[Paraview] XDMF using ORIGIN_DXDYDZ switches X- and Z-dimension

[Jens]

Hi Jerry,

this mixed definition results in quite a bit of problems.

What if I add a point-cloud using POLYVERTEX(saved as
2d-hdf5-array..each row one xyz-coords) to a scalar-field on a uniform
grid (saved as 3d-hdf5-array). The scalar-field will switch X and Z, but
the xyz-points of the point-cloud will not.

I have to save the data in the hdf5-file for 3d-data with flipped XZ and
for point-data without.

Greetings
Jens


Jerry Clarke schrieb:
> Jens,
> 
> It's a little confusing but it's done that way to be consistent.
> The first dimension in Topology is the slowest changing dimension not
> necessarily "X" (Think Plot3D curvilinear K-J-I). i.e. Topology and
> Geometry are not the same.
> 
> 
> Jerry
> 
> Jens wrote:
>> Hi XMF-User,
>>
>> I wonder why the following test-case, switches X and Z-dimension-size in
>> 3.4.x and 3.6.1.
>> X should be  10.0, but shows 100.0 in paraview.
>> Z should be 100.0, but shows  10.0 in paraview.
>> Any idea?
>>
>> Greetings
>> Jens
>>
>> ---------------
>> <?xml version="1.0" ?>
>> <Xdmf Version="2.0" xmlns:xi="http://www.w3.org/2001/XInclude" >
>>
>>   <Domain Name="XDMF-Test">
>>     <Grid Name="1-1-1_t0.0" Type="Uniform">
>>
>>        <Topology
>>           TopologyType="3DCORECTMESH" Dimensions="10 50 100">
>>        </Topology>
>>
>>        <Geometry GeometryType="ORIGIN_DXDYDZ">
>>           <DataItem DataType="Float" Dimensions="3" Format="XML">
>>              10 50 100
>>           </DataItem>
>>           <DataItem DataType="Float" Dimensions="3" Format="XML">
>>              1.0 1.0 1.0
>>           </DataItem>
>>        </Geometry>
>>
>>     </Grid>
>>   </Domain>
>>
>> </Xdmf>
>> _______________________________________________
>> 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
>