Omega359 commented on issue #8800: URL: https://github.com/apache/arrow-datafusion/issues/8800#issuecomment-1890781678
> @Omega359 do you think anything still left that can be improved? Yes. Hiding rename case behaviour behind a flag that as far as I can find isn't prominently noted in the docs isn't great. It's really not obvious as a user of the api why it isn't working when you provided the 'correct' name. Having to dig into the source code, while awesome to have it available, is not ideal to solving an issue like this, especially for users who are still getting up to speed on rust. I didn't have a solution to my issue until a core developer popped into the discord channel to help. A few possible solutions I see of which I'll help with #2 time permitting: 1. Fix the naming so that it behaves as (well, to me anyways) one would expect - either match the name regardless of case or match based on case. The current implementation seems to conflate sql behaviour with dataframe behaviour. 2. Update the user guide to make it clear what the default behaviour is - and if that isn't the desired behaviour, how to change it. Even better would be to link each dataframe (and sql) with a page that outlines the function in more detail with examples, caveats and any corner case behaviour. 3. Update the api docs with information related to any flags that could impact behaviour/performance/etc and link to the examples provided in #2 (or datafusion-examples, whatever works) -- 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]
