-
Notifications
You must be signed in to change notification settings - Fork 503
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix for UWUW Macro Conflict #3150
Conversation
Looks like a file is not being found
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for this update @ahnaf-tahmid-chowdhury! My only request is that we change the new precompiler value from DAGMC_UWUW
to OPENMC_UWUW
, using the OPENMC_
prefix as something of a namespace for these precompiler values. We tend to do this with the CAPI functions as well.
However, some tests are failing with the following error: ERROR: Material with name/ID 'fuel' not found for volume (cell) 1 It seems that in the recent OpenMC version, all materials must include all the same properties, can't keep anything empty. We need to update the dagmc.h5m file, but I’m not sure how to do that. |
I might be wrong here but that error looks like the openmc materials provided don't include a material with the name 'fuel'. I don't know why this passed previously but I guess checking the model.materials contain a material with the name fuel would help check and then perhaps the material name can be changed to fuel if it does not exist. |
Unfortunately, it was turned off when this was created. I have recently turned it on and found that the related tests are now failing, as it was not updated over time. We should consider turning on all features in our CI to ensure everything is working properly. |
I looked into this. I think the |
Looks like that test is passing now. I'll take a look at the changes again shortly. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
A couple of tweaks for default behavior and one correction to the config file. Ideally we'd our OpenMCCMake.config
with some small application that builds on top of the OpenMC installation. @ahnaf-tahmid-chowdhury would you mind making an issue for that task?
Co-authored-by: Patrick Shriwise <[email protected]>
Co-authored-by: Patrick Shriwise <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One small issue with a removed const
on a DAGMC method, but then I think this looks good.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for this update @ahnaf-tahmid-chowdhury 👍🏻
Description:
This PR resolves the following issues in the OpenMC DAGMC interface:
Macro Conflict Fix:
UWUW
macro was causing conflicts during compilation as it was treated as a numeric constant.UWUW
has been replaced withUWUW_HPP
to avoid naming conflicts with numeric constants.Changes:
In
CMakeLists.txt
, the line:has been updated to:
All relevant files that referenced
UWUW
have been updated accordingly.Const Qualifier Removal:
const
qualifier was removed from the functionuwuw_assign_material
in both the declaration (indagmc.h
) and the definition (indagmc.cpp
).