Re: [Lynx-dev] [Patch] Lynx won't compile against zlib 1.2.5.1

[Thomas Dickey]

On Mon, 19 Sep 2011, Nikos Chantziaras wrote:

> On 09/19/2011 04:35 AM, Thomas Dickey wrote:
> > On Mon, 19 Sep 2011, Nikos Chantziaras wrote:
> >
> > > On 09/19/2011 04:13 AM, Thomas Dickey wrote:
> > > > On Sun, 18 Sep 2011, Keith Bowes wrote:
> > > >
> > > > > Je 2011-Sep-17 je 08:09, Thomas Dickey skribis:
> > > > > > ON/OFF are on the other hand unnecessary in lynx altogether - making
> > > > > > them
> > > > > > TRUE/FALSE would be simpler (similar to BOOL vs BOOLEAN, but different
> > > > > > since the latter are typed).
> > > > >
> > > > > Curious: a lot of programs use glib to get around this kind of thing
> > > > > (booleans, machine word-size differences, unified interfaces for things
> > > > > like regular expressions, etc.). Have you guys considered it?
> > > >
> > > > The BOOL vs BOOLEAN stuff is in an area where glib fails to help
> > > > (native Win32 ports).
> > >
> > > The best idea IMO is to check for "bool". If it's not there, typedef
> > > it. If
> >
> > in a native port, there's no autoconf - just a makefile.
> > Lynx has a few of those.
>
> I use autoconf in my native ports too.  I can cross-compile it from Linux or 
> compile it natively (on Windows with mingw/msys, on OS X and BeOS/Haiku 
> as-is.)

Lynx does that too.  Which native ports are you referring to?

(I've tried installing BeOS and Haiku a few times in a VM, but the result 
won't boot - my BeOS are on older machines which I'd like to obsolete, but 
cannot until I have a workable VM).

> Having a single source and build system base saves me a lot of headaches.

actually, dropping MinGW would simplify my life, particularly for its
Windows headers.   But just to be nice, I'll weave through its thicket
of inconsistent cut/paste header definitions to get something usable.

--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net