support sum_int in TiFlash

[windtalker]

What problem does this PR solve?

Issue Number: close #10720

Problem Summary:

What is changed and how it works?

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

None

Summary by CodeRabbit

• New Features

  • Added SumInt aggregation function support for integer sum calculations.

• Chores

  • Updated Docker container image references to new registry paths.
  • Improved branch name sanitization in Docker deployment scripts to handle special characters properly.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR adds support for the SumInt aggregation function in TiFlash by extending aggregation handlers to recognize and process SumInt expressions, adding comprehensive test coverage, and updating Docker image registry paths with branch name sanitization in test scripts.

Changes

Cohort / File(s)	Summary
SumInt Aggregation Support contrib/tipb, dbms/src/Debug/MockExecutor/AggregationBinder.cpp, dbms/src/Debug/MockExecutor/FuncSigMap.cpp, dbms/src/Flash/Coprocessor/DAGUtils.cpp	Extended aggregation function handling to treat SumInt identically to Sum by adding mapping entries, updating conditional checks in buildAggFunc, and including SumInt in aggregation-function recognition in DAGUtils.
SumInt Aggregation Tests dbms/src/Flash/Coprocessor/tests/gtest_sum_int_agg_func.cpp	New GoogleTest file covering SumInt expression mapping, string rendering, partial result handling, and return type inference for integer inputs across multiple aggregation modes.
Docker Registry Updates tests/docker/cluster.yaml, tests/docker/cluster_disable_new_collation.yaml, tests/docker/cluster_new_collation.yaml, tests/docker/cluster_tidb_fail_point.yaml, tests/docker/tiflash-tagged-image.yaml	Updated container image references from hub.pingcap.net registry to us-docker.pkg.dev/pingcap-testing-account registry for pd, tikv, tidb, and tiflash services.
Test Script Branch Sanitization tests/docker/run.sh, tests/docker/util.sh	Added environment variable exports that sanitize branch and tag names by replacing forward slashes with dashes for PD_BRANCH, TIKV_BRANCH, TIDB_BRANCH, and TAG variables.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Suggested reviewers

• kolafish
• JinheLin

Poem

🐰 Hop hooray! SumInt arrives in TiFlash today,
Aggregations aligned in every single way,
Tests sing and dance with return types true,
While Docker containers pull fresh and new! 🌟

🚥 Pre-merge checks | ✅ 3 | ❌ 3

❌ Failed checks (2 warnings, 1 inconclusive)

Check name	Status	Explanation	Resolution
Out of Scope Changes check	⚠️ Warning	Most changes are in-scope for sum_int support, but updates to Docker image registries in cluster YAML and shell scripts appear unrelated to the sum_int feature implementation.	Move Docker registry updates to a separate PR or provide justification for including them with the sum_int feature implementation.
Docstring Coverage	⚠️ Warning	Docstring coverage is 10.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Description check	❓ Inconclusive	The PR description uses the required template structure but leaves critical sections incomplete: the problem summary and commit message are empty, and no detailed explanation of changes is provided.	Complete the 'Problem Summary' and 'What is changed and how it works' sections with specific details about the sum_int implementation.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title 'support sum_int in TiFlash' clearly and concisely describes the main objective of the PR, which is adding support for the sum_int aggregation function in TiFlash.
Linked Issues check	✅ Passed	The code changes properly implement sum_int support across TiFlash components: AggregationBinder, FuncSigMap, DAGUtils, and include comprehensive unit tests covering type handling and partial result behavior.
Merge Conflict Detection	✅ Passed	✅ No merge conflicts detected when merging into feature/release-8.5-materialized-view

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

No actionable comments were generated in the recent review. 🎉

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[wuhuizuo]

/retest

[coderabbitai]

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

dbms/src/Debug/MockExecutor/AggregationBinder.cpp (1)

205-210: ⚠️ Potential issue | 🟡 Minor

SumInt should infer output type signedness from input type, not set ColumnFlagUnsigned unconditionally.

In buildAggFunc (lines 205-210), SumInt is grouped with Count and Sum to set the output type to TypeLongLong with both ColumnFlagUnsigned and ColumnFlagNotNull. However, based on the return type tests in dbms/src/Flash/Coprocessor/tests/gtest_sum_int_agg_func.cpp, the output signedness should match the input: summing Int64 should produce Int64, not UInt64. The current unconditional ColumnFlagUnsigned is incorrect. Consider examining the input type and only setting the unsigned flag when the input type is unsigned.

Additionally, compileAggregation (lines 272–293) does not handle the "sum_int" function name in its checks. While tests currently bypass this via buildAggFunc, adding "sum_int" to the conditions alongside "sum" would improve consistency.

🧹 Nitpick comments (2)

dbms/src/Flash/Coprocessor/DAGExpressionAnalyzer.cpp (1)

1738-1743: Log level escalation from TRACE to DEBUG may increase log noise in production.

Implicit casts are common in query processing; logging each one at DEBUG level could be verbose. If this change is a temporary debugging aid (consistent with the "[DNM] for test" PR title), consider reverting to TRACE before merging.

dbms/src/Flash/Coprocessor/tests/gtest_sum_int_agg_func.cpp (1)

37-47: Unused template parameter TExpectedOutput.

TExpectedOutput is declared but never referenced in checkSumIntReturnType. The actual verification uses string comparison. Either remove the unused parameter or use it for a type-level assertion (e.g., verifying nested->getReturnType() matches DataTypeNumber<TExpectedOutput>).

♻️ Option A: Remove unused parameter

-template <typename TInput, typename TExpectedOutput>
+template <typename TInput>
 void checkSumIntReturnType(const String & expected_output, const String & expected_output_nullable)

Then update callers:

-    checkSumIntReturnType<Int8, Int64>("Int64", "Nullable(Int64)");
+    checkSumIntReturnType<Int8>("Int64", "Nullable(Int64)");

[wuhuizuo]

wait for #10721

[xzhangxian1008]

LGTM

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: gengliqi, xzhangxian1008

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [gengliqi]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ti-chi-bot]

[LGTM Timeline notifier]

Timeline:

• 2026-02-13 02:42:38.536452892 +0000 UTC m=+159111.616443597: ☑️ agreed by gengliqi.
• 2026-02-13 05:38:06.081915435 +0000 UTC m=+169639.161906140: ☑️ agreed by xzhangxian1008.

[wuhuizuo]

/retest