danielcweeks commented on PR #9884: URL: https://github.com/apache/iceberg/pull/9884#issuecomment-1986717618
@jbonofre I don't necessarily think this is a good idea. The issue is that if we start using schemes that are not well-defined or standardized, then you can run into compatibility problems. For example, if we write new schemes, then other libraries like python, rust, and go also need to have similar behavior. Some of those may rely on libraries (like arrow) that may not support arbitrary schemes. This could be even more problematic for vendor solutions that don't allow for changing the supported schemes (e.g. bigquery or snowflake). I feel like it's better to identify and add explicitly defined schemes so that we don't run into compatibility issues. I think we should identify specific cases where we would need this as opposed to having a well curated set of schemes. I did see an example of what those might be in the issue. Can we expand on the use case with specific examples before adopting this? -- 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: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
