# Remove dataset points via programmable filter
**URL:** https://discourse.paraview.org/t/remove-dataset-points-via-programmable-filter/8639
**Category:** ParaView Support
**Created:** [December 16, 2021, 4:14pm UTC](https://discourse.paraview.org/t/remove-dataset-points-via-programmable-filter/8639 "2021-12-16T16:14:02Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@Captain\_Convergence](https://discourse.paraview.org/u/Captain_Convergence)
#### Post date: [December 16, 2021, 4:14pm UTC](https://discourse.paraview.org/t/remove-dataset-points-via-programmable-filter/8639/1 "2021-12-16T16:14:02Z")
Dear all,
I am trying to remove points from a dataset based on geometrical conditions. In order to do so I have to get the coordinates of the dataset, select which ones fulfill my conditions and then generate **a new dataset with different number of pointData**. I tried an approach like this one:
```auto
import vtk
from vtkmodules.vtkCommonDataModel import vtkDataSet
import vtk.numpy_interface.dataset_adapter as dsa
import numpy as np
x_old = inputs[0].GetPointData().GetArray('x')
z_old = inputs[0].GetPointData().GetArray('z')
x_new=[]
z_new=[]
cont=0
for i in range(len(x_old)) :
if x_old[i]>0.12 :
x_new.append(x_old[i])
z_new.append(z_old[i])
cont=cont+1
output.PointData.append(x_new, 'x_new')
output.PointData.append(z_new, 'z_new')
```
However I think that I am having errors because the length of **x\_new** is not the same as the initial one ( **x\_old** ). How could I generate a new dataset with some of the previous data points with their respective arrays? Thank you in advance!
---
### Author:  [@Captain\_Convergence](https://discourse.paraview.org/u/Captain_Convergence)
#### Post date: [December 17, 2021, 4:44pm UTC](https://discourse.paraview.org/t/remove-dataset-points-via-programmable-filter/8639/2 "2021-12-17T16:44:31Z")
Is there any example of generating unstructured data from a set of coordinates in python? I couldn’t find an example like this.