David,
Have you had time to look at this yet? If not, I should get a chance today
to scour the CMakeLists.txt files to see if the issue lies there. Notably, I
have tried it with 4 64-bit installs of boost now (1_36_0 to 1_39_0) and
CMake reports finding version 1.33.1 each time.
Regards,
Hello,
Is there a way to suppress the message you get each time you open pvpython
i.e.:
paraview version 3.7.0, Date: 2009-10-03
Regards,
Adriano
===
Adriano Gagliardi MEng PhD
Business Sector Leader
Computational Aerodynamics
Aircraft Research Association
Hi Pat,
Yes -- I should have gone with my first suspicion. It is indeed a memory
leak. I'm obviously not cleaning stuff up properly... which doesn't surprise
me because my script really is kind of a hack... I have attached it.
Basically it reads a directory of vtk files, loops over each,
Hello,
In PARAVIEW_DIR\lib\paraview-3.8\site-packages\paraview\servermanager.py
comment the lines 2947 and 2949.
if hasattr(sys, ps1):
# session is interactive.
print vtkSMProxyManager.GetParaViewSourceVersion();
Aurélien.
2011/2/11 Adriano Gagliardi agaglia...@ara.co.uk
Hello,
Thanks, Aurelian. I don't like random hacks, but if that is the only option
for the moment, so be it.
Thanks again.
===
Adriano Gagliardi MEng PhD
Business Sector Leader
Computational Aerodynamics
Aircraft Research Association Ltd.
Manton Lane
Bedford
Tel:
Out of curiosity, why would you want to disable the message? It's only
shown when pvpython is run as an interactive session, never when a
script argument is specified or with pvbatch.
Utkarsh
2011/2/11 Adriano Gagliardi agaglia...@ara.co.uk:
Thanks, Aurelian. I don't like random hacks, but if
I receive the message when executing a Python script with pvpython or
pvbatch after 'from paraview.simple import *'. Whether I run it as:
pvpython ./my-script.py
pvbatch ./my-script.py
or if I include the executable at the top of the Python script ie:
#!/usr/bin/env pvpython or #!/usr/bin/env
That's odd. Maybe you have something in your pythonrc or pythonstartup that
is defining sys.ps1. Try adding this before the import paraview line in
your script:
import sys
del sys.ps1
Pat
On Fri, Feb 11, 2011 at 9:09 AM, Adriano Gagliardi agaglia...@ara.co.ukwrote:
I receive the message
I've tried compiling the git source on a 32 and a 64 bit machine. Both
run debian testing.
I've tried with MPI2PY ON and OFF
I'm using MPI YES, Python YES
Both stop with the error (see attach)
Any ideas?
Ricardo Reis
'Non Serviam'
PhD candidate @ Lasef
Computational Fluid
It looks like you need to install the mpi headers for the version of mpi you
have installed on your machine.
On Fri, Feb 11, 2011 at 1:10 PM, Ricardo Reis rr...@aero.ist.utl.pt wrote:
I've tried compiling the git source on a 32 and a 64 bit machine. Both run
debian testing.
I've tried
Interesting. Unfortunately, I cannot reproduce this. Would you mind
helping me debug this? You'll need a debug build of ParaView built
from source however. I'd recommend building using the latest
git-master.
Thanks
Utkarsh
On Fri, Feb 11, 2011 at 12:39 PM, Moore, Branden J (GE Global
Research)
Richard,
The problem is indeed due to offscreen rendering. I've added a bug
(http://paraview.org/Bug/view.php?id=11847). Until this is resolved,
the work-around would be to uncheck Use offscreen rendering for
screenshots in Edit | settings dialog (Render View page).
Utkarsh
On Thu, Feb 10,
On Fri, 11 Feb 2011, Robert Maynard wrote:
It looks like you need to install the mpi headers for the version of mpi you
have installed on your machine.
I have them. I regularly compile mpi programs on these machines, either my
own or from third parties...
I have checked, this directory is
Would it be possible for you to post the CMakeCache.txt for the build?
On Fri, Feb 11, 2011 at 2:00 PM, Ricardo Reis rr...@aero.ist.utl.pt wrote:
On Fri, 11 Feb 2011, Robert Maynard wrote:
It looks like you need to install the mpi headers for the version of mpi
you
have installed on your
Have you tried using Plot On Intersection Curves filter?
