Re: [PATCH v13 00/12] port tag.c to use ref-filter APIs

> On Tue, Aug 25, 2015 at 4:05 AM, Junio C Hamano <gitster@pobox.com> wrote:
>> Junio C Hamano <gitster@pobox.com> writes:
>>
>>> Karthik Nayak <karthik.188@gmail.com> writes:
>>>
>>>> On Mon, Aug 24, 2015 at 10:57 PM, Junio C Hamano <gitster@pobox.com> wrote:
>>>>> Karthik Nayak <karthik.188@gmail.com> writes:
>>>>> ...
>>>>>> +     performed. If used with '--quote' everything in between %(align:..)
>>>>>> +     and %(end) is quoted.
>>>> ...
>>>> I might have misunderstood you, But based on the discussion held here
>>>> (thread.gmane.org/gmane.comp.version-control.git/276140)
>>>> I thought we wanted everything inside the %(align) .... %(end) atoms
>>>> to be quoted.
>>>
>>> Perhaps I misunderstood your intention in the doc.
>>>
>>> I took "everything in between %(align:...) and %(end) is quoted" to
>>> mean that
>>>
>>>       %(if:empty)%(align)%(end)%(then)Empty%(else)Not Empty%(end)
>>>
>>> can never satisfy %(if:empty), because %(align)%(end) would expand
>>> to a string that has two single-quotes, that is not an empty string.
>>>
>>> If that is not what would happen in the "branch --list" enhancement,
>>> then the proposed behaviour is good, but the above documentation would
>>> need to be updated when it happens, I think.  It at least is misleading.
>>
>> OK, now I checked the code, and I _think_ the recursive logic is
>> doing the right thing (modulo minor nits on comment-vs-code
>> discrepancy and code structure I sent separately).
>>
>
> For the current code %(if:empty)%(align)%(end)%(then)Empty%(else)Not Empty%(end)
> would print non-empty, I guess the documentation holds in that case.
> Not sure if we require it to print non-empty.