What worked before
For about 2 years I use dpf.operators.result.mapdl_section() to get the section thickness of all shells in my models.
dpfdatasource = dpf.DataSources(rstpath)
model = dpf.Model(data_sources=dpfdatasource)
msh = model.metadata.meshed_region
sections = msh .property_field('section')
shellthickness = dpf.operators.result.mapdl_section(properties_name='Thickness', section=sections, data_sources=dpfdatasource).outputs.properties_value()[0] #Thickness per section number
This has worked fine, even for models with stress read only surface shells (which have no section properties).
The problem
I use the same rst files as before. Since I've installed Ansys 2025R2, mapdl_section only works for models without the surface shells. When read only surface shells are in the model (which don't have section properties), I get the following error:
ansys.dpf.gate.errors.DPFServerException: SectionsDataProvider output is not well defined.
mapdl_section_properties:32<-sections_data_provider:35
I checked with colleagues who are using 2024R2 and didn't install 2025R2 yet. For them it still works fine. I've checked that python and ansys-dpf-core version are the same.
Workaround
I have found a workaround in the mean time: run mapdl_section() without the section parameter, if a section number in sections in not in shellthickness, then I know it is a stress read only section and I set the thickness to 0.
Question
I would like to know the reason for this error. I was under the impression that pyDPF only uses the license settings of installed ansys versions. In that case, installing a different Ansys version should not cause a change in running my pyDPF scripts on the same rst files. So apparently pyDPF uses some functions of the Ansys installation?
Thanks for helping me find the reason for this issue!
