[Xlock-develop] xlockmore-5.30ALPHA05 released

[Jouk Jansen]

bagleyd at gwyn.tux.org wrote on 9-FEB-2010 17:40:46.28

>> David it seems that I'm not allowed to send to your E-mail address!!!
>Sometimes my mail stops working.  I dont know why.... I think if you
>send it with the "gwyn." part it fails more often (bagleyd at gwyn.tux.org as
>opposed to bagleyd at tux.org.  It usually works.
>Unfortunately if you hit reply on my mails it puts the gwyn part automatically.
>Jouk, I am not sure if this is related to the problem you had/have.
My mail-client think the mails come from bagleyd at gwyn.tux.org. Since it told
me that it was "grey-listing and come back later" I tried again, but it
failed again. I CC this mail to bagleyd at tux.org.


>> I will test on OpenVMS later this week.
>Ok worried here because I was messing around with Intrinsics.h.... its
>not used on the Unix side with the OpenGL.  But I think VMS uses it... not
>sure why.
The beta version just works fine. I have no idea if or why  Intrinsics.h is
used on VMS. Normally the X-windowing stuff is not that much different on
VMS and Unix apart form the "old" version on VMS.


>> >I updated all the gl modes to have the WIN32 defines.  I think I got
>> >all the "low hanging fruit", though some of these others may not
>> >be that hard to get to work.
>> >
>> >GL modes added except:
>> >  fire.c: xpm dependencies, works but no ground and tree
>> >  glplanet.c: xbm stuff
> >  invert.c: problems linking C++
>> 
>> What is the linking problem? Maybe I know a solution, since my own programs
>> are a mixture of F90, C and C++ (main is in F90) and they link under cygwin
>> with and without the -mno-cygwin option. 
>Yeah I dunno what was going on here.  solitaire.cc fails to link as well
>and that is no problem without invert.
>
>Here's how to reproduce in BETA version:
>change xlock/mode.h:  make MODE_invert be defined
>change modes/glx/Makefile.win32: add invert.c to C source and
>  uncomment the CC_SRC
>then to make -f Makefile.win32 in main dir.
>
>g++-3 -DWIN32 -O2 -mno-cygwin -mwindows -Wall -o xlock95.scr xlock/*.o modes/*.o modes/glx/*.o win32/*.o win32/*.res -L/usr/lib/w32api -lcrypt -lopengl32 -lglu32
>modes/glx/i_figureeight.o:i_figureeight.cc:(.bss+0x0): multiple definition of `_randommode'
>modes/solitaire.o:solitaire.cc:(.bss+0x0): first defined here
>modes/glx/i_linkage.o:i_linkage.cc:(.bss+0xc): multiple definition of `_randommode'
>modes/solitaire.o:solitaire.cc:(.bss+0x0): first defined here
>modes/glx/i_sphere.o:i_sphere.cc:(.bss+0x0): multiple definition of `_randommode'
>modes/solitaire.o:solitaire.cc:(.bss+0x0): first defined here
[snip]
>collect2: ld returned 1 exit status
>make: *** [xlock95.scr] Error 1

I see that the definition of randmode in win32/xapi.h is not "extern".
However I do not see where xapi.h is used. Not defining this "extern" and
using the .h file in more than one .c file would give similar messages on my
OpenVMS box. I noticed that different compilers handle this problem
differently. So maybe even gcc and g++.

If this failed randommode may be a "reserved" word. So I would change all
occurencies in my source by something else.

I have no time to test this right now.



               Jouk


Pecunia olet!

>------------------------------------------------------------------------------<

  Jouk Jansen
		 
  joukj at hrem.nano.tudelft.nl

  Technische Universiteit Delft        tttttttttt  uu     uu  ddddddd
  Kavli Institute of Nanoscience       tttttttttt  uu     uu  dd    dd
  Nationaal centrum voor HREM              tt      uu     uu  dd     dd
  Lorentzweg 1                             tt      uu     uu  dd     dd
  2628 CJ Delft                            tt      uu     uu  dd     dd
  Nederland                                tt      uu     uu  dd    dd
  tel. 31-15-2782272                       tt       uuuuuuu   ddddddd

>------------------------------------------------------------------------------<