On 31/05/2020 19:19, Kyle Meyer wrote:
You've been sending a diff,
presumably from the point you branched off of to the tip of your branch.
In that case, you're already presenting each iteration you've sent as
one change; it just lacks a commit message.
right, that's indeed what I did, and this is also what I thought, so no
need to rebase, squash or whatever, as long as I make sure that the diff
I'm sending you is about this single issue, and let's agree on the
commit message, because after all I'm adding a function to a software I
don't really know.
I hope to send an updated patch soon, that will also include the docs.
I have no strong opinion on workflows, just trying to understand the one
used here.
btw: if I had write permissions to the repositories, I would be adding
test cases, and reviewing the docstrings, some of which I find
misleading. your remark on setf/setq could also be addressed in the
code. and some of the code ought to be refactored, as to allow for unit
tests.
ciao, MF
