Hello,
Currently, the temporal (transient) vtkHDF file does not fully support field data with multiple components and tuples. Only the first tuple with all its components is read.
The main issue is that the number of components and tuples can change from one step to another, so we need a way to know their size for each step.
Here is a proposal to add a new group in the hdf format Steps/FieldDataSize. In this group, each field data could have a dataset representing the component and tuple size. For example:
For a field data dataset named FieldData/MyField (with 2 steps):
It should have offsets Steps/FieldDataOffsets/MyField:
And a new dataset specifying its size, Steps/FieldDataSize/MyField, with the first column representing component size and the second column representing tuple size:
Any comments or suggestions are welcome!
Thanks,