brianrackle edited a comment on issue #929: URL: https://github.com/apache/arrow-rs/issues/929#issuecomment-967735847
> But I think this issue should be about CSV handling of these values and was mistakenly worded to refer to JSON. > > For CSV there is not a formal spec, so the expected behavior is undefined. My thoughts are that since `2.` can be interpreted by an Int then it should be and the usual schema inference rules will upgrade from Int to Float only if there are other values in the column which can only be represented by Float. After playing with the solution some more inferring as an Int would mean changing the native parse logic which currently does not support converting "2." to an Int. So I followed the path of least resistance and went ahead with doing a conversion which is already supported and infer "2." as a Float so that the existing native parsing logic can be used. -- 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]
