[
https://issues.apache.org/jira/browse/ARROW-15026?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17456691#comment-17456691
]
Andreas Rappold commented on ARROW-15026:
-----------------------------------------
Thanks, I tried but wasn't fully successful, see
[https://github.com/notEvil/arrow/tree/ARROW-15026]
With 2 additional header files exposed to the python build (the internal and
one dependency) it compiles and does prevent conversions. At this stage I
thought someone more experienced should take over because it would take me ages
to get all the details right.
> [Python] datetime.timedelta to pyarrow.duration('us') silently overflows
> ------------------------------------------------------------------------
>
> Key: ARROW-15026
> URL: https://issues.apache.org/jira/browse/ARROW-15026
> Project: Apache Arrow
> Issue Type: Bug
> Components: Python
> Reporter: Andreas Rappold
> Priority: Major
> Fix For: 7.0.0
>
>
>
> Hi! This reproduces the issue:
> {code:java}
> # python 3.9.9
> # pyarrow 6.0.1
> import datetime
> import pyarrow
> d = datetime.timedelta(days=-106751992, seconds=71945, microseconds=224192)
> pyarrow.scalar(d)
> # <pyarrow.DurationScalar: datetime.timedelta(days=-106751992, seconds=71945,
> microseconds=224192)>
> pyarrow.scalar(d).as_py() == d
> # True
> d2 = d - datetime.timedelta(microseconds=1)
> pyarrow.scalar(d2)
> # <pyarrow.DurationScalar: datetime.timedelta(days=106751991, seconds=14454,
> microseconds=775807)>
> pyarrow.scalar(d2).as_py() == d2
> # False{code}
> Other conversions (e.g. to int*) raise an exception instead. I didn't check
> if duration overflows for too large timedeltas. If its easy to fix, point me
> in the right direction and I try to create a PR. Thanks
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
