# Not able to read transient UnstructuredGrid data in Paraview
**URL:** https://discourse.paraview.org/t/not-able-to-read-transient-unstructuredgrid-data-in-paraview/14409
**Category:** ParaView Support
**Tags:** vtkhdf
**Created:** [April 16, 2024, 6:46am UTC](https://discourse.paraview.org/t/not-able-to-read-transient-unstructuredgrid-data-in-paraview/14409 "2024-04-16T06:46:35Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@mrx](https://discourse.paraview.org/u/mrx)
#### Post date: [April 16, 2024, 6:46am UTC](https://discourse.paraview.org/t/not-able-to-read-transient-unstructuredgrid-data-in-paraview/14409/1 "2024-04-16T06:46:35Z")
Hi,
I am trying to write a vtkhdf file containing transient PointData. I read the manual and tried to write a vtkhdf file with dummy data. But I am not able to read that in Paraview, Although I do see the pressure dataset and the timesteps as well in Paraview.
The writer python code is as follows
```auto
try:
import h5py as h5
import numpy as np
import os
import sys
except:
print( "Failed to import")
fType = 'f'
idType = 'i8'
class HfuVtkHdf:
def writeHdf(self):
try:
hdfObj = h5.File('test.vtkhdf', 'w')
# Generate the root content which will have the Version and Type attributes
root = hdfObj.create_group('VTKHDF', track_order=True)
root.attrs['Version'] = (2,1)
ascii_type = 'UnstructuredGrid'.encode('ascii')
root.attrs.create('Type', ascii_type, dtype=h5.string_dtype('ascii', len(ascii_type)))
self.generate_geometry_structure(root)
hdfObj.close()
print( "done..." )
except:
hdfObj.close()
print("Exception in writeHdf")
def generate_geometry_structure(self,f):
try:
root = f
crd = np.array([[0., 0., 0.],
[5., 0., 0.],
[5., 5., 0.]])
res = np.array([1.0 ,3.0 ,5.0 ,6.0 ,7.0 ,8.0 ,9.0 ,11.0 ,13.0])
crd = crd.astype(np.float32)
res = res.astype(np.float32)
cnn = np.array([0, 1, 2])
cnn = cnn.astype(np.int64)
osets = np.array([0, 3])
osets = osets.astype(np.int64)
etypes = np.zeros( 1, dtype='i' ) + 5
etypes = etypes.astype(np.uint8)
mat = [0]
dset1 = root.create_dataset('NumberOfPoints', (1,), maxshape=(None,), dtype=idType)
dset1[0] = 3
dset4 = root.create_dataset('NumberOfConnectivityIds', (1,), maxshape=(None,), dtype=idType)
dset4[0] = 3
dset5 = root.create_dataset('NumberOfCells', (1,), maxshape=(None,), dtype=idType)
dset5[0] = 1
dset2 = root.create_dataset('Types', shape = etypes.shape, data = etypes, maxshape=(None,), dtype=np.uint8)
dset3 = root.create_dataset('Points', shape = crd.shape, data = crd, maxshape=(None,3), dtype=fType)
dset6 = root.create_dataset('Offsets', shape=osets.shape,data = osets, maxshape=(None,), dtype=idType)
dset7 = root.create_dataset('Connectivity', shape=cnn.shape,data = cnn, maxshape=(None,), dtype=idType)
pData = root.create_group('PointData')
pData.create_dataset('pressure', shape= res.shape, data = res, maxshape=(None,), dtype=fType)
steps = root.create_group('Steps')
steps.attrs['NSteps'] = 3
timeSteps = np.array([0, 3, 6])
timeSteps = timeSteps.astype(np.int64)
steps.create_dataset('Values', shape= timeSteps.shape, data = timeSteps, maxshape=(None,), dtype=fType)
demotypes = np.zeros( 3, dtype='i' )
demotypes = demotypes.astype(np.int64)
singleDSs = ['PartOffsets', 'PointOffsets']
for name in singleDSs:
#steps.create_dataset(name, (0,), maxshape=(None,), dtype=idType)
steps.create_dataset(name, shape= demotypes.shape, data = demotypes, maxshape=(None,), dtype=fType)
nTopoDSs = ['CellOffsets', 'ConnectivityIdOffsets']
for name in nTopoDSs:
#steps.create_dataset(name, (0,4), maxshape=(None,4), dtype=idType)
steps.create_dataset(name, shape= demotypes.shape, data = demotypes, maxshape=(None,), dtype=fType)
pData = steps.create_group('PointDataOffsets')
prOffsets = np.array([0, 3, 6])
prOffsets = prOffsets.astype(np.int64)
pData.create_dataset('pressure', shape= prOffsets.shape, data = prOffsets, maxshape=(None,), dtype=fType)
except:
print("Exception in generate_geometry_structure")
if __name__ == ' __main__':
print( "Starting...")
obj = HfuVtkHdf( )
obj.writeHdf()
```
The error is as follows:
ERROR: In vtkExecutive.cxx, line 730
vtkPVCompositeDataPipeline (00000201EBCC3060): Algorithm vtkFileSeriesReader (00000201EB97CE40) returned failure for request: vtkInformation (00000201F039C550)
Debug: Off
Modified Time: 452274
Reference Count: 1
Registered Events: (none)
Request: REQUEST\_DATA
FORWARD\_DIRECTION: 0
ALGORITHM\_AFTER\_FORWARD: 1
FROM\_OUTPUT\_PORT: 0
Attaching the hdf file as well:
[test.vtkhdf](https://discourse.paraview.org/uploads/short-url/A4Eba5KcuNEw89k1bthbD5bJAyi.vtkhdf) (41.9 KB)
Please tell me what I am missing. Thanks!
---
### Author:  [@Louis\_Gombert](https://discourse.paraview.org/u/Louis_Gombert)
#### Post date: [April 16, 2024, 8:07am UTC](https://discourse.paraview.org/t/not-able-to-read-transient-unstructuredgrid-data-in-paraview/14409/2 "2024-04-16T08:07:50Z")
Hi, you’re not too far off! The `NumberOfPoints`, `NumberOfConnectivityId` and `NumberOfCells` datasets need to be of size 3 (your number of timesteps), and the offsets arrays should be of `IdType` type. The following file should open in ParaView.
[test.vtkhdf](https://discourse.paraview.org/uploads/short-url/cXYgARCw7FFIxuOR8w2xqwTz5DB.vtkhdf) (41.8 KB)
Hope this helps!
fyi @Lucas_Givord @mwestphal
---
### Author:  [@mrx](https://discourse.paraview.org/u/mrx)
#### Post date: [April 16, 2024, 8:36am UTC](https://discourse.paraview.org/t/not-able-to-read-transient-unstructuredgrid-data-in-paraview/14409/3 "2024-04-16T08:36:52Z")
Thanks a lot! That worked for me!