Re: Multitype support in bison

[Hans Aberg]

On 13 Oct 2007, at 01:10, Ilyes Gouta wrote:

> I'm new to Bison. I'm writing an interpreter for a lightweight C-like
> language using flex and bison. My language has to support the *two*
> types *int* and *float* used by the identifiers.

You might look into the "Grammar Summary" of C++/C. Standards can be  
gotten from ANSI at a low prices. You might also netsearch for  
"n1943.pdf" (free version of a C revision).

There is also info about a C++ Yacc grammar at:
  http://www.parashift.com/c++-faq-lite/compiler- 
dependencies.html#faq-38.11

> My question is, do I have to write separate rules for both types in my
> .y file, where each set of rules is dedicated for a given type, or do
> I have to handle the differentiation between those types in the
> grammatical actions associated with a unified set of rules for these
> two types?
>
> Here is an example of the grammar, I coded for bison:
>
> expr: CONSTINT                      { $<ival>$ = $<ival>1; }
>     | CONSTFLOAT                   { $<fval>$ = $<fval>1; }
>     | IDENTIFIER                   { $<ival>$ = getvalue($<id>1); }
>     | expr ADD expr                { $<ival>$ = $<ival>1 + $<ival>3; }
>     | expr SUB expr                { $<ival>$ = $<ival>1 - $<ival>3; }
> ;
>
> If I opt for the second case, I would, for example, modify the
> function getvalue to distinguish between the two types given the
> identifier's name, stored in id.

So these standards have only one set of rules for all types. You  
might hev to write more complicated actions, to do type-checking  
there, though.

> Any ideas guys?

And when you come to loops and jumps, you must build code closures in  
the actions, that can be executed afterwards.

  Hans Aberg