Utkarsh
On Thu, Feb 10, 2011 at 1:32 PM, Dan Bower d...@gps.caltech.edu wrote:
Hi all,
I am plotting a spherical data volume using structuredgrid. I also
want to plot line data on the surface of the sphere. Presently the
line data
On Fri, 11 Feb 2011, Robert Maynard wrote:
Would it be possible for you to post the CMakeCache.txt for the build?
sure
attached
Ricardo Reis
'Non Serviam'
PhD candidate @ Lasef
Computational Fluid Dynamics, High Performance Computing, Turbulence
http://www.lasef.ist.utl.pt
I ended up writing a vtkpython script to convert line data from a GMT
format to a vtkPolyData structure. Write out the vtp file and load
into Paraview. Seems to work well.
Cheers,
Dan
On Fri, Feb 11, 2011 at 11:21 AM, Utkarsh Ayachit
utkarsh.ayac...@kitware.com wrote:
Have you tried using
I suggest replacing the Calculator and DescriptiveStatistics filters
with one python programmable filter (or python calculator if you are
using 3.10). That way the data type isn't changed and Fetch will do
what you expect it to - produce a standard vtkDataSet with 1
point/cell.
The python
Thanks Michael.
I guess you are right that extra time is due to overlapping the ghost
layers.
If you want more performance, you should use .pvti and .vti instead .pvtr
and .vtr to store rectilinear data.
My grid is not necessarily uniform even in one direction. If I remember
correctly, *.vti
If there are no suggestions in terms of Amazon machines, perhaps some info
somewhere regarding offscreen rendering for a headless paraview server? I
have searched through the mail archives to the best of my ability, and
followed the instructions on the server build wiki, and I am just not quite
Hi Matthew,
The ParaView Users Guide has a section on setting up offscreen rendering
using OSMesa which you can read at:
http://www.paraview.org/Wiki/ParaView/Users_Guide/Parallel_Rendering#Offscreen_software_rendering_via_OSMesa
On Fri, Feb 11, 2011 at 4:56 PM, Matthew Dillon
Thank you, I have been using that section as a guideline for getting this
thing set up. Perhaps someone on the list has a recommendation for a known
version of mesa that works with PV 3.8/3.10?
On Fri, Feb 11, 2011 at 1:03 PM, Robert Maynard
robert.mayn...@kitware.comwrote:
Hi Matthew,
The
Try MesaLib-7.6.1.tar.gz
Make sure you configure with --driver osmesa. The result should be a
libOSMesa and libGLU, no libGL. Paraview should link only with libOSMesa,
no to libGL. When you're done build paraview, an ldd on the binary should
reveal no dependencies on libGL, only libOSMesa.
Also, when compiling set VTK_USE_X to OFF. An ldd on your binary should
show no X11 library dependencies either. When compiled without X, you dont
need to specify any --use-offscreen-rendering flag to pvserver. It is
possible to compile with X and libGL and OSMesa, but it sounds like you are
Pat,
Alrighty, I will make sure to do that. I think the best option is to do a
pure osmesa build (it sounds like its simpler at the end of the day...) but
I am certainly up for recommendations otherwise. Thanks
Matt
On Fri, Feb 11, 2011 at 1:37 PM, pat marion pat.mar...@kitware.com wrote:
Also,
Cross post from vtk developers list. I apologize for the spam. Please do
not reply to this message!
-- Forwarded message --
From: pat marion pat.mar...@kitware.com
Date: Fri, Feb 11, 2011 at 5:49 PM
Subject: new widget to spy on your vtk objects
To: vtkdev
Can I still do volume rendering with osmesa, or is this a gpu thing?
On Fri, Feb 11, 2011 at 1:42 PM, Matthew Dillon mrdil...@alaska.edu wrote:
Pat,
Alrighty, I will make sure to do that. I think the best option is to do a
pure osmesa build (it sounds like its simpler at the end of the
Hmm, I was greeted with a few warnings when generating the cmake build
files:
/begin quote
WARNING: Target vtkRendering requests linking to directory
/home/ubuntu/osmesa/lib. Targets may link only to libraries. CMake is
dropping the item.
WARNING: Target vtkRendering requests linking to
The OSMESA_LIBRARY should also point to libOSMesa.so
(btw you can also compile leaving OPENGL_gl_LIBRARY and OPENGL_glu_LIBRARY
empty. I found I needed to do this when compiling static to avoid link errors
in, I think, the XDMF library).
On 2/11/11 6:32 PM, Matthew Dillon mrdil...@alaska.edu
Hi,
A couple of bugs turn up when building 3.8.1 using portland group
compiler (see below) I attached a patch for each, hoping they can make
it into the 3.10 release if they are not already there.
Burlen
[1]
[ 84%] Building CXX object
30 matches
Mail list logo
