Re: [PATCH v2] receive-pack: optionally deny case clone refs

From: David Turner <dturner@twopensource.com>
Date: Jun 4, 2014, 17:24 UTC
Message-ID: <1401902684.18134.144.camel@stross>
In-Reply-To: <538EB748.3050300@viscovery.net>

On Wed, 2014-06-04 at 08:06 +0200, Johannes Sixt wrote:

Show 7 quoted lines

> > +receive.denyCaseCloneBranches::
> > +	If set to true, git-receive-pack will deny a ref update that creates
> > +	a ref which is the same but for case as an existing ref.  This is
> > +	useful when clients are on a case-insensitive filesystem, which
> > +	will cause errors when given refs which differ only in case.
> 
> Shouldn't this better be named 'receive.denyCaseCloneRefs'?

Yes.  I'll fix that.

> How about 'denyCaseInsensitiveRefs', 'denyIgnoreCaseRefs'?

I don't love these, because it's not the case-insensitivity that's being denied but the duplication.

> Is this entry really so important that it must be the first in the list of
> receive.deny* list, which is not alphabetically sorted?

I think the list should be sorted alphabetically, so that we don't have to worry about what is most important.

<snip issues that I'll fix when I reroll>

Show 5 quoted lines

> BTW, on a case-insensitive file system, is there not a chance that 'git
> rev-parse CaseClone' succeeds even though the ref is stored in
> victim/.git/refs/heads/caseclone? Perhaps you should inspect the output of
> 'git for-each-ref' for the expected result? (Mental note: At least a
> case-preserving file system is required to run the test.)

I'll look into that and fix if necessary.  Thanks.

<snip more issues that I'll fix when I reroll>