Re: push parser

[Bob Rossi]

On Wed, Sep 27, 2006 at 04:30:43PM -0700, Paul Eggert wrote:
> Bob Rossi <address@hidden> writes:
> 
> > There are no need for global variables, you are correct. The patch I
> > originally made simply did not do away with the 4 global variables that
> > the user traditionally used. I was a afraid this would break backwards
> > compatibility for many users.
> 
> OK, but we shouldn't need to worry about doing away with global
> variables in push parsers, since they're supposed to be pure.  We need
> to keep globals only for traditional, impure parsers.

OK, agreed. There are 4 globals of interest. 
  int yychar;
  YYSTYPE yylval;
  int yynerrs;
  YYLTYPE yylloc;
I know what to do with yylval and yylloc. That's because we have the
user declare them and pass them into the parser. I also know what to
do with yychar, because the user also passes that data into the parser.
What should I do with yynerrs?

> >>  {
> >>    struct yypvars *ctx = yypvarsinit ();
> >>    int status;
> >>    do {
> >>      YYTOKTYPE token_info;
> >>      status = yypushparse (ctx, my_lex (&token_info), &token_info);
> >>    } while (status == YYPUSH_MORE);
> >>    free (ctx);
> >>    return status;
> >>  }
> >
> > Be careful with the above examples, function parameters can be 
> > evaluated in either direction, and in the case above, if the compiler
> > evaluates right to left, you will get undesired results.
> 
> No, there's no problem here, since the example passes token_info's
> address to yypushparse, not its value.

Oops. I thought I had one up on you :).

Thanks,
Bob Rossi