maartenbreddels commented on pull request #7357:
URL: https://github.com/apache/arrow/pull/7357#issuecomment-642451423
Few questions about the preferred workflows (didn't see this in the contribs
guidelines):
* should Jira issues and PR's be 1 tot 1, or can I open multiple PR's
referencing the same Jira issue
* 1 commit per PR, or multiple commits per PR if related, but each commit
one idea/topic/functionlity?
I have to dive a bit into how the kernels work, but with the current
implementation of ascii_lower and upper I see performance issues (we should be
able to write into a pre-allocated buffer, since we know the byte length). I'm
happy first getting some basic functionality in, and optimize later, it will
surface the issues around the framework a bit more I think.
So would you prefer to get in some basic string functions first, and
optimize later? In that case, we can rename this PR and merge it, I'll open a
few more PRs, and once a few functions are in, we have a baseline for
benchmarking and can optimize.
----------------------------------------------------------------
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:
[email protected]
