Show 14 quoted lines
> Steffen Prohaska <prohaska@zib.de> writes:
>
>> What's left is a new switch "--current". Less code, easy
>> to explain.
>
> But won't that force the "current" people always type that from
> the command line, as your previous point was that your earlier
> patch to say "remote.$there.push = HEAD" does not work that way?
> If that configuration works as expected, then I'd 100% agree
> that we would not need push.defaultRefs. Either you do not have
> "push" at all if your preference is --matching, or you do have
> "push = HEAD" if your preference is --current. But if it
> doesn't (which was what I gathered from your earlier response),
> having a configuration would help them, wouldn't it?
Brief recap, to check if I understand things correctly:
1. If you use "matching" more often, then it should be enough to provide
remote.<remotename>.push with refspec or wildcard refspec. "git push"
would push matching. If one wants to push only current branch, one
would use "git push --current" or "git push <remotename> HEAD".
Question: what to do if there is no remote.<remotename>.push? Assume
1-1 matching?
2. If you use "current" more often, then it should be anough (after
correcting git; although it was written that it is quite a bit of work)
to provide "remote.<remotename>.push = HEAD", or
"push.defaultRefs = current" if one wants to set this up for all
remotes, or perhaps "remote.*.push = HEAD". "git push" would push
current. If one wants to push matching, one would use "git push
--matching"... although for matching one needs remote configured...
Jakub Narebski <jnareb@gmail.com>