                                           OptExprBuilder builder,
                                           Map<ScalarOperator, SubqueryOperator> subqueryPlaceholders,
                                           boolean useSemiAnti) {
        ColumnRefOperator columnRefOperator = expressionMapping.get(expression);
        if (columnRefOperator != null) {
            return columnRefOperator;
        }
}

SQL like

with cc as (select *, 1 as a from t0) select v1, row_number() over (order by a) from cc

didn't replace the a to 1.

Fixes #issue

What type of PR is this:

  • BugFix
  • Feature
  • Enhancement
  • Refactor
  • UT
  • Doc
  • Tool

Does this PR entail a change in behavior?

  • Yes, this PR will result in a change in behavior.
  • No, this PR will not result in a change in behavior.

If yes, please specify the type of change:

  • Interface/UI changes: syntax, type conversion, expression evaluation, display information
  • Parameter changes: default values, similar parameters but with different default values
  • Policy changes: use new policy to replace old one, functionality automatically enabled
  • Feature removed
  • Miscellaneous: upgrade & downgrade compatibility, etc.

Checklist:

  • I have added test cases for my bug fix or my new feature
  • This pr needs user documentation (for new or modified features or behaviors)
    • I have added documentation for my new feature or new function
  • This is a backport pr

Bugfix cherry-pick branch check:

  • I have checked the version labels which the pr will be auto-backported to the target branch
    • 3.3
    • 3.2
    • 3.1
    • 3.0
    • 2.5

This is an automatic backport of pull request #50734 done by [Mergify](https://mergify.com). ## Why I'm doing: Failed to execute query like ``` with cc as (select 1 as a) select percentile_approx(1, cc.a) from cc; ``` Because of ``` if (!functionCallExpr.getChild(1).getType().isNumericType() || !functionCallExpr.getChild(1).isConstant()) { throw new SemanticException( "percentile_approx requires the second parameter's type is numeric constant type"); } ```

What I'm doing:

After #49724, we will replace columnRef to constant if it's possible, but sometimes the translator may directly invoke visit() method instead of visitSlot() method.
So return constant if it's possible in visit() method.

Actually, this pr still can't resolve the constant folding totally. The below method can still return a columnRef even when it can be replaced by a constant.

public static ScalarOperator translate(Expr expression, ExpressionMapping expressionMapping,
                                           List<ColumnRefOperator> correlation, ColumnRefFactory columnRefFactory,
                                           ConnectContext session, CTETransformerContext cteContext,
                                           OptExprBuilder builder,
                                           Map<ScalarOperator, SubqueryOperator> subqueryPlaceholders,
                                           boolean useSemiAnti) {
        ColumnRefOperator columnRefOperator = expressionMapping.get(expression);
        if (columnRefOperator != null) {
            return columnRefOperator;
        }
}

SQL like

with cc as (select *, 1 as a from t0) select v1, row_number() over (order by a) from cc

didn't replace the a to 1.

Fixes #issue

What type of PR is this:

  • BugFix
  • Feature
  • Enhancement
  • Refactor
  • UT
  • Doc
  • Tool

Does this PR entail a change in behavior?

  • Yes, this PR will result in a change in behavior.
  • No, this PR will not result in a change in behavior.

If yes, please specify the type of change:

  • Interface/UI changes: syntax, type conversion, expression evaluation, display information
  • Parameter changes: default values, similar parameters but with different default values
  • Policy changes: use new policy to replace old one, functionality automatically enabled
  • Feature removed
  • Miscellaneous: upgrade & downgrade compatibility, etc.

Checklist:

  • I have added test cases for my bug fix or my new feature
  • This pr needs user documentation (for new or modified features or behaviors)
    • I have added documentation for my new feature or new function
  • This is a backport pr

@mergify mergify bot added the conflicts label Sep 9, 2024
Copy link
Contributor Author

mergify bot commented Sep 9, 2024

Cherry-pick of 01f048a has failed:

On branch mergify/bp/branch-3.2/pr-50734
Your branch is up to date with 'origin/branch-3.2'.

