Re: pull requests for completion functions

[Oliver Kiddle <okiddle@xxxxxxxxxxx>]

Alex George wrote:
> No objections here. I think it would be especially useful for large
> patches. However, if someone were to use git{hub,lab} to submit a PR,
> should they send a message about it to the list? If so, should discussion
> about the patch take place on the list, or on git{hub,lab}? What should the
> mailing list subject be? Should the body simply contain a link to the PR?

The motivation is to make it easier for casual contributors to submit
completion function related patches. I was planning to keep interaction
related to pull/merge requests on git{hub,lab} and only move to the
list if questions related to the core parts of completion come up. When
discussing a proposed patch, it's often useful to quote parts of both
patch and commentary so I'm not sure how constructive it would be if the
patch is on gitlab and the explanatory remarks go to the list.

Sending completion function patches to the mailing list will also
continue to be perfectly acceptable. I'll probably continue to do things
that way myself. Perhaps we should lower the threshold of what gets
pushed unposted with completion functions to reduce noise on the list(?)

I'm not sure I see a problem with large patches on the mailing list but
until now, my experience with git{lab,hub} is relatively limited. So I'm
open to suggestions.

Oliver