Re: [PATCH v15 05/13] ref-filter: implement an `align` atom

From: Matthieu Moy <matthieu.moy@grenoble-inp.fr>
Date: Sep 2, 2015, 15:50 UTC
Message-ID: <vpqoahkom5a.fsf@anie.imag.fr>
In-Reply-To: <CAOLa=ZQxWO6+_N6TKgZ09HVGiNNDy29p6kU56Y2TX8Y+kYTcsA@mail.gmail.com>

Karthik Nayak <karthik.188@gmail.com> writes:

Show 22 quoted lines

> On Wed, Sep 2, 2015 at 2:15 PM, Matthieu Moy
>>> @@ -754,6 +816,42 @@ static void populate_value(struct ref_array_item *ref)
>>>                       else
>>>                               v->s = " ";
>>>                       continue;
>>> +             } else if (skip_prefix(name, "align", &valp)) {
>>> +                     struct align *align = &v->align;
>>> +                     struct strbuf **s;
>>> +
>>> +                     if (valp[0] != ':')
>>> +                             die(_("format: usage %%(align:<width>,<position>)"));
>>> +                     else
>>> +                             valp++;
>>
>> I agree with Junio that skip_prefix(name, "align:" ...) is simpler for
>> the same thing.
>
> Correct me if mistaken, but Junio wanted the first skip_prefix to
> check for "align:" rather than "align", but that would mean we don't have
> the error printing.
>
> Are you sure we want to skip that?

See my other message: you convinced me ;-).

-- 
Matthieu Moy
http://www-verimag.imag.fr/~moy/