[vtk-developers] Re: VTK Python Installation [was: VTK 5.0 branch in mid-August]

David Gobbi dgobbi at atamai.com
Tue Aug 2 14:01:47 EDT 2005


Prabhu Ramachandran wrote:

>"David" == David Gobbi <dgobbi at atamai.com> writes:
>[...]
>    David> To put it briefly, if our current batch of *Python.so files
>    David> are split into *Python.so and *PythonD.so files, and if all
>    David> these files are put into the right places (where "right" is
>    David> defined by Python and UNIX conventions and not by any other
>    David> ideals), then the system will work without any of these
>    David> things that Prabhu and myself refer to as "hacks".
>
>    David> As it stands right now, the *Python.so files need to be in
>    David> python's path as well as the system runtime-library path at
>    David> the same time.  This is only possible by using
>    David> LD_LIBRARY_PATH, by doing nasty hacks in the .py files, or
>    David> by (shudder) using rpath.
>
>Many thanks for "entering the fray".  I am in total agreement with
>what you say.
>
>>From what I have tested with my current build, the os.chdir() trick
>will work only if '.' is in the rpath.  If not, when you import
>libvtkRenderingPython.so it will not be able to find the other
>libvtk*Python.so's that it links to.
>  
>
If there was a way to use rpath to force python modules to search for 
.so files in their
own directory, then that would be fine.  But as far as I know, rpath 
with relative
paths doesn't work this way, and using a chdir trick just doesn't seem 
right.
However, if it works, it is worth considering unless it is known to 
cause other
things to break.






More information about the vtk-developers mailing list