Re: ANSI CL support

[Pascal J. B. <pj...@in...>]

Raymond Toy <toy...@gm...> writes:

>>>>>> "Pascal" == Pascal J Bourguignon <pj...@in...> writes:
>
>     Pascal> Raymond Toy <toy...@gm...> writes:
>     >> A friend pointed this out to me:
>     >> 
>     >> (apropos "-ANSI") ->
>     >> *COERCE-FIXNUM-CHAR-ANSI*                  variable
>     >> *FLOATING-POINT-CONTAGION-ANSI*            variable
>     >> *FLOATING-POINT-RATIONAL-CONTAGION-ANSI*    variable
>     >> *LOOP-ANSI*                                variable
>     >> *MERGE-PATHNAMES-ANSI*                     variable
>     >> *PARSE-NAMESTRING-ANSI*                    variable
>     >> *PHASE-ANSI*                               variable
>     >> *PRINT-EMPTY-ARRAYS-ANSI*                  variable
>     >> *PRINT-PATHNAMES-ANSI*                     variable
>     >> *PRINT-SPACE-CHAR-ANSI*                    variable
>     >> *PRINT-UNREADABLE-ANSI*                    variable
>     >> *SEQUENCE-COUNT-ANSI*                      variable
>     >> 
>     >> The ANSI CL spec is now almost 20 years old and hasn't changed since
>     >> then. I didn't check to see what all of variables actually do, but
>     >> isn't it time to make these default to T? Or at least some of them?
>
>     Pascal> Always run clisp with -ansi (for the case you also give -norc),
>     Pascal> and put (setf custom:*ansi* t) in ~/.clisprc.lisp
>
> Yes, my friend mentioned that to me as well. But that doesn't explain
> why this is not the default. Does anyone really want a Common Lisp
> implementation that isn't?

I'd agree, IMO the default should be ANSI, if only for the unsuspecting
newbies and other occasional users.


-- 
__Pascal Bourguignon__
http://www.informatimago.com/