Skip to content

fix(python): Improve schema inference for null-first rows#22923

Draft
5vb76 wants to merge 3 commits intopola-rs:mainfrom
5vb76:main
Draft

fix(python): Improve schema inference for null-first rows#22923
5vb76 wants to merge 3 commits intopola-rs:mainfrom
5vb76:main

Conversation

@5vb76
Copy link
Copy Markdown

@5vb76 5vb76 commented May 24, 2025

Closes #21537

Updated the test suite so that running make test now produces the expected output.
Enhance schema‐inference robustness: When infer_schema_length=1 and the first row contains only NULL values, Polars will now scan subsequent rows to determine each column’s actual type—preventing premature failures and improving the user experience.
Test updates: Modified test_infer_schema_length to validate this new, more resilient behavior.

@github-actions github-actions Bot added fix Bug fix python Related to Python Polars labels May 24, 2025
@5vb76
Copy link
Copy Markdown
Author

5vb76 commented May 25, 2025

Hi, reviewers.
I already fix the issue #21537, but all failing checks show I cannot pass the one test named 'test_frame_from_pydecimal_and_ints'. Could u kindly help me with that? I can pass 'make test' in my own repo.

@5vb76 5vb76 marked this pull request as draft May 29, 2025 03:15
@ritchie46 ritchie46 force-pushed the main branch 3 times, most recently from ddf5907 to d0914d4 Compare September 27, 2025 11:06
@ritchie46 ritchie46 force-pushed the main branch 3 times, most recently from 90ceb7b to e9fce55 Compare October 26, 2025 16:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

fix Bug fix python Related to Python Polars

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Decimals in DataFrame() with orient="row" result in nulls

1 participant