# CGNS plugin support for ZonePointers / time-varying topology
**URL:** https://discourse.paraview.org/t/cgns-plugin-support-for-zonepointers-time-varying-topology/17480
**Category:** Development
**Created:** [March 18, 2026, 8:07am UTC](https://discourse.paraview.org/t/cgns-plugin-support-for-zonepointers-time-varying-topology/17480 "2026-03-18T08:07:19Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@nitawa](https://discourse.paraview.org/u/nitawa)
#### Post date: [March 18, 2026, 8:07am UTC](https://discourse.paraview.org/t/cgns-plugin-support-for-zonepointers-time-varying-topology/17480/1 "2026-03-18T08:07:19Z")
Dear ParaView experts and developers,
We are trying to visualize a CGNS file that follows the CGNS SIDS standard for time-varying unstructured meshes with changing topology.
The recommended CGNS SIDS approach for this case is discussed in Chapter 11: _Time-Dependent Flow, Section 11.5.3 “Example - Adapted Unstructured Mesh_ at this link: [CGNS-SIDS](https://cgns.org/standard/SIDS/time.html) and consists in:
- Create a new `Zone_t` each time the topology changes;
- Store `TimeValues`, `NumberOfZones`, and `ZonePointers` under `BaseIterativeData_t`;
- Store the corresponding `FlowSolutionPointers` under each zone’s `ZoneIterativeData_t` and use `Null` entries in `ZoneIterativeData_t` for time steps where a zone is inactive.
We checked that the CGNS files (see attachment) is accepted by `cgnscheck` tool, which means that the structure is valid per the CGNS SIDS standard.
However, ParaView displays all zones at every time step simultaneously, instead of showing only the zone referenced by `ZonePointers` for the current time.
This is consistent with a similar issue reported earlier in VisIt at this link: [visit #20838](https://github.com/visit-dav/visit/discussions/20838), where a maintainer confirmed that currently the Visit CGNS reader does not support the `ZonePointers` construct.
We also found a related thread here from 2023 [paraview #11427](https://discourse.paraview.org/t/cgns-problem-properly-displaying-a-file-with-adaptive-mesh-solution/11427) where a Kitware developer
confirmed that there is currently a limitation if I am not mistaken.
Thus these questions:
1. Does the current ParaView CGNS plugin support `BaseIterativeData_t`/`ZonePointers` for time-varying topology and if not are there any plans to support in the future ? I understand from the discussion on [visit #20838](https://github.com/visit-dav/visit/discussions/20838), that Visit will;
2. In the meantime, is there a recommended workaround for this use case such that we are able to visualize a topology-changing unstructured mesh in CGNS with ParaView?
We provide a minimal example CGNS file if that would help to better understand the issue.
In advance, many thanks for your valuable feedback and your help!
Best regards.
[multi\_zone.tar.gz](https://discourse.paraview.org/uploads/short-url/A0MIw75YBvykYpD30JDjvLfKlvc.gz) (47.4 KB)
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [March 18, 2026, 8:47am UTC](https://discourse.paraview.org/t/cgns-plugin-support-for-zonepointers-time-varying-topology/17480/2 "2026-03-18T08:47:03Z")
> [@nitawa](#):
>
> Does the current ParaView CGNS plugin support `BaseIterativeData_t`/`ZonePointers` for time-varying topology and if not are there any plans to support in the future ? I understand from the discussion on [visit #20838](https://github.com/visit-dav/visit/discussions/20838), that Visit will;
It doesnt support it as stated by Tiffany and Michael in the thread you linked.
We are definitely open to add support to it, but there is no immediate plan as far as I know.
> In the meantime, is there a recommended workaround for this use case such that we are able to visualize a topology-changing unstructured mesh in CGNS with ParaView?
You can load the time steps individually, but that is not ideal I agree.