Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Chris McGuire on 1/12/2007 7:55 PM: > That was it, worked fine! > Hooray - the gnulib wchar module has seen its first happy customer. I'll check in the remaining m4 patches shortly; I've been camping on the ss_sp vs. ss_base patch for a

Re: Building m4 on BSDI 4.0.1

Bruno Haible clisp.org> writes: > > 2007-01-12 Eric Blake byu.net> > > > > Provide a robust . Further simplifications are now > > possible in other modules, but not included here. > > Thanks. Looks fine, except 2 nits: OK, checked in with those nits fixed, along with a typo fix in

Re: Building m4 on BSDI 4.0.1

Eric Blake <[EMAIL PROTECTED]> writes: > Here's my first cut at it. It currently assumes that absolute > exists. I'm not sure whether it is worth trying to check for > HAVE_WCHAR_H, or to leave that until we actually get a report of a system > that needs it. Yes, I'd leave it alone until then,

Re: Building m4 on BSDI 4.0.1

Eric Blake wrote: > 2007-01-12 Eric Blake <[EMAIL PROTECTED]> > > Provide a robust . Further simplifications are now > possible in other modules, but not included here. > * modules/wchar: New module. > * m4/wchar.m4: New file. > * lib/wchar_.h: Likewise. > *

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Paul Eggert on 1/11/2007 10:39 AM: > Eric Blake <[EMAIL PROTECTED]> writes: > >> At this point, I really think it is easier to generate a replacement >> that takes care of the intricacies, > > Yes, I'd like a wchar module that provides

Re: Building m4 on BSDI 4.0.1

Eric Blake <[EMAIL PROTECTED]> writes: > At this point, I really think it is easier to generate a replacement > that takes care of the intricacies, Yes, I'd like a wchar module that provides wchar.h, so that we can remove HAVE_WCHAR_H from most of the code. > Also, there are 8 files in lib/* th

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Eric Blake on 1/10/2007 11:24 AM: > > Still not enough; Chris reported another failure offlist. wctype.m4's > AC_SUBST > (HAVE_WINT_T) was resulting in an empty string, even though wint_t.m4's > AC_DEFINE(HAVE_WINT_T, 1, ...) was now g

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Paul Eggert on 1/10/2007 3:27 PM: > > But the rest of wctype_.h uses the @...@ substitution method; this is > the tradition for .h files that might get installed into an environment > that doesn't have config.h. > > I installed this furt

Re: Building m4 on BSDI 4.0.1

Eric Blake <[EMAIL PROTECTED]> writes: > The obvious solution (I hope) - only use one of the two substitution methods. But the rest of wctype_.h uses the @...@ substitution method; this is the tradition for .h files that might get installed into an environment that doesn't have config.h. I insta

Re: Building m4 on BSDI 4.0.1

Eric Blake byu.net> writes: > > Ahh - here we go. My earlier patch fails, because it was guarded by > HAVE_WINT_T, but your configure output showed that we guessed that bit > wrong. By the way, the config.log file produced by configure is generally > more useful than the captured output of con

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Chris McGuire on 1/9/2007 9:26 PM: > > It looks like the following gnulib patch is needed. Chris, can you > > confirm one last thing? Check that the following compiles for you (gcc -c > > file.c), but that when you comment out the line w

Re: Building m4 on BSDI 4.0.1

Booo to Paul. ;) Eric, I'm working on your tests now. Thanks guys, Chris At 12:37 PM 1/9/2007, Paul Eggert wrote: >Eric Blake <[EMAIL PROTECTED]> writes: > >> maybe it >> is worth factoring out wchar.m4 into its own file and using that from all >> other modules that use wchar_t or . > >Sounds l

Re: Building m4 on BSDI 4.0.1

Eric Blake <[EMAIL PROTECTED]> writes: > maybe it > is worth factoring out wchar.m4 into its own file and using that from all > other modules that use wchar_t or . Sounds like a lot of work for an operating system whose original supplier stopped support for many years ago.

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Paul Eggert on 1/9/2007 12:23 AM: > Eric Blake <[EMAIL PROTECTED]> writes: > >> BSD/OS 4.1 has a bug: and must be included before >> - . */ >> + . >> + BSDI 4.0.1 has a bug: must be included before . */ >> +# include > >

Re: Building m4 on BSDI 4.0.1

Eric Blake <[EMAIL PROTECTED]> writes: > BSD/OS 4.1 has a bug: and must be included before > - . */ > + . > + BSDI 4.0.1 has a bug: must be included before . */ > +# include That patch looks fine to me, except that the comment should say "BSD/OS" rather than "BSDI". The operating

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Chris McGuire on 1/8/2007 12:59 PM: >> The second one is for you - can you locate a system header that >> defines wchar_t on your platform? Something like the following will be a >> good start, >> >> $ grep -E '(typedef|define).*wchar_t'

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Chris McGuire on 1/3/2007 7:46 AM: >>> source='mbchar.c' object='mbchar.o' libtool=no \ >>> DEPDIR=.deps depmode=gcc /bin/bash ../depcomp \ >>> gcc -I. -g -O2 -c mbchar.c >>> In file included from mbchar.h:157, >>>

Re: Building m4 on BSDI 4.0.1

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Chris, and happy new year, According to Chris McGuire on 1/3/2007 7:41 AM: >> I guess the following hack forces configure to ignore wchar.h: >> >> ./configure ac_cv_header_wchar_h=no > I guess you may have found something there. The fact that thi