My last reply was incomplete.  Here is the full thought that was going
through my head:

I'm just a user and have no root access, so all extra format libraries are
in a non-standard place.  When compiling ImageMagick, one of the TIFF
tests is to find the file "tiffio.h".  The TIFF library already installed
did not have this file, so I downloaded a current version and compiled it.
The ImageMagick configure required minor tweaking to find the "tiffio.h"
file in the nonstandard place.  During compilation or linking the already
installed older version
of the TIFF
library was used instead of the new one I had just compiled.  The
offending parameter was __eprintf.  I was able to change the order of
libraries during compilation so it found the current TIFF in the
non-standard directory first and used that one.   ImageMagick compiled
correctly.  

When I then compiled DX, I had to change the order of compilation flags in
the configure script near line 3435 to ac_link=`'${CC-cc} -o
conftest${ac_exeext} conftest.$ac_ext $LIBS  $CFLAGS $CPPFL
AGS $LDFLAGS1>&5'.

Additionally the Makefile in src/exec/dxexec had to be slightly modified
to add -lMesaGL to the library line.

Good luck-
Randy




On Wed, 9 Aug 2000, Thomas Ruedas wrote:

> Randy Polson <[EMAIL PROTECTED]>:
> >When I had trouble with compilation and __eprintf it was because the TIFF
> >library was
> >different than expected.  I don't know if this is the case here.
> I don't think so. At least I downloaded again the latest tiff version
> from the OpenDX site, recompiled it and rebuilt OpenDX with it, but the
> error remains. I didn't see any complaints about TIFF being incomplete
> in configure's output. As the errormsg names libGL.a as the culprit,
> maybe I should use the latest Mesa lib directly from the Mesa site? Or
> are there incompatibilities with certain versions?
> -- 
> ------------------------------------------------------------------------
> Thomas Ruedas
> Institute of Meteorology and Geophysics, J.W.Goethe University Frankfurt
> e-mail: [EMAIL PROTECTED]
> http://www.geophysik.uni-frankfurt.de/~ruedas/
> ------------------------------------------------------------------------
> 

Reply via email to