Re: [RFC] branch: list branches by single remote

Junio C Hamano venit, vidit, dixit 24.08.2011 20:34:

Michael J Gruber [off-list ref] writes:

quoted

quoted

As suggested, I've just called it "--glob" for now.

Well, again, what's the point in replicating

http://permalink.gmane.org/gmane.comp.version-control.git/172228

and how is it different?
As I've mentioned, I've been in the middle of polishing that up.

It is not unusual for a similar itch to happen to different people
independently.

If this were something you reposted even a WIP re-polish within the past
two weeks, I would understand and even sympathise with your irritation,
but please don't expect everybody to dig back FOUR MONTHS worth of mail
backlog to find an topic that may or may not be abandoned by the original
author.  Perhaps we would need a weekly posting of topics people have

Well, I've mentioned it earlier in this thread, though without a link.
It's the time when several of us are off-line for a couple weeks. I've
also mentioned it in the thread about filtering by remote.

On a side note, the list here usually does require people to search
back, and much longer than this, on frequently requested issues. But how
does a newcomer discover "frequently"? Right, by searching back years,
not months.

posted, found to be not quite ready yet, and are still being polished and
not abandoned [*1*]?

Having said that, I still appreciate that you posted a link to the
previous topic:

  http://thread.gmane.org/gmane.comp.version-control.git/172226

so that the discussions in this thread to scratch the same "itch" can
benefit from the points raised in the previous thread that need to be
considered.

The old thread talks about renaming existing options and transition plans
to make the "listing" mode of "branch" and "tag" more similar, which may
be a good plan in the longer term.

I however can see that teaching "--glob" to both "branch" and "tag" (in
other words, "tag -l" would become a synonym for "tag --glob") an equally
good longer term plan.

I don't care about the names, but I'd hate to introduce more
inconsistencies. The implementation is a non-brainer, it's really
something only Apple could get a patent on. The other Michael and I came
up with basically the same patch because it follows "automatically". But
the ui is important.

So I'll take the opportunity and discuss this further in the cover
letter for the upcoming series.

Michael