From: Johannes Löthberg <johannes@kyriasis.com>
Date: Fri, 05 Jun 2015 17:27:36 GMT
Subject: Re: [PATCH v3] receive-pack: Create a HEAD ref for ref namespace
Message-ID: <20150605172736.GA30419@leeloo.kyriasis.com>
In-Reply-To: <xmqqh9qmrs2j.fsf@gitster.dls.corp.google.com>

On 05/06, Junio C Hamano wrote:
>Johannes Löthberg <johannes@kyriasis.com> writes:
>
>> +
>> +		echo "ref: refs/namespaces/new_namespace/refs/heads/master" >expect &&
>> +		test_cmp expect ../bare/refs/namespaces/new_namespace/HEAD  &&
>
>Use "symbolic-ref refs/namespaces/new_namespace/HEAD"; HEAD is not
>required to be expressed as a textual symref.
>

Gotcha.

>> +
>> +		(
>> +			printf "%s commit\t%s\n" $(git rev-parse master^) \
>> +			                         refs/heads/master &&
>> +			printf "%s commit\t%s\n" $(git rev-parse master) \
>> +			                         refs/namespaces/new_namespace/HEAD &&
>> +			printf "%s commit\t%s\n" $(git rev-parse master) \
>> +			                         refs/namespaces/new_namespace/refs/heads/master
>> +		) >expect &&
>
>Use of 'printf' is clever and I like it.  Have you considered
>letting it do the iteration as well?  I.e.
>
>	printf "%s commit\t%s\n" \
>        	one two \
>                three four \
>                five six \
>	>expect &&
>
>might be easier to read.
>

Didn't think about that actually. Will do.

>> +		git -C ../bare for-each-ref refs/ >actual &&
>> +		test_cmp expect actual
>> +	) &&
>> +	GIT_NAMESPACE=new_namespace git clone bare clone &&
>> +	(
>> +		cd clone &&
>> +		git show
>
>We can accept any random commit at HEAD as long as it exists at this
>point?  Don't we need to make sure that a ref whose tip is still "one"
>is not propagated to this new clone?
>

Oh crap, I just remembered that I forgot to address that part, sorry.

>> +	)
>> +'
>> +
>> +
>>  test_done

-- 
Sincerely,
  Johannes Löthberg
  PGP Key ID: 0x50FB9B273A9D0BB5
  https://theos.kyriasis.com/~kyrias/