[Paraview] Vtu file issue

[Moreland, Kenneth]

(Replying back to the ParaView mailing list because, contrary to popular opinion, I don't know everything. :)

Now that I'm looking at it, implementing this in ParaView is a bit harder than I originally thought.  However, it is pretty easy to write a VTK (not ParaView) Python script to convert the one data set to a series of data files.  I have attached such a script.  You need to run this with vtkpython (as opposed to, say pvpython or in the ParaView GUI).  I'm not sure if this executable comes with the ParaView binaries, but it definitely something that is built from the ParaView source.

The "right" way to implement this would be to create a new reader for ParaView that used the vtu format and the odd array-name-is-time convention.  It should be pretty trivial for a developer, but not something I would want to be someone's first experience with ParaView.

-Ken


On 4/15/10 8:24 PM, "wastrel at gmail.com" <wastrel at gmail.com> wrote:

By the way I attached the original data. Let me know what's your take on that, if you had a chance to take a look. Thanks!



On Thu, Apr 15, 2010 at 9:18 PM,  <wastrel at gmail.com> wrote:
Thanks for the reply.

Now I am thinking to split the original vtu file to a series of vtu files according time series,

I can use the reader to get the data info as follow,
reader=XMLUnstructuredGridReader(FileName=fname);

when writing each individual vtu file, I need to delete other variables except the one corresponding to current time instance. Now the problem is that I can not get the data, all I got from the reader is data information. Do you know how to change the data structure of reader so XMLUnstructureedGridWriter(Input=reader,FileName=fname) can be used to generate the new files?

I am totally new to Paraview, and any suggestions is welcome.

Thanks,

Ken



On Thu, Apr 15, 2010 at 3:20 PM, Moreland, Kenneth <kmorel at sandia.gov> wrote:
>
> ParaView has no direct mechanism to recognize those multiple variables as a time series.  The most straightforward way to do something like that would be to create a filter that reports a time series based on the variables it encounters in RequestInformation and then selects the appropriate variable (with proper name change) in RequestData.
>
> Unless you are already developing C++ plugins, the quickest way to develop such a filter would be to use the Python programmable filter.
>
> http://www.paraview.org/Wiki/Python_Programmable_Filter
>
> -Ken
>
>
> On 4/14/10 7:09 PM, "wastrel at gmail.com" <wastrel at gmail.com> wrote:
>
> Hi,
>
> I am dealing with a set of data with vtu format. There is a variable changing over time, but somehow its value was stored separately at each time instance and with different variable names. Say the variable is x, and at each time instance its value was stored with x_001, x_002,x_003,....x_100. So now I am wondering how to combine these to a single variable x with time indices in Paraview. Your input is highly appreciated!
>
> Ken
>
>
>
>
>    ****      Kenneth Moreland
>     ***      Sandia National Laboratories
> ***********
> *** *** ***  email: kmorel at sandia.gov
> **  ***  **  phone: (505) 844-8919
>     ***      web:   http://www.cs.unm.edu/~kmorel <http://www.cs.unm.edu/%7Ekmorel>
>





   ****      Kenneth Moreland
    ***      Sandia National Laboratories
***********
*** *** ***  email: kmorel at sandia.gov
**  ***  **  phone: (505) 844-8919
    ***      web:   http://www.cs.unm.edu/~kmorel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20100416/e5edaff6/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: TimeConvert.py
Type: application/octet-stream
Size: 825 bytes
Desc: TimeConvert.py
URL: <http://www.paraview.org/pipermail/paraview/attachments/20100416/e5edaff6/attachment.obj>