[
https://issues.apache.org/jira/browse/ARROW-8111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17058819#comment-17058819
]
Wes McKinney edited comment on ARROW-8111 at 3/13/20, 3:04 PM:
---------------------------------------------------------------
Right, but in terms of creating a public API to expose this functionality, I
think it would be better to be explicit about what format is expected rather
than unconditionally making an inference (in this particular case, the
inference could be wrong if the data is of non-US origin). If the user
indicates a format that is not supported yet we can say so
was (Author: wesmckinn):
Right, but in terms of creating a public API to expose this functionality, I
think it would be better to be explicit about what format is expected rather
than unconditionally making an inference (in this particular case, the
inference could be wrong if the data is of international origin). If the user
indicates a format that is not supported yet we can say so
> [C++][CSV] Support DD/MM/YYYY date format
> -----------------------------------------
>
> Key: ARROW-8111
> URL: https://issues.apache.org/jira/browse/ARROW-8111
> Project: Apache Arrow
> Issue Type: Improvement
> Components: C++
> Reporter: Artem Alekseev
> Priority: Major
>
> Currently, date parser supports only YYYY-MM-DD format. For our workload we
> need DD/MM/YYYY format. It is obvious that CSV parser should support
> different date formats, so we may start from implementing DD/MM/YYYY format.
> Also, we may use some date parsing library, which would solve the problem for
> us.
> Also, we may need to somehow specify a format for every column in CSV parser.
> If you have any implementation ideas in mind, please share, so that I can
> implement it.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
