Re: What's cooking in git.git (May 2013, #04; Wed, 15)

From: Junio C Hamano <gitster@pobox.com>
Date: May 21, 2013, 15:27 UTC
Message-ID: <7vy5b8qro0.fsf@alter.siamese.dyndns.org>
In-Reply-To: <87a9nobyae.fsf@linux-k42r.v.cablecom.net>

Thomas Rast <trast@inf.ethz.ch> writes:

Show 8 quoted lines

> Junio C Hamano <gitster@pobox.com> writes:
>
>> In addition to your topic, it may be a good idea to notice this at
>> the Porcelain level (e.g. "checkout -b" and "branch", but not at the
>> "update-ref" level) and warn or even die if a Porcelain tries to
>> create a branch with such a name.
>
> But if we're going there and deprecating some ways of naming refs,

This is not about deprecating but is discouraging a confusing use at the Porcelain level (i.e. enforcing one policy decision that may not apply to those who wrote their own workflow using plumbing commands, knowing and following the "full 40-hex is taken as an object name, not a dwimmed ref" rule---to them there is no ambiguities).

I think I see where you are going and I do not necessarily disagree with that, but it is a separate topic.

> please also disallow some other funny things in the same go.  Michael
> suggested this earlier in some thread: the fewer ways we have of legally
> spelling refnames, the more syntax is available for revision syntax.