You are currently cherry-picking commit 01f048a733.
  (fix conflicts and run "git cherry-pick --continue")
  (use "git cherry-pick --skip" to skip this patch)
  (use "git cherry-pick --abort" to cancel the cherry-pick operation)

Changes to be committed:
	modified:   fe/fe-core/src/main/java/com/starrocks/sql/analyzer/FunctionAnalyzer.java
	modified:   fe/fe-core/src/main/java/com/starrocks/sql/optimizer/transformer/SqlToScalarOperatorTranslator.java
	modified:   fe/fe-core/src/main/java/com/starrocks/sql/optimizer/validate/TypeChecker.java
	modified:   fe/fe-core/src/test/java/com/starrocks/sql/analyzer/AnalyzeAggregateTest.java
	modified:   fe/fe-core/src/test/java/com/starrocks/sql/plan/DistinctAggTest.java
	modified:   fe/fe-core/src/test/java/com/starrocks/sql/plan/WindowTest.java

Unmerged paths:
  (use "git add/rm <file>..." as appropriate to mark resolution)
	both modified:   fe/fe-core/src/test/java/com/starrocks/analysis/SelectStmtTest.java
	both modified:   fe/fe-core/src/test/java/com/starrocks/sql/plan/AggregateTest.java
	both modified:   fe/fe-core/src/test/java/com/starrocks/sql/plan/JoinTest.java
	both modified:   fe/fe-core/src/test/java/com/starrocks/sql/plan/SetTest.java
	deleted by us:   test/sql/test_agg_function/R/test_percentile_approx
	deleted by us:   test/sql/test_agg_function/T/test_percentile_approx

To fix up this pull request, you can check it out locally. See documentation: https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/reviewing-changes-in-pull-requests/checking-out-pull-requests-locally

@wanpengfei-git wanpengfei-git enabled auto-merge (squash) September 9, 2024 08:17
@mergify mergify bot closed this Sep 9, 2024
auto-merge was automatically disabled September 9, 2024 08:18

Pull request was closed

Copy link
Contributor Author

mergify bot commented Sep 9, 2024

@mergify[bot]: Backport conflict, please reslove the conflict and resubmit the pr

@mergify mergify bot deleted the mergify/bp/branch-3.2/pr-50734 branch September 9, 2024 08:18
@packy92 packy92 restored the mergify/bp/branch-3.2/pr-50734 branch September 9, 2024 08:45
@packy92 packy92 reopened this Sep 9, 2024
@wanpengfei-git wanpengfei-git enabled auto-merge (squash) September 9, 2024 10:58
@packy92 packy92 closed this Sep 9, 2024
auto-merge was automatically disabled September 9, 2024 11:11

Pull request was closed

@mergify mergify bot deleted the mergify/bp/branch-3.2/pr-50734 branch September 9, 2024 11:11
@packy92 packy92 restored the mergify/bp/branch-3.2/pr-50734 branch September 11, 2024 05:56
@packy92 packy92 reopened this Sep 11, 2024
@wanpengfei-git wanpengfei-git enabled auto-merge (squash) September 11, 2024 05:57
@packy92 packy92 force-pushed the mergify/bp/branch-3.2/pr-50734 branch from 3eefdad to 85157e1 Compare September 11, 2024 10:54
Signed-off-by: packy92 <[email protected]>
Signed-off-by: packy92 <[email protected]>
Copy link

sonarcloud bot commented Sep 13, 2024

@wanpengfei-git wanpengfei-git merged commit b6e1c46 into branch-3.2 Sep 19, 2024
30 of 31 checks passed
@wanpengfei-git wanpengfei-git deleted the mergify/bp/branch-3.2/pr-50734 branch September 19, 2024 09:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants