[Enhancement](parquet)update runtime filter when read next parquet row group. #59053
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 |
hubgeter
force-pushed
the
join_rf_parquet_2
branch
from
b0d62d6 to
fcacd4c
Compare
Contributor
Possible file(s) that should be tracked in LFS detected: 🚨The following file(s) exceeds the file size limit:
Consider using |
github-actions
bot
added
the
lfs-detected!
Contributor
Author
|
run buildall |
github-actions
bot
removed
the
lfs-detected!
TPC-H: Total hot run time: 32192 ms |
TPC-DS: Total hot run time: 176509 ms |
ClickBench: Total hot run time: 27.14 s |
Contributor
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
morningman
previously approved these changes
Dec 16, 2025
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. |
There was a problem hiding this comment.
Pull request overview
This PR enhances Parquet row group filtering by fetching the latest join runtime filters when creating each row group reader, rather than only at the file level. This allows for more efficient data filtering as runtime filters become available during query execution.
Key Changes:
- Added runtime filter update mechanism that is called when creating new Parquet row group readers
- Introduced test coverage to verify that runtime filters are applied across multiple row groups
Reviewed changes
Copilot reviewed 9 out of 11 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| regression-test/suites/external_table_p0/hive/test_parquet_join_runtime_filter.groovy | Adds comprehensive test suite for verifying runtime filter application across Parquet row groups |
| docker/thirdparties/docker-compose/hive/scripts/create_preinstalled_scripts/run84.hql | Creates test tables for runtime filter validation |
| be/src/vec/exec/scan/scanner.cpp | Initializes total runtime filter count during scanner construction |
| be/src/vec/exec/scan/file_scanner.h | Updates method signature to track runtime filter changes |
| be/src/vec/exec/scan/file_scanner.cpp | Implements runtime filter update callback for Parquet readers |
| be/src/vec/exec/format/parquet/vparquet_reader.h | Adds callback mechanism and lazy read context update method |
| be/src/vec/exec/format/parquet/vparquet_reader.cpp | Implements runtime filter updates when creating row group readers |
| be/src/vec/exec/format/parquet/vparquet_group_reader.h | Extends LazyReadContext to include partition and missing column information |
| be/src/runtime_filter/runtime_filter_consumer_helper.h | Exposes runtime filter count accessor |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| _output_tuple_desc(_local_state->output_tuple_desc()), | ||
| _output_row_descriptor(_local_state->_parent->output_row_descriptor()), | ||
| _has_prepared(false) { | ||
| _total_rf_num = cast_set<int>(_local_state->_helper.runtime_filter_nums()); |
There was a problem hiding this comment.
The function name cast_set is ambiguous and doesn't clearly convey its purpose. Consider renaming to something more descriptive like set_int_value or using direct casting syntax like static_cast<int>.
Suggested change
| _total_rf_num = cast_set<int>(_local_state->_helper.runtime_filter_nums()); | |
| _total_rf_num = static_cast<int>(_local_state->_helper.runtime_filter_nums()); |
Comment on lines
+414
to
+412
| new_lazy_read_ctx.fill_partition_columns = std::move(_lazy_read_ctx.fill_partition_columns); | ||
| new_lazy_read_ctx.fill_missing_columns = std::move(_lazy_read_ctx.fill_missing_columns); |
There was a problem hiding this comment.
Moving from _lazy_read_ctx and then assigning back to it on line 416 leaves the source in a moved-from state. Consider copying these values instead of moving, or restructure to avoid the circular dependency.
Suggested change
| new_lazy_read_ctx.fill_partition_columns = std::move(_lazy_read_ctx.fill_partition_columns); | |
| new_lazy_read_ctx.fill_missing_columns = std::move(_lazy_read_ctx.fill_missing_columns); | |
| new_lazy_read_ctx.fill_partition_columns = _lazy_read_ctx.fill_partition_columns; | |
| new_lazy_read_ctx.fill_missing_columns = _lazy_read_ctx.fill_missing_columns; |
Comment on lines
356
to
361
| // when create new row group reader, call this function to get lasted runtime filter conjuncts. | ||
| std::function<Status(bool*, VExprContextSPtrs&)> _call_late_rf_func = [](bool* changed, | ||
| VExprContextSPtrs&) { | ||
| *changed = false; | ||
| return Status::OK(); | ||
| }; |
There was a problem hiding this comment.
The default lambda implementation always sets changed to false and returns OK. Consider documenting why this is the default behavior or making it more explicit with a named static function.
Suggested change
| // when create new row group reader, call this function to get lasted runtime filter conjuncts. | |
| std::function<Status(bool*, VExprContextSPtrs&)> _call_late_rf_func = [](bool* changed, | |
| VExprContextSPtrs&) { | |
| *changed = false; | |
| return Status::OK(); | |
| }; | |
| // when creating a new row group reader, call this function to get the latest runtime filter conjuncts. | |
| // The default implementation does nothing, sets 'changed' to false, and returns OK. | |
| // This is used when no late runtime filter is required. | |
| static Status default_late_rf_func(bool* changed, VExprContextSPtrs&) { | |
| *changed = false; | |
| return Status::OK(); | |
| } | |
| std::function<Status(bool*, VExprContextSPtrs&)> _call_late_rf_func = default_late_rf_func; |
morningman
force-pushed
the
join_rf_parquet_2
branch
from
71cefb9 to
44fa16d
Compare
Contributor
|
run buildall |
TPC-H: Total hot run time: 32772 ms |
TPC-DS: Total hot run time: 176928 ms |
ClickBench: Total hot run time: 27.49 s |
Contributor
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
hubgeter
force-pushed
the
join_rf_parquet_2
branch
from
44fa16d to
2937fe0
Compare
Contributor
Author
|
run buildall |
github-actions
bot
removed
the
approved
TPC-H: Total hot run time: 32265 ms |
TPC-DS: Total hot run time: 177264 ms |
ClickBench: Total hot run time: 27.39 s |
Contributor
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
16 tasks
morningman
pushed a commit
that referenced
this pull request
Dec 19, 2025
Contributor
Author
|
run buildall |
hubgeter
force-pushed
the
join_rf_parquet_2
branch
from
0a86b5b to
deb08e8
Compare
Contributor
Author
|
run buildall |
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
morningman
pushed a commit
that referenced
this pull request
Dec 25, 2025
morningman
pushed a commit
that referenced
this pull request
Dec 25, 2025
morningman
pushed a commit
that referenced
this pull request
Dec 26, 2025
github-actions
bot
added
the
approved
Contributor
|
PR approved by at least one committer and no changes requested. |
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
CalvinKirs
approved these changes
Dec 31, 2025
16 tasks
morningman
pushed a commit
that referenced
this pull request
Jan 7, 2026
### What problem does this PR solve? Problem Summary: case from pr : #59053
hubgeter
added a commit
to hubgeter/doris
that referenced
this pull request
Jan 9, 2026
…w group. (apache#59053) Problem Summary: This pull request achieves better filtering by fetching the latest join runtime filter when creating the Parquet row group reader. Previously, the join runtime filter was fetched at the Parquet file level.
hubgeter
added a commit
to hubgeter/doris
that referenced
this pull request
Jan 9, 2026
### What problem does this PR solve? Problem Summary: case from pr : apache#59053
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.
What problem does this PR solve?
Problem Summary:
This pull request achieves better filtering by fetching the latest join runtime filter when creating the Parquet row group reader. Previously, the join runtime filter was fetched at the Parquet file level.
Release note
None
Check List (For Author)
Test
Behavior changed:
Does this need documentation?
Check List (For Reviewer who merge this PR)