[Paraview] Maybe a silly question?

[David E DeMarle]

The ex2 object is a reader, it won't have data in it. The ex2 object
should have a control that you can use to specify which arrays to
read/ignore from the file but that is different from what you are
asking for.
In any case when you update the reader it it will produce data, and
that you already know how to fetch and inspect.

David E DeMarle
Kitware, Inc.
R&D Engineer
28 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-371-3971 x109



On Wed, Mar 23, 2011 at 11:50 PM,  <wastrel at gmail.com> wrote:
> Thanks David, I made a programmable filter and got it running. It works.
> Now I have a new question, how to get a point_data array from a ex2 object?
> For vtk object, I know I can fetch it to local, like
> from paraview.simple import servermanager as sm
> data=sm.Fetch(reader)
> Then there is data.GetPointData().GetArray(). But I could not find such
> routines when I tried to do the same to ex2 object.
> Any idea?
>
>
> On Wed, Mar 23, 2011 at 5:01 PM, David Thompson <dcthomp at sandia.gov> wrote:
>>>
>>> Thanks again, and I took a look on the programmable filter code. In my
>>> case the block consists of more than 1 million tetrahedrons and it will be
>>> time consuming to run iterations. Is there a way that I can export the cells
>>> and their corresponding point information to a numpy array? The ultimate
>>> goal is to calculate the volume of each tetrahedron.
>>>
>>> Your help is much appreciated!
>>
>> There is a VTK-numpy bridge which is documented on the wiki. I don't
>> recall exactly where but I'm sure a search engine will find it for you.
>>
>>        David
>>
>>> On Wed, Mar 23, 2011 at 4:45 PM, David Thompson <dcthomp at sandia.gov>
>>> wrote:
>>>
>>> I have an Exodus II type file loaded into Paraview and trying to get the
>>> point ID for a group of cells. I know for vtk type object there are
>>> functions like GetCell(), GetPoint()
>>>
>>> I assume you're talking about using Python from ParaView?
>>>
>>> Yup, that's exactly what I am doing here.
>>>
>>> There are 2 ways to use Python in ParaView: writing a programmable filter
>>> and scripting the user interface. The former is evaluated on the server
>>> where you can get access to (at least a portion of) the actual dataset. The
>>> latter is evaluated on the client, which does not have access to the
>>> datasets -- instead it has proxy objects that send messages to the servers,
>>> telling them what to do. So, you can access the points and cells from within
>>> a programmable filter, but not from within a client-side script.
>>>
>>>
>>>
>>> and just wonder if there are such kind of routines available for Exodus
>>> II files.
>>>
>>> Yes, the ExodusII reader outputs VTK objects in a multiblock dataset. You
>>> can use GetCell(), GetPoint(), and such on each individual block... you just
>>> have to decide on a particular block. The ExodusII reader separates things
>>> into blocks because each one may have different point and/or cell variables.
>>> There are actually 2 levels of blocks; the top level segregates datasets
>>> into groups that correspond to exodus blocks vs. sets. The second level
>>> breaks individual blocks or sets into separate datasets. Does that help?
>>>
>>>
>>> I am new in dealing with ex2 file, could you please be detailed? There is
>>> just one block in the dataset, so how to get the vtk object out from it so I
>>> can access the points and cells?
>>>
>>>
>>> See the section of this page
>>>  http://www.vtk.org/Wiki/Python_Programmable_Filter
>>> named "Dealing with Composite Datasets". It describes how to iterate over
>>> all of the blocks in the dataset and run a function (in this case, one named
>>> "flatten") on each block.
>>>
>>>       David
>>>
>>
>
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
>