Re: [PATCH/RFC 3/6] http-backend: handle refspec argument

From: David Turner <dturner@twopensource.com>
Date: Apr 19, 2016, 18:57 UTC
Message-ID: <1461092277.5540.108.camel@twopensource.com>
In-Reply-To: <CAPig+cRSE-BCPdrSbrCYmTcT6EsabKnekr2GEAbmBsc5=jxnHQ@mail.gmail.com>

On Sat, 2016-04-16 at 21:51 -0400, Eric Sunshine wrote:

Show 16 quoted lines

> On Fri, Apr 15, 2016 at 3:19 PM, David Turner <
> dturner@twopensource.com> wrote:
> > +               if (refspec) {
> > +                       struct strbuf interesting_refs =
> > STRBUF_INIT;
> > +                       strbuf_addstr(&interesting_refs, "-
> > -interesting-refs=");
> > +                       strbuf_addstr(&interesting_refs, refspec);
> > +                       argv_array_push(&argv,
> > interesting_refs.buf);
> > +                       strbuf_release(&interesting_refs);
> > +               }
> 
>     if (refspec)
>         argv_array_pushf(&interesting_refs,
>             "--interesting-refs=%s", refspec);

Will fix, thanks.

Show 10 quoted lines

> >                 argv_array_push(&argv, ".");
> >                 run_service(argv.argv, 0);
> >                 argv_array_clear(&argv);
> > @@ -841,6 +905,19 @@ int main(int argc, char **argv)
> > +               if (starts_with(arg, "--interesting-refs=")) {
> > ...
> > +                       continue;
> > +               }
> 
> Is this leaking the string list?

Yes, intentionally. interesting_refspec is a global that we look at later.