RE: Compile Problems with Harold's Server...

[Suhaib Siddiqi <ssiddiqi at inspirepharm dot com>]

> Actually I think the patch Harold submitted was because the w32api was
> wrong.
> 

If so, then w32api should in Cygwin should be patched without #ifdef for
Xserver???

Suhaib


> Rob
> 
> ----- Original Message -----
> From: "Suhaib Siddiqi" <ssiddiqi@inspirepharm.com>
> To: "'Alexander Gottwald'" <gottwald@prudsys.com>
> Cc: <cygwin-xfree@cygwin.com>
> Sent: Wednesday, April 11, 2001 11:18 PM
> Subject: RE: Compile Problems with Harold's Server...
> 
> 
> >
> >
> > > I think there were two possibilies.
> > >
> > > a) leave the w32api includes unchanged and modify whole xc-tree
> > > b) fix the occurences in w32api and leave the x sources untouched
> > >
> > > since the first would have been a lot of work, I think the second
> was
> > > the best.
> >
> > Nope, first is the best.  You do not need to modify whole 
> 250 MB code
> > of xc-tree it is ONLY Harold code in xc/programs/Xserver/hw/xwin.
> Therefore
> > that needs to fixed for namespace issues, as John reported.
> >
> > I do not think it is a good idea that folks at Cygwin start patching
> the
> > headers
> > for every software out there.
> >
> > >
> > > Of course these changes should only be made when 
> compiling for X (or
> > > even just the XServer) and not for normal development.
> > >
> > > Maybe the inclusion of the affected windows headers can be
> > > stopped by adding
> > > some #define and #undef, but this is for the future when the
> > > XServer is
> > > running correctly and no other work has to be done.
> >
> >
> > The code in hw/xwin needs to fixed.  That is the best way to go.
> >
> > Suhaib
> >
> > >
> > > Bye
> > >     ago
> > >
> > > --
> > > Alexander Gottwald
> > > gottwald@prudsys.com
> > > www.gotti.org
> > >
> >
>