RST built without geometry data

[mcMunich]

Before submitting the issue

• I have checked for Compatibility issues
• I have searched among the existing issues
• I am using a Python virtual environment

Description of the bug

A common method to avoid extraneous data in a .rst is to use the APDL command /CONFIG,NORSTGM,1 to avoid writing extra data to the .rst file. But when you do that, you are then unable to use pydpf to read in the data. Some geo off .rst data should be attached to this ticket.

https://ansyshelp.ansys.com/account/Secured?returnurl=/Views/Secured/corp/v251/en/ans_cmd/Hlp_C_CONFIG.html

This is a not unhead of variation of FE usage to minimize storage requirements and while there are work arounds, it would be useful if some controls could be used in pydpf to import the data and allow numerical (but not graphical) post processing.

Pydpf is designed to optimize grabbing the data. In this case, the users have optimized earlier by setting conditions on the saveing of the data. So while it creates problems for pydpf, these are users actively interested in optimization.

Steps To Reproduce

Create RST in APDL but set /CONFIG,NORSTGM,1
or read in the attached data
Import .rst into pydpf

Which Operating System causes the issue?

Windows

Which DPF/Ansys version are you using?

Ansys 2024 R2

Which Python version causes the issue?

3.12

Installed packages

all required.

[PProfizi]

Hi @mcMunich could you give an example of the PyDPF-Core script your users tried for loading the data?
A dpf.Model will indeed try and load a mesh, but in theory you do not need one to be there for result operators to work if using a dpf.DataSources directly instead of a dpf.Model.
Basically you need to go the "operators-only" route and not use the Model helper class.
If it turns out the request is for the Model to still work without a mesh present, then we can write an enhancement request for that.

[mcMunich]

@PProfizi

This was the defaults script used.

from ansys.dpf import core as dpf
import os
result_path = r"D:\temp\apdl_nogeo_rst\BEAMER.rst"
model = dpf.Model(result_path)
print(model)

And apparently I can use DataSources to import the .rst.
Is there an example how to then process using operators with the geometry-less .rst. I just find a csv import using DataSources in the examples and the operators used are specific to csv input data.

my_data_sources = dpf.DataSources(server_file_path)
import_csv_operator = dpf.operators.serialization.csv_to_field()

[ayush-kumar-423]

@PProfizi, I think we should add an enhancement to the dpf.Model class to support RSTs without geometry data.

@mcMunich using the following code for any operator should work:

ds = dpf.DataSources(rst_path)
stress = dpf.operators.result.stress(data_sources=ds)

[PProfizi]

@ayush-kumar-423 indeed I will translate this issue as an enhancement request for the dpf.Model to support models without a mesh, yet I am not clear on the implications it would have down the line, so it may not be that easy to change.

@mcMunich concerning examples on how to use operators directly, we do have a small example here, as well as a dedicated documentation page here.

We do plan to release a new tutorials section soon, which will target this exact sort of question thanks to, among other topics, a tutorial on how to use the operators.