jorisvandenbossche opened a new issue, #35448: URL: https://github.com/apache/arrow/issues/35448
### Describe the bug, including details regarding any error messages, version, and platform. https://github.com/apache/arrow/pull/11358 refactored the timezone offset handling in string parsing (both in read_csv, casting str->timestamp and strptime). The general rule was that if there is a `%z` in the format string you provide to `strptime`, we apply the offset and return a timestamp type with tz=UTC. But this seems to work only in certain cases: ``` # space before %z -> works >>> pc.strptime(["5/1/2020 +0100", None, "12/11/1900 -0130"], format="%m/%d/%Y %z", unit='us').type TimestampType(timestamp[us, tz=UTC]) # no space before %z -> doesn't work >>> pc.strptime(["5/1/2020+0100", None, "12/11/1900-0130"], format="%m/%d/%Y%z", unit='us').type TimestampType(timestamp[us]) ``` ### Component(s) C++ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
