seddonm1 commented on pull request #9243: URL: https://github.com/apache/arrow/pull/9243#issuecomment-779029428
@jorgecarleitao OK I have had a good look at the new API which I believe is on the whole good (and will allow some impressive optimisations) but has led to some very big functions trying to pattern match all the cases: https://github.com/seddonm1/arrow/blob/aaf76f5a7d7a871a2cdb040839376ac1aac29c0b/rust/datafusion/src/physical_plan/string_expressions.rs#L125 So `ltrim` as two different signatures: [`utf8`/`largeutf8`] or [`utf8`/`largeutf8`, `utf8`/`largeutf8`]. Once you expand this out for these signatures and the fact that each argument could be either `array` or `scalar` I am getting 16 unique combinations. I found when trying to actually implement the string functions that the generic case is actually not that common as most functions have different signatures. Could you have a look and advise? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
