Re: [PATCH v2 15/16] fsck: reduce word legos to help i18n

[Duy Nguyen <pclouds@gmail.com>]

On Tue, Nov 6, 2018 at 4:41 AM Junio C Hamano <gitster@pobox.com> wrote:
> >  static int fsck_error_func(struct fsck_options *o,
> >       struct object *obj, int type, const char *message)
> >  {
> > -     objreport(obj, (type == FSCK_WARN) ? "warning" : "error", message);
> > -     return (type == FSCK_WARN) ? 0 : 1;
> > +     if (type == FSCK_WARN) {
> > +             fprintf_ln(stderr, "warning in %s %s: %s",
> > +                        printable_type(obj), describe_object(obj), message);
> > +             return 0;
> > +     }
> > +
> > +     fprintf_ln(stderr, "error in %s %s: %s",
> > +                printable_type(obj), describe_object(obj), message);
> > +     return 1;
>
> Make it look more symmetrical like the original, perhaps by
>
>         if (type == FSCK_WARN) {
>                 ...
>                 return 0;
>         } else { /* FSCK_ERROR */
>                 ...
>                 return 1;
>         }
>
> Actually, wouldn't it be clearer to see what is going on, if we did
> it like this instead?
>
>         const char *fmt = (type == FSCK_WARN)
>                 ? N_("warning in %s %s: %s")
>                 : N_("error in %s %s: %s");
>         fprintf_ln(stderr, _(fmt),
>                    printable_type(obj), describe_object(obj), message);
>         return (type == FSCK_WARN) ? 0 : 1;
>
> It would show that in either case we show these three things in the
> message.  I dunno.

Specifying "type == FSCK_WARN" twice triggers me (what if we add a
third fsck type?) so I just turn this to a switch/case block instead
(and get to know the third fsck type FSCK_IGNORE).
-- 
Duy