harryplumer opened a new issue, #26567: URL: https://github.com/apache/airflow/issues/26567
### Apache Airflow Provider(s) amazon ### Versions of Apache Airflow Providers `apache-airflow-providers-amazon==5.1.0` ### Apache Airflow version 2.3.4 ### Operating System Linux ### Deployment Astronomer ### Deployment details _No response_ ### What happened Once https://github.com/apache/airflow/pull/25083 was merged, when using CSV as the output format, null strings started appearing as `"None"` in the actual CSV export. This will cause unintended behavior in most use cases for reading the CSV including uploading to databases. Certain databases such as Snowflake allow for things like `NULL_IF` on import however there are times where you would want the actual string "None" to be in the field and there would be no way at that point to distinguish. Before:  After:  ### What you think should happen instead The strings should be empty as they did previously. I understand the implementation of the recent PR for parquet and propose that we add an additional condition to line 138 of the `sql_to_s3.py` file restricting that to only if the chosen output is parquet. ### How to reproduce Run the `SqlToS3Operator` with the default output format of `CSV` on any query that selects a column of type string that allows null. Look the outputted CSV in S3. ### Anything else Every time we select a nullable column with the `SqlToS3Operator` ### Are you willing to submit PR? - [X] Yes I am willing to submit a PR! ### Code of Conduct - [X] I agree to follow this project's [Code of Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md) -- 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]
