[ARROW-14448] [Python] Update pyarrow.array() docstring note on timestamp (timezone) conversion

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 7.0.0
• Component/s: Python
• Labels: pull-request-available
• External issue URL: https://github.com/apache/arrow/issues/18906
• Assignee: Sanjiban Sengupta
• Reporter: Joris Van den Bossche
• Created: 22/Oct/21 18:52
• Updated: 11/Jan/23 08:40
• Resolved: 11/Jan/22 09:52

Description

The current pa.array(..) docstring has the following note:

Localized timestamps will currently be returned as UTC (pandas’s native representation). Timezone-naive data will be implicitly interpreted as UTC.

This is IMO quite confusing. While technically somewhat correct (we do store UTC values under the hood for timezone aware data, and values "as if" in UTC for timezone naive data), it's not what the user actually gets returned. The type of the returned array does preserve the timezone if possible, or will have no timezone (also no UTC) when converted from naive timestamps.

Issue Links

• links to: GitHub Pull Request #12078