On Thursday 26 October 2006 21:35, Joel E. Denny wrote:
> On Thu, 26 Oct 2006 henrik.sorensen balcab.ch
wrote:
> > On Thursday 26 October 2006 20:45, Joel E. Denny
wrote:
> > > On Thu, 26 Oct 2006 henrik.sorensenbalcab.ch
wrote:
> > > > On Thursday 26 October 2006 02:27, Joel
E. Denny wrote:
> > >
> > > That's nice. I'm wondering why the grammar
doesn't use $-1 and drop
> > > the mid-rule entirely, but I haven't test
this. I'm thinking there may
> > > always be a solution similar to these to
avoid USE for mid-rules.
> >
> > The grammar
> >
> > procoptionlist:
> > { $<pol>$ = newProcOptionList(); }
> > procoption
> > { $$ = $<pol>1; }
> >
> > | procoptionlist procoption {$$=$1;}
> > | procoptionlist ',' {$$=$1;} procoption
{$$=$1;}
> >
> > ;
> >
> > procoption:
> > MAIN {
if(setTristateAttribute(&($<pol>0)->main,1)<
0) YYERROR; }
> > ;
> >
> > so the procoption can be separated by either
blanks or comma.
>
> Sorry, I should've looked back at your original post.
Thanks for the
> clarification.
>
> Ok, so pretend you didn't actually need the $$=$1 in
the final semantic
> action. In this case, there would be no opportunity to
use $3
> legitimately, and so USE($3) would be necessary to
suppress the warning.
> In other words, scratch my above conjecture: there may
actually be cases
> where USE is necessary for mid-rule values.
>
> This brings me back to my earlier question. How often
do these issues
> arise?
I use this construct all the time, and the problem will
arise whenever an
option can be preceeded by anything than blanks.
> That is, do you believe this warning is too much
trouble to be
> worthwhile?
The warning would be ok, if it was only true.
> Surely it's useful in many cases, but could something
be
> changed about it to accommodate the troublesome cases?
Or is it ok as is?
the best thing would to detect that the $3 in rule
procoptionlist is
referenced as $0 in procoption
Henrik
_______________________________________________
help-bisongnu.org http
://lists.gnu.org/mailman/listinfo/help-bison
|