Re: too many warnings from Bison CVS for Pike

[Hans Aberg]

On 14 Feb 2006, at 22:24, Joel E. Denny wrote:

> On Thu, 9 Feb 2006 address@hidden wrote:
>
> > Dynamic C-parserstack. This is what is Bison implements if
> > __cplusplus is not defined. However, this should work under C++ as
> > well for POD types. By the segment:
> >
> > #if (! defined (yyoverflow) \
> >       && (! defined (__cplusplus) \
> >          || (defined (YYLTYPE_IS_TRIVIAL) && YYLTYPE_IS_TRIVIAL \
> >               && defined (YYSTYPE_IS_TRIVIAL) &&  
> > YYSTYPE_IS_TRIVIAL)))
> >
> > /* A type that is properly aligned for any stack member.  */
> > union yyalloc
> > {
> >    short int yyss;
> >    YYSTYPE yyvs;
> >      YYLTYPE yyls;
> > };
> > ...
> >
> > you already have a test against non-POD types, as these cannot be
> > used in a union. So if one would want to use this dynamic stack with
> > a POD type, just remove the (! defined (__cplusplus) above.
>
> I'm getting a little lost in this thread, so forgive me if I'm
> repeating....  Why is it necessary to remove the ! defined  
> (__cplusplus)?
> If %union is defined, then YYSTYPE_IS_TRIVIAL.  If also
> YYLTYPE_IS_TRIVIAL, then union yyalloc and YYSTACK_RELOCATE() will be
> defined.  In this case, defined(__cplusplus) is irrelevant.
>
> Am I missing something?

If you say that if %union is invoked, and that defines  
YYSTYPE_IS_TRIVIAL, which in its turn causes the dynamic C- 
parserstack to be used, then I missed that. :-)

> > But one might supply a macro, say CPP_SEMANTIC_POD, which signals
> > that the C++ semantic type is a POD, and that the dynamic C-
> > parserstack should be used.
>
> %union seems to be enough.

If %union using a C++ compiler automatically invokes the dynamic C- 
parserstack, then there would only be the need for such an option for  
user that want to use a POD type without %union.

But is not the combination of YYLTYPE_IS_TRIVIAL and  
YYSTYPE_IS_TRIVIAL the same as the macro CPP_SEMANTIC_POD I  
suggested? Why do you need two macros? Is not enough with one macro  
YYTYPES_ARE_POD, which, when defined, invoked the dynamic C-parser?  
Then %union defines YYTYPES_ARE_POD, triggering the dynamic stack.  
And if somebody defines YYTYPES_ARE_POD, and tries to put in a non- 
POD types in at least one of the semantic or location types, a  
conforming C++ compiler will generate an error message when it  
encounters the definition of union yyalloc above.

  Hans Aberg