[PATCH STABLE V2] update: allow update to existing branches with invalid names (issue3710)

[Kevin Bullock]

On 28 Nov 2012, at 2:22 PM, Tim Henigan wrote:

> On Wed, Nov 28, 2012 at 1:45 PM, Pierre-Yves David
> <pierre-yves.david at logilab.fr> wrote:
>> I forget to add that the patch looks good otherwise. Except, I've one
>> more question about this patch. What about others way to create branch
>> other than "hg branch". Do we want to deny them? In particular, shall we
>> accept to import patch that create new invalid branch name.
> 
> As Matt stated in his notes about this issue [1], we should be strict
> in what we accept from the user, but generous with stuff already
> created.
> 
> In my opinion, a patch that includes an invalid branch name should be
> accepted.  It may be a commit that was already created, tested
> (hopefully), and distributed by the author.

Anyone else have arguments specifically around this issue? I'm ambivalent, but tilting slightly towards agreeing.

If others agree, then I'll add a test that explicitly covers this.

pacem in terris / мир / शान्ति / ‎‫سَلاَم‬ / 平和
Kevin R. Bullock

> [1]: http://bz.selenic.com/show_bug.cgi?id=3710#c6