Hi Everyone,
I have a question about paraview python calculator strain.
My questions are:
1. If want to calculate strain form a displacement field (3 component
vecotor),
is it correct to use directly
strain(deformationfield_vector)
2. If the above is correct, then what kind of strain is it?
That sounds odd. I tried the same with git-master Paraview and it
seems to save out N independent blocks just fine. Can you copy+paste
the header from one of the vti files?
Utkarsh
On Mon, Aug 22, 2011 at 6:06 PM, Jesus Pulido jpul...@ucdavis.edu wrote:
I am trying to save a generated dataset
Burlen,
Good points. At the risk of stating the obvious let's note that we'll
be able to distribute different pvsc versions for the server for
different OSs and ParaView versions without any issue. Now locating
ssh/xterm/plink etc are indeed real issues.
I don't like the cmake approach, it
Have you tried using vtkErrorMacro or vtkWarningMacro?
Utkarsh
On Thu, Aug 18, 2011 at 1:13 PM, Federico Milano fmil...@gmail.com wrote:
Hi. Is there any way to log messages into the Paraview window that opens
when an error occurs and then inspect this log if something went wrong in
the
Folks,
For the longest time ParaView has been using an odd/even version
number scheme: N.{odd}.* was treated as a development version while
N.{even}.* was treated as a release version. There were a couple of
problems:
* One could never tell two development checkouts apart. Since
development
I know for our HPC access, they require a custom version of SSH that has been
security enhanced. So if PV used a built-in libssh, we would be unable to use
PV to connect through to them.
I don't have a better solution, just wanted to point out that packaging libssh
without an option to use a
Tim,
That's was exactly what I was concerned about when it comes to using
libssh. Good to know that there are indeed such use-cases and not
merely hypothetical ones.
Utkarsh
On Tue, Aug 23, 2011 at 11:18 AM, Tim Gallagher
tim.gallag...@gatech.edu wrote:
I know for our HPC access, they require
Sure. Here is the header for the .pvti file
VTKFile type=PImageData version=0.1 byte_order=LittleEndian
PImageData WholeExtent=0 255 0 255 0 255 GhostLevel=0 Origin=0 0 0
Spacing=1 1 1
PPointData Scalars=RTData
PDataArray type=Float32 Name=RTData/
/PPointData
Piece Extent=0
Jesus,
I verified that it's indeed a bug with 3.10.1. However it has been
fixed since and the fix will be included in the upcoming 3.12.
Utkarsh
On Tue, Aug 23, 2011 at 11:24 AM, Jesus Pulido jpul...@ucdavis.edu wrote:
Sure. Here is the header for the .pvti file
VTKFile type=PImageData
Hi Utkarsh,
It sounds like you need a very flexible solution, or even multiple
solutions, given the variety of use cases you have to support.
You hit on a great point re python approach
allows us to write code to discover informationabout the client/server
and take appropriate action(s).
John, I missed it. thanks for pointing it out.
Burlen
On 08/22/2011 01:27 PM, Biddiscombe, John A. wrote:
Burlen
Something far easier and less intrusive might be to add a file dialog
element to the pvsc syntax so that the user could easily browse for
Have a look at
We just pick a random port between 2000 and 65000 ... the chances are pretty
slim (but not absolutely zero) for stepping on another process, right? 8-)
Rick Angelini
USArmy Research Laboratory
Computation ScienceEngineering Branch/
ARL/DSRC (DoD Shared
Yes, and this has worked OK, at least no one complained, but I think
that it's success has more to do with a small number of users involved
rather than the strength of this approach.
On 08/23/2011 08:52 AM, Angelini, Richard C USA CIV (US) wrote:
We just pick a random port between 2000 and
Hi Blaise,
The Lion compatible ParaView 3.10.1 binary is now available.
http://paraview.org/paraview/resources/software.html
Direct Link:
http://paraview.org/files/v3.10/ParaView-3.10.1-OSX-10.7-x86_64.dmg
Note that this binary does not include MPI. In the next minor release we
will most
Problem: When building a long pipeline on a large dataset, the rendering
of the result from each added filter takes a long time, for an intermediate
result that is of no interest.
Question: Is there a way to force ParaView (3.10.1) to use Outline
representation for each added filter? Or
Hmm,
I haven't been able to reproduce it yet. I suspected a simple numpy
dependency problem, but the code in vtkPythonProgrammableFilter.cxx
where numpy is imported does so in a try block. That code should
suffice and doesn't look like the code you've posted.
Which version of ParaView, which OS
Hi Sudhir,
I'm glad you managed to deploy successfully ParaViewWeb specially with
the latest version of activemq-cpp.
Regarding your issue I have an idea, but I might need also some more
feedback from you.
First of all, what kind of data did you tried to render ? Did you
used, one of our sample
17 matches
Mail list logo
