[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RFC: enum instead of #define for tokens
From: |
Akim Demaille |
Subject: |
RFC: enum instead of #define for tokens |
Date: |
02 Apr 2002 16:34:35 +0200 |
User-agent: |
Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Common Lisp) |
Hi,
We are still trying to solve different defects of Bison, and designing
the C++ calling convention. Amongst the differences between C and
C++, given that we are not bound to backward compatibility, there is
no doubt that we will #define the tokens: they will be const int or
enums. But it is extremely tempting to move the C parser to enums
too.
Except for extremely bizarre scenarios which I'm able to imagine, but
fail to see the real interest, I see no difference between #define and
enums. One such scenario would be for instance to have parts of the
scanner compiled only when #ifdef SOME_TOKEN. Are there people using
such idioms?
More generally, does anybody see any argument against moving Bison to
enums instead of #define?
Does POSIX mandate something?
Thanks!
- RFC: enum instead of #define for tokens,
Akim Demaille <=
- Re: RFC: enum instead of #define for tokens, Miles Bader, 2002/04/02
- RE: enum instead of #define for tokens, Wayne Green, 2002/04/02
- Re: RFC: enum instead of #define for tokens, Paul Eggert, 2002/04/02
- Re: RFC: enum instead of #define for tokens, Akim Demaille, 2002/04/03
- Re: RFC: enum instead of #define for tokens, Paul Eggert, 2002/04/03
- Re: RFC: enum instead of #define for tokens, Akim Demaille, 2002/04/03
- Re: RFC: enum instead of #define for tokens, Hans Aberg, 2002/04/03
- Re: RFC: enum instead of #define for tokens, Akim Demaille, 2002/04/04
- Re: RFC: enum instead of #define for tokens, Hans Aberg, 2002/04/04