shehabgamin commented on PR #14268: URL: https://github.com/apache/datafusion/pull/14268#issuecomment-2629245515
> If the approach looks good to you too, we can go ahead! Sounds good! This works for now since we'll be working towards a better design in the future. I may add a new `TypeSignatureClass` in this pr to expose `default_cast_for` so that it's accessible by downstream users. I'll also update the doc comments to accurately reflect the behavior of `Coercible`. I'll get to this sometime in the next few hours. Taking a detour to review and comment on https://github.com/apache/datafusion/pull/14392. > We already have user defined coercion for functions, #14296 is for binary op or case expression and so on. > > If we agree to use User-defined coercion for functions everywhere, modify coerce types are the easier fix that can go into release quicker. Another better design is to redesign a good method for the function trait, I will also think about the design too. I think a redesign is the path forward. As long as we offer flexibility for customization (e.g., users can override the default signature), then it doesn't really matter what the default implemented signature is, because we won't have to worry about conflicting requirements. -- 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: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
