Re: problem compiling bison with gcc 3.0.5

[Martin Trautmann]

Hi,

please send only to the list. I subscribed.

Hans Aberg wrote:
> > + Why can't such a conflict be avoided with %prec precedences?
>
> It has to do with the precedence algorithm used; see for example
>   http://www.cs.vu.nl/~dick/PTAPG.html
> and the "Modern Compiler Design" book by the same authors.
I downloaded it, but I will need some time to read it.
My problem is, that I want to create two parallel expression parsers:
exp:
    simple_exp  %prec normal
  | complex_exp
;
simple_exp: simple_exp '+' simple_exp
  | '+' simple_exp %prec UMINUS
  | any_simple_token
;
complex_exp: complex_exp '+' complex_exp
  | '+' complex_exp %prec UMINUS
  | simple_exp %prec very_very_low
  | any_complex_token
;

The complex expression creates parse trees in the memory but it 
shouldn't be used for simple expressions like 1 + 1;
The generated parser works like I want it, but why isn't it possible to 
tell with %prec very_very_low that it shouldn't be used when there is an 
alternative rule with %prec normal.

> >   * passing some objects to the yyparse() function
>
> Check YYPARSE_PARAM, Bison manual sec 4.2.4.
I am using YYPARSE_PARAM at the moment, but its very nasty. Its only
possible to pass a void* pointer. This would be very easy to make the
type changeable.
The problem with the void* pointer is that the casts are quite long and
I don't like that in the actions. So I created lots of functions that
get the void* pointer and the other parameters, cast the pointer and 
call the right function of the casted object.


bye

Martin