<html>
All the VTK Kalgan dashboards including Borland static and shared with
tcl are<br>
being built with the nightly CMake from CVS. As a side note,
this test<br><br>
<font color="#0000FF"><u>TestEmptyInput</u></font> <br><br>
is failing with a floating point exception on borland.<br>
The class that causes this is:<br>
vtkQuadricClustering<br>
I do not have a good debugger for borland, so I can not tell you
where,<br>
but I suspect a divide by zero. <br><br>
-Bill<br><br>
<br>
At 10:22 PM 10/24/2002 -0400, Bill Lorensen wrote:<br>
<blockquote type=cite class=cite cite>I would avoid the nightly cmake. I
had a similar problem. I've reverted to using -rLatestRelease.<br><br>
Bill<br><br>
At 09:44 PM 10/24/02 +0100, John Biddiscombe wrote:<br>
>> Well that's pants.<br>
><br>
>Strong words indeed!<br>
><br>
>>Are we talking a Borland build here? What CMake<br>
>> version/CVS? What VTK version/CVS?<br>
><br>
>CMake from CVS earlier today, vtk from CVS earlier today. Borland
build, but<br>
>I am using my own LocalBorlandGenerator and GlobalBorlandGenerator
and not<br>
>the ones supplied with Cmake (though this is not the problem I'm
sure).<br>
><br>
>I'll experiment with wiping a few directories (+caches) and
recreating<br>
>stuff.<br>
><br>
>JB<br>
><br>
><br>
>_______________________________________________<br>
>vtk-developers mailing list<br>
>vtk-developers@public.kitware.com<br>
><a href="http://public.kitware.com/mailman/listinfo/vtk-developers" eudora="autourl">http://public.kitware.com/mailman/listinfo/vtk-developers</a>
<br><br>
<br>
_______________________________________________<br>
vtk-developers mailing list<br>
vtk-developers@public.kitware.com<br>
<a href="http://public.kitware.com/mailman/listinfo/vtk-developers" eudora="autourl">http://public.kitware.com/mailman/listinfo/vtk-developers</a>
</blockquote></html>