[ARROW-13224] [Python][Doc] Documentation missing for pyarrow.dataset.write_dataset

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 5.0.0
• Component/s: Documentation, Python
• Labels: pull-request-available
• External issue URL: https://github.com/apache/arrow/issues/28911
• Assignee: Weston Pace
• Reporter: Weston Pace
• Created: 30/Jun/21 18:24
• Updated: 11/Jan/23 08:31
• Resolved: 20/Jul/21 13:38

Description

I don't believe this is meant to be internal.  pyarrow.parquet.write_to_dataset uses this (if use_legacy_dataset=False) but the parquet API doesn't expose the same features.  A new example should also probably be added to the Tabular Datasets section of the docs explaining why write_dataset can take in a scanner (e.g. memory preserving, ability to write a dataset from flight or any record batch source, etc.)

Issue Links

• is duplicated by: ARROW-13207 [Python][Doc] Dataset documentation still suggests deprecated scan method as the preferred iterative approach
• links to: GitHub Pull Request #10693