Fix MDI forces structure #401
Conversation
There was a problem hiding this comment.
Hi Taylor, everything looks good except for the GAMESS changes. The if "system__mwords" breaks one of the memory setting tests.
qcengine/programs/tests/test_canonical_config.py::test_local_options_memory_gib[gamess-model2-keywords2-none] PASSED [ 20%]
qcengine/programs/tests/test_canonical_config.py::test_local_options_memory_gib[gamess-model2-keywords2-dsl] FAILED [ 40%]
The ValueError addition is a little strange in that it's a Python error and not one of the qcng errors. Distributed error handling is convoluted (to me, at least), and with none of the other harnesses using this pattern plus MDI being in a class of its own in terms of how it runs things, I'm not sure how the ValueError fits in. Could you describe the circumstances in which these changes came about to help me understand them? Thanks!
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| natoms : int, optional | ||
| New number of atoms. If None, receive through MDI. | ||
| """ | ||
| natom = len(self.molecule.geometry) |
Check notice
Code scanning / CodeQL
Unused local variable Note
Description
This is a simple fix to ensure that the MDI interface properly formats the forces prior to communication.
Changelog description
Fix bug in the shape of the MDI forces structure
Status