[Paraview] Query about TableToStructuredGrid filter

[Utkarsh Ayachit]

Opps, I made a typo :).

Here are the filter parameters again:

Whole Extent (0, 149), (0, 151), (0, 2)
X Column: X
Y Column: Y
Z Column: Z

Utkarsh


On Mon, Oct 5, 2009 at 4:51 PM, Utkarsh Ayachit
<utkarsh.ayachit at kitware.com> wrote:
> Lester,
>
> Sorry for the late reply.
>
> I was able to load your attached dataset without any issues with the
> following property values for the TableToStructuredGrid filter:
>
> Whole Extent (0, 140), (0, 151), (0, 2)
> X Column: X
> Y Column: Y
> Z Column: Z
>
> Utkarsh
>
> On Fri, Jul 24, 2009 at 6:54 AM, Lester Anderson
> <lester_anderson1963 at hotmail.com> wrote:
>> Hi Utkarsh
>>
>> Thanks for the info, I can now see how it works from your sample file. I did
>> have to re-order the entries in my test CSV file in order to get it plotting
>> as a cube and not as a distorted one (with the original format). It does
>> appear that the Table to Structured Grid expects the CSV data to be in a
>> particular order, but it does not seem to be able to handle data values
>> entered in effect randomly, even though the positional information (XYZ) is
>> explicit.
>>
>> I did apply a quick test on some real data, a few sampled layers as the
>> whole data are quite chunky! Even though I enter the extents of the data
>> properly, it fails with the the same error message. I have attached a zipped
>> data file for you to see, it is only 3 data layers (z=0, -100, -200), with
>> the units in metres (XYZ).
>>
>> X: 291000 - 321000
>> Y: 2026000 - 2056000
>> Z: -200 - 0 (negative to get depth in the right direction!)
>>
>> I already have the whole dataset as a VTK file (thanks to Berk's help), but
>> this is a useful test of real-world data. The whole dataset forms a volume
>> 30km x 30km x 10km (XYZ) so pretty big!
>>
>> If you can work out why it is not working on the Table to structure grid,
>> that would be a great help. Table to points is perfectly fine with the
>> extents exactly as expected. As longs as the data are on a regular grid,
>> then Paraview should be able to convert this to a structured grid
>> irrespective of the order of the data points. Perhaps it would be useful to
>> have some clarification on the way Paraview processes the data and whether
>> or not we are expected to preformat the input CSV file.
>>
>> Thanks
>>
>> Lester
>>
>>
>>> Date: Thu, 23 Jul 2009 15:23:51 -0400
>>> Subject: Re: [Paraview] Query about TableToStructuredGrid filter
>>> From: utkarsh.ayachit at kitware.com
>>> To: lester_anderson1963 at hotmail.com
>>> CC: paraview at paraview.org
>>>
>>> Attached is a sample csv file.
>>> Since the CSV file has 8 points, we can create a grid with extents
>>> (2x2x2). Hence set the Whole Extent on the "Table To Structured Grid"
>>> filter to be (0, 1, 0, 1, 0, 1). Then choose the columns to use for
>>> the point coordinates and hit apply.
>>>
>>> I guessing what you are missing is setting of the "Whole Extent" on the
>>> panel.
>>>
>>> Utkarsh
>>>
>>> On Thu, Jul 23, 2009 at 11:25 AM, Lester
>>> Anderson<lester_anderson1963 at hotmail.com> wrote:
>>> > Hello
>>> >
>>> > It is good to see that there is now CSV import functionality, which
>>> > works
>>> > fine for displaying points (e.g. using Glyphs), but if you want to do
>>> > more
>>> > interesting work, you need a structured grid. The option to take the
>>> > loaded
>>> > CSV data and convert it to a Structured Grid is listed (Table to
>>> > Structure
>>> > Grid). However, this always fails with the same error message (Windows
>>> > XP
>>> > and Vista)
>>> >
>>> > ERROR: In ..\..\..\src\VTK\Graphics\vtkTableToStructuredGrid.cxx, line
>>> > 98
>>> > vtkPTableToStructuredGrid (0A195768): The input table must have exactly
>>> > 1
>>> > rows. Currently it has 8 rows.
>>> >
>>> >
>>> > ERROR: In ..\..\..\src\VTK\Filtering\vtkExecutive.cxx, line 757
>>> > vtkCompositeDataPipeline (0A1952A0): Algorithm
>>> > vtkPTableToStructuredGrid(0A195768) returned failure for request:
>>> > vtkInformation (09987D80)
>>> > Debug: Off
>>> > Modified Time: 9473124
>>> > Reference Count: 1
>>> > Registered Events: (none)
>>> > Request: REQUEST_DATA
>>> > FORWARD_DIRECTION: 0
>>> > ALGORITHM_AFTER_FORWARD: 1
>>> > FROM_OUTPUT_PORT: 0
>>> >
>>> > I have tested this on a multilayer volume and it fails as above: the
>>> > Table
>>> > to Points always works. What we want is to have Paraview
>>> > interpolate the data between the points such that a volume can be sliced
>>> > etc. Is there a problem with this option?
>>> >
>>> > Lester
>>> >
>>> >
>>> > ________________________________
>>> > Windows Live Messenger: Celebrate 10 amazing years with free winks and
>>> > emoticons. Get Them Now
>>> > _______________________________________________
>>> > 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
>>> >
>>> >
>>
>> ________________________________
>> Celebrate a decade of Messenger with free winks, emoticons, display pics,
>> and more. Get Them Now
>