[fix](repeat plan) fix repeat output slot order inconsistent with its input expression #60045
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Contributor
|
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
Contributor
Author
|
run buildall |
TPC-H: Total hot run time: 31367 ms |
TPC-DS: Total hot run time: 173815 ms |
ClickBench: Total hot run time: 26.69 s |
Contributor
FE UT Coverage ReportIncrement line coverage |
Contributor
FE Regression Coverage ReportIncrement line coverage |
There was a problem hiding this comment.
Pull request overview
This pull request fixes a critical bug in the Repeat node's output slot ordering in the Nereids planner. The BE requires that the repeat node's output slots follow a specific order: input expressions + GroupingID + other grouping functions. The previous physical translator implementation did not guarantee this ordering, which caused bad cast exceptions in the backend when executing queries with GROUPING SETS.
Changes:
- Refactored the PhysicalPlanTranslator to ensure correct output slot ordering for Repeat nodes
- Added comprehensive regression tests covering the fixed scenarios
- Fixed the logic to explicitly add group by expressions first, followed by aggregate function slots, then GroupingID, and finally grouping function slots
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java | Refactored visitPhysicalRepeat method to ensure correct output slot ordering: group by expressions → aggregate function slots → grouping ID → grouping function slots |
| regression-test/suites/nereids_p0/repeat/test_repeat_output_slot.groovy | Added regression test with two SQL queries using GROUPING SETS to verify the fix |
| regression-test/data/nereids_p0/repeat/test_repeat_output_slot.out | Expected output for the regression tests, including query plans and results |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java
Outdated
Show resolved
Hide resolved
fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java
Show resolved
Hide resolved
fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java
Outdated
Show resolved
Hide resolved
morrySnow
reviewed
Jan 21, 2026
Contributor
morrySnow
left a comment
morrySnow
left a comment
There was a problem hiding this comment.
add a ut for translate repeat
fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java
Outdated
Show resolved
Hide resolved
fe/fe-core/src/main/java/org/apache/doris/nereids/glue/translator/PhysicalPlanTranslator.java
Outdated
Show resolved
Hide resolved
Contributor
Author
|
run buildall |
Contributor
Author
|
run buildall |
TPC-H: Total hot run time: 31381 ms |
TPC-DS: Total hot run time: 173591 ms |
ClickBench: Total hot run time: 26.56 s |
Contributor
FE UT Coverage ReportIncrement line coverage |
Contributor
FE Regression Coverage ReportIncrement line coverage |
starocean999
approved these changes
Jan 22, 2026
github-actions
bot
added
the
approved
Contributor
|
PR approved by at least one committer and no changes requested. |
Contributor
|
PR approved by anyone and no changes requested. |
morrySnow
approved these changes
Jan 23, 2026
github-actions bot
pushed a commit
that referenced
this pull request
Jan 23, 2026
… input expression (#60045) maybe relate PR: #21168 BE requires that the repeat node's output slot order should be inconsistent with its input expressions. That is output slots = input expressions + GroupingID + other grouping functions. But physical translator not ensure this requirement. Then sometimes the repeat may have bad cast exception. for sql: SELECT 100000 FROM db2.table_9_50_undef_partitions2_keys3_properties4_distributed_by5 GROUP BY GROUPING SETS ( (col_datetime_6__undef_signed, col_varchar_50__undef_signed) , () , (col_varchar_50__undef_signed) , (col_datetime_6__undef_signed, col_varchar_50__undef_signed) ); the above sql will have wrong ouput slot order then BE will have exceptions: (1105, 'errCode = 2, detailMessage = (172.20.57.146) [E-7412]assert cast err:[E-7412] Bad cast from type:doris::vectorized::ColumnVector<(doris::PrimitiveType)26> to doris::vectorized::ColumnStr<unsigned int> 0#doris::Exception::Exception(int, std::basic_string_view<char, std::char_traits<char> > const&, bool) at /home/zcp/repo_center/doris_master/doris/be/src/common/exception.cpp:0\n\t1# ...
yujun777
added a commit
to yujun777/doris
that referenced
this pull request
Jan 23, 2026
… input expression (apache#60045) maybe relate PR: apache#21168 BE requires that the repeat node's output slot order should be inconsistent with its input expressions. That is output slots = input expressions + GroupingID + other grouping functions. But physical translator not ensure this requirement. Then sometimes the repeat may have bad cast exception. for sql: SELECT 100000 FROM db2.table_9_50_undef_partitions2_keys3_properties4_distributed_by5 GROUP BY GROUPING SETS ( (col_datetime_6__undef_signed, col_varchar_50__undef_signed) , () , (col_varchar_50__undef_signed) , (col_datetime_6__undef_signed, col_varchar_50__undef_signed) ); the above sql will have wrong ouput slot order then BE will have exceptions: (1105, 'errCode = 2, detailMessage = (172.20.57.146) [E-7412]assert cast err:[E-7412] Bad cast from type:doris::vectorized::ColumnVector<(doris::PrimitiveType)26> to doris::vectorized::ColumnStr<unsigned int> 0#doris::Exception::Exception(int, std::basic_string_view<char, std::char_traits<char> > const&, bool) at /home/zcp/repo_center/doris_master/doris/be/src/common/exception.cpp:0\n\t1# ...
16 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
intro by #59116
BE requires that the repeat node's output slot order should be inconsistent with its input expressions.
That is output slots = input expressions + GroupingID + other grouping functions.
But physical translator not ensure this requirement. Then sometimes the repeat may have bad cast exception.
for sql:
the above sql will have wrong ouput slot order:
then BE will have exceptions:
relate PR: #59116 may cause repeat's group by expression order not match repeat's output expression
Release note
None
Check List (For Author)
Test
Behavior changed:
Does this need documentation?
Check List (For Reviewer who merge this PR)