Re: AC_PROG_CC problems on Windows with MKS Shell

[Lars J. Aas]

On Fri, Jan 19, 2001 at 03:15:02PM +0100, Akim Demaille wrote:
: Lars> I think I would prefer it if we used verbose names instead of
: Lars> _X, _E, etc. then.  Maybe also drop "TEST" and mark the macro
: Lars> name as a predicate in some other way.  I'd also consider
: Lars> including FILE / DIR in the macro name...
: 
: AS_EXECUTABLE_IF
: 
: That's too heavy, IMHO, and makes it useless hairy to write composed
: tests where we would have naturally used && and || etc.
: 
: But maybe I get your idea wrong.  Could you make it explicit?

No, you got the idea, except *_IF() isn't good since it's a predicate
that "evaluates" to true or false, not a macro that has IF and IF-NOT
arguments to run depending on a test...  

In lisp/scheme, you name predicates e.g. "is-integer?" or just "integer?".
The C convention is for Guile to use names like SCM_INTEGER_P(value)
where _P means it is a predicate.  I don't think _P is something we
should use, and using "?" means changing the m4 word syntax, so it's
probably not an option either.  We're really left with using IS_* to
mark predicates, so something like AS_IS_EXECUTABLE(file) would be
my suggestion.

: Mine was AS_TEST_X(FILE) == test (-x|-f) $1.  I think keeping TEST is
: not too bad, that's what we do with AS_DIRNAME, AS_MKDIR etc.   ?

Sure, but test is in a slightly different category.  Since we're working
with non-portable flags here, is it always the case that the flag is
either supported or does not exist, or can a flag have different
meanings for different shells?

  Lars J
-- 
Innovation is one percent inspiration and ninetynine percent perspiration,
and in my case; twice that...  -- Norville Barnes, `The Hudsucker Proxy'