search

mosdef-hub / mbuild

A hierarchical, component based molecule builder
https://mbuild.mosdef.org
Other
171 stars 79 forks source link

Add warnings for deprecated file formats (protobuf, hoomdxml, Hoomd-Blue and LAMMPS simulation writers) #1188

Closed chrisjonesBSU closed 2 months ago

chrisjonesBSU commented 3 months ago

PR Summary:

Per the last few dev meetings, we've been discussing what needs to be done before releasing 1.0. We decided to clean up and consolidate conversion.py which involves moving the backend to GMSO wherever possible, and removing .hoomdxml and .pb2 readers and writers. Also, we decided to remove any mbuild.format writers that are now duplicated in GMSO such as create_hoomd_simulation, create_hoomd_forcefield, and write_lammpsdata.

We decided that the first step was including the necessary deprecation warnings in a minor release, which this PR gets started. I think this covers everything that will be removed from mBuild, and any other changes to readers and writers will just involve changing the backend and won't require a deprecation warning. Let me know if I'm missing anything, or if we need to expand in the warning messages.

I can start making the next round of changes in a separate PR.

CalCraven commented 2 months ago

Ran this locally to test the failing tests, which are all related to packmol. They pass with versions: packmol 20.010 h508aa58_0 conda-forge

For the CI, looks like we're installing packmol[version='>=20'], and get packmol 20.14.4 h4c1867c_0 conda-forge

So that may be the source of the issues, external to the PR.

CalCraven commented 2 months ago

Here's the error file in the log that packmol spits out.

STOP: Maximum number of GENCAN loops achieved.

--------------------------------------------------------------------------------

 Packmol was not able to find a solution to your
 packing problem with the desired distance tolerance.

 First of all, be sure if the molecules fit in the
 regions specified and if the constraints were set
 correctly. 

 Secondly, try simply running it again with a different 
 seed for the random number generator of the initial 
 point. This is done by adding the keyword seed to the
 input file, as in:
chrisjonesBSU commented 2 months ago

The messages were updated with the full path to the relevant GMSO writers.