> The latest changes in CVS lead to the following errors when
> compiling on a Mac: [...]
Thanks, I hope this is fixed now in the CVS. Please test.
Werner
___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff
The latest changes in CVS lead to the following errors when compiling
on a Mac:
g++ -I. -I. -I/Users/Robert/groff/src/include
-I/Users/Robert/groff/src/include -DHAVE_CONFIG_H -g -O2 -c color.cpp
In file included from color.cpp:30:
/usr/include/unistd.h:183: error: declaration of `int getopt
> But `false' is part of POSIX. That should be enough.
> You cannot use the original Bourne style for much; it does not
> even have shell functions.
Depends on your requirements, and how portable you want to keep
your script. You *cannot* blindly rely on POSIX, as there are
still many platforms
Keith MARSHALL <[EMAIL PROTECTED]> schrieb am 21.06.05 12:44:20:
>
> > I'll simply use a modified version of Keith's suggestion:
> >
> > foo.ps: foo.ms
> >groff --some-options foo.ms >$@ \
> >|| ($(RM) $@; false)
>
> Indeed, you can use `false', instead of `exit
> I took a brief, incomplete look. In all places I've seen, the caller either
> knows the string length, or takes case to append a NUL before calling
> contents().
Thanks!
Werner
___
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/
> I'll simply use a modified version of Keith's suggestion:
>
> foo.ps: foo.ms
>groff --some-options foo.ms >$@ \
>|| ($(RM) $@; false)
Indeed, you can use `false', instead of `exit 1', as I suggested.
But note that `false' is *not* a shell builtin in standard Bourn
On Mon Jun 20 21:47:48 2005, [EMAIL PROTECTED] wrote:
>
> > When soelim wrote error messages about being unable to open files,
> > the reported file name tended to have garbage characters appended to
> > it. Please apply.
>
> Done, thanks.
>
> > ('const char * string::contents() const' is devio
On Tue Jun 21 00:32:15 2005, [EMAIL PROTECTED] wrote:
>
> > Okay, I'm seriously confused here, [...]
>
> I hope that my other mail helps you to understand the problem.
Yes, it did. Thanks.
...
> > Anyway, that isn't as important as I thought. What I /know/ to be
> > different in CVS groff compa
On Tue Jun 21 08:09:46 2005, [EMAIL PROTECTED] wrote:
>
> > > Looking into the `tmac/s' file from Solaris I see that the
> > > start-up code *always* sets PO. Since groff allows the setting of
> > > `PO' on the command line, this is not an option for groff.
> >
> > So this would lead to the rule
