Re: GNATS needs some work (autoconf/59)

[Tim Van Holder]

> It would be shame if we release Autoconf with open bugs in the GNATS
> database.
> 
> I tried yesterday to analyze or close what I felt I can. But I don't feel
> competent e.g. in the DOS and Fortran problems.
> 
> So lets go to
> http://sources.redhat.com/cgi-bin/gnatsweb.pl?database=autoconf
> and fix whatever we can.
I'd like to volunteer to handle anything DOS/DJGPP-related if
possible.

As far as bug 59 (the pathsep thing) goes, I'm already looking
into a good way to do this. The way I see it, there are two
different issues that need to be handled separately.
On the one hand, there is the pathsep on the build system, which
should be used by autoconf, automake-generated makefiles and
other build scripts (such as autoconf's testsuite). Then there
is the pathsep on the target, as would be AC_DEFINE'd for use in
a program.

A big problem is deciding which pathsep is best in each case.
For the build pathsep, I'd originally thought to use a test for
a DOS/Win-ish environment (test -n "$COMSPEC$ComSpec"), but
this breaks for recent cygwin, as it seems to always use ':' as
pathsep (using its "/cygdrive/x/" system to replace "x:/"-style
DOS drivespecs).
So by my current thinking, I'd use uname as a basis for setting
an ac_pathsep variable, as this is fairly low-cost and could
therefore be done unconditionally at autoconf startup. It would
be necessary to be able to distinguish between (recent) cygwin
and other win32 environments (mingw32, for example) though.
Annoyingly, this might lead to duplicating some of the hair that
is now in config.guess. An AC_PATHSEP macro would be added that
would substitute a PATH_SEPARATOR variable, for use in automake
(particularly its texinfo rules) et al.

For the target pathsep (AC_TARGET_PATHSEP), I'd use
AC_CANONICAL_TARGET to decide (same as is done for AC_CYGWIN etc
now), then AC_DEFINE a PATH_SEPARATOR macro (I'm not sure whether
or not this would clash with the AC_SUBST'ed PATH_SEPARATOR though).

Mark, if you have any further thoughts on this, please mail
them to the autoconf list for review (starting with the macro you
have). Thanks.