Github user twalthr commented on the issue: https://github.com/apache/flink/pull/5240 Thanks for the feedback @fhueske. I hope I could address most of it. I think we should merge this PR (if you agree) and add more PRs for this issue as the next steps. I suggest the following subtasks: - Add validation for the CSV format - Add full CsvTableSourceFactory support (incl. proctime, rowtime, and schema mapping) - Add a JSON schema parser to the JSON and logic for creating a table source from it - Add validation for the JSON format - Add validation for the Rowtime descriptor - Add validation for StreamTableDescriptor - Add validation for BatchTableDescriptor - Add KafkaTableSource factories What do you think?
---