Github user fhueske commented on the pull request:
https://github.com/apache/flink/pull/426#issuecomment-77358774
Those are very good points! Having an annotation for POJO fields to define
their order in the PojoTypeInfo (and therefore the order in which the fields
are mapped to CSV fields) is a nice idea.
The String[] parameter to select POJO fields would still be good because it
- makes the POJO type and the CSV field independent of each other and
- allows to fill only a subset of POJO fields.
Having a way to provide logic to parse a String into a
custom/not-natively-supported type is also important for POJOs. I would
however, make that independent of the POJO and add it as a parameter to the
CsvReader. That way you don't need to touch existing POJO code and make it
again independent of the data source (imagine CSV files with different date
formats).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
