Miscellaneous bugfixes and cleanup#475
Conversation
# Conflicts: # RELEASE_NOTES.md
|
Hi @brbass - Our CI is reporting a build failure (attached snippet). Have you seen this before? |
Fixed! I think it is probably a difference of compilers since this warning didn't cause an error in my local build. |
|
We have some CI builds enforcing warnings as errors just to keep things clean, so that's probably the difference. |
| JohnsonCookDamagePolicy(): | ||
| UpdatePolicyBase<Dimension>({SolidFieldNames::flaws, | ||
| SolidFieldNames::plasticStrain}) { | ||
| UpdatePolicyBase<Dimension>() { |
There was a problem hiding this comment.
Why remove these dependencies? I believe this policy does depend on both the flaws and plastic strain -- was there a circular dependency resulting?
Also as an aside the JC damage and strength are not well tested, so be cautious using them.
|
|
||
| // Call the ancestor. | ||
| SPHBase<Dimension>::initializeProblemStartupDependencies(dataBase, state, derivs); | ||
| SPH<Dimension>::initializeProblemStartupDependencies(dataBase, state, derivs); |
There was a problem hiding this comment.
This one is a C++ question for me: the SPH class does not have an initializeProblemDependencies method, but it's ancestor (SPHBase) does. What happens by calling the SPH:: declaration here?
There was a problem hiding this comment.
If there is no SPH::declaration, then it will call SPHBase::declaration. But if there is an SPH::declaration in the future, this will call that instead. Seems like an awful bug to track down, so this is just being nice to our future selves.
|
Hi @brbass, we are hopeful to get this into our next release. I'm ready to take any changes you might have in response to the questions/comments raised in this PR. If no changes are required, please let me know that as well. Thanks! |
Responded to all! Feel free to merge whenever you guys think it is ready. |
Summary
ToDo :
RELEASE_NOTES.mdwith notable changes.