Fix stale num_rows argument breaking mypy in cudf-polars duplicated-output path by galipremsagar · Pull Request #23303 · NVIDIA/cudf

galipremsagar · 2026-07-16T19:03:00Z

Description

DataFrame.from_table lost its num_rows parameter in #23234 (row counts are now inferred from the pylibcudf table, which carries them even for zero-column tables), but the duplicated-output path added in #23114 still passes num_rows=0. The two PRs merged around the same time, so this surfaced only after both landed: mypy now fails on every PR's check-style job (engine/core.py:840: Unexpected keyword argument "num_rows"), and the path would raise TypeError at runtime.

plc.copying.empty_like already produces a 0-row table (including for zero-column inputs, verified), so dropping the argument preserves the intended "freshly-allocated empty same-schema frame" semantics exactly.

Checklist

I am familiar with the Contributing Guidelines.
New or existing tests cover these changes.
The documentation is up to date with these changes.

copy-pr-bot · 2026-07-16T19:03:04Z

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

coderabbitai · 2026-07-16T19:03:49Z

📝 Walkthrough

Walkthrough

drop_if_replicated now creates empty non-root outputs from the original table while preserving column names, dtypes, and execution stream.

Changes

Replicated output handling

Layer / File(s)	Summary
Schema-preserving empty frame construction `python/cudf_polars/cudf_polars/engine/core.py`	The duplicated-output path uses `plc.copying.empty_like` and preserves the original DataFrame schema and stream metadata.

Estimated code review effort: 1 (Trivial) | ~3 minutes

Possibly related PRs

rapidsai/cudf#23234: Both PRs adjust empty DataFrame construction to preserve row or schema information.

Suggested labels: improvement

Suggested reviewers: madsbk, brandon-b-miller, pointkernel, vyasr

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title clearly states the core fix: removing a stale num_rows argument in the cudf-polars duplicated-output path.
Description check	✅ Passed	The description matches the change and explains the mypy/runtime issue and why dropping num_rows is correct.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

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

galipremsagar · 2026-07-16T19:07:03Z

/okay to test f900238

galipremsagar · 2026-07-17T00:53:52Z

/okay to test b4053f6

galipremsagar · 2026-07-17T01:02:08Z

/okay to test b4053f6

…ction DataFrame.from_table lost its num_rows parameter in NVIDIA#23234 (row counts are now inferred from the pylibcudf table, which carries them even for zero-column tables), but the duplicated-output path added in NVIDIA#23114 still passed num_rows=0, breaking mypy on every PR and raising TypeError at runtime on that path. empty_like already produces a 0-row table (including for zero-column inputs), so the argument was redundant.

galipremsagar · 2026-07-17T01:09:53Z

/okay to test 3a35541

galipremsagar · 2026-07-17T01:15:40Z

/merge

wence- · 2026-07-17T10:24:20Z

/merge

galipremsagar · 2026-07-17T12:13:51Z

/merge

galipremsagar · 2026-07-17T13:06:35Z

/merge

galipremsagar · 2026-07-17T13:10:03Z

/merge

galipremsagar · 2026-07-17T13:11:30Z

/merge

galipremsagar · 2026-07-17T13:42:33Z

/merge

…utput path (NVIDIA#23303) `DataFrame.from_table` lost its `num_rows` parameter in NVIDIA#23234 (row counts are now inferred from the pylibcudf table, which carries them even for zero-column tables), but the duplicated-output path added in NVIDIA#23114 still passes `num_rows=0`. The two PRs merged around the same time, so this surfaced only after both landed: mypy now fails on every PR's `check-style` job (`engine/core.py:840: Unexpected keyword argument "num_rows"`), and the path would raise `TypeError` at runtime. `plc.copying.empty_like` already produces a 0-row table (including for zero-column inputs, verified), so dropping the argument preserves the intended "freshly-allocated empty same-schema frame" semantics exactly. Authors: - GALI PREM SAGAR (https://github.com/galipremsagar) Approvers: - Matthew Roeschke (https://github.com/mroeschke) - Lawrence Mitchell (https://github.com/wence-) URL: NVIDIA#23303

galipremsagar requested a review from a team as a code owner July 16, 2026 19:03

galipremsagar added bug Something isn't working Python Affects Python cuDF API. non-breaking Non-breaking change cudf-polars Issues specific to cudf-polars labels Jul 16, 2026

galipremsagar requested a review from madsbk July 16, 2026 19:03

github-project-automation Bot added this to cuDF Python Jul 16, 2026

github-actions Bot assigned galipremsagar Jul 16, 2026

GPUtester moved this to In Progress in cuDF Python Jul 16, 2026

mroeschke approved these changes Jul 16, 2026

View reviewed changes

TomAugspurger mentioned this pull request Jul 16, 2026

Update to 26.10 #23305

Merged

coderabbitai Bot mentioned this pull request Jul 16, 2026

Use OrderScheme metadata to select order-aware groupby execution #23306

Merged

galipremsagar changed the base branch from main to release/26.08 July 17, 2026 01:01

galipremsagar requested review from a team as code owners July 17, 2026 01:01

galipremsagar requested review from KyleFromNVIDIA, PointKernel, brandon-b-miller and vyasr July 17, 2026 01:01

galipremsagar changed the base branch from release/26.08 to main July 17, 2026 01:03

galipremsagar changed the base branch from main to release/26.08 July 17, 2026 01:08

galipremsagar removed request for a team, KyleFromNVIDIA, PointKernel, brandon-b-miller and vyasr July 17, 2026 01:09

galipremsagar added the 5 - Ready to Merge Testing and reviews complete, ready to merge label Jul 17, 2026

wence- approved these changes Jul 17, 2026

View reviewed changes

galipremsagar changed the title ~~Fix stale num_rows argument breaking mypy in cudf-polars duplicated-output path~~ Fix stale num_rows argument breaking mypy in cudf-polars duplicated-output pat Jul 17, 2026

galipremsagar changed the title ~~Fix stale num_rows argument breaking mypy in cudf-polars duplicated-output pat~~ Fix stale num_rows argument breaking mypy in cudf-polars duplicated-output path Jul 17, 2026

galipremsagar added bug Something isn't working and removed bug Something isn't working labels Jul 17, 2026

rapids-bot Bot merged commit 7b13484 into NVIDIA:release/26.08 Jul 17, 2026
404 of 414 checks passed

github-project-automation Bot moved this from In Progress to Done in cuDF Python Jul 17, 2026

josephine-wolf-oberholtzer added this to cuDF Python Aug 12, 2026

josephine-wolf-oberholtzer moved this to Done in cuDF Python Aug 12, 2026

Uh oh!

Conversation

galipremsagar commented Jul 16, 2026

Description

Checklist

Uh oh!

copy-pr-bot Bot commented Jul 16, 2026

Uh oh!

coderabbitai Bot commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Uh oh!

galipremsagar commented Jul 16, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

wence- commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

galipremsagar commented Jul 17, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

coderabbitai Bot commented Jul 16, 2026 •

edited

Loading