On Thu, 2016-05-12 at 09:06:45 +0200, Julia Lawall wrote:
> > > @removal@
> > > identifier F, I;
> > > type T;
> > > @@
> > >  F(...,
> > > -  const
> > >    T I,
> > >   ...
> > >  );
> 
> Oops, now I see why this doesn't parse.  When you make a rule for a
> prototype, you need to specify the return type (which could be just T1).
> Otherwise, the parser can't tell the difference between this and a
> function call.

You're right that does it. Looks to me then that the grammar
documentation should not indicate the fn_ctype as optional for function
prototypes (http://coccinelle.lip6.fr/docs/main_grammar006.html)?

        Thanks,
        Sören
_______________________________________________
Cocci mailing list
[email protected]
https://systeme.lip6.fr/mailman/listinfo/cocci

Reply via email to