Support Pydantic v1 and v2 #752
Conversation
Codecov ReportPatch coverage:
Additional details and impacted files@@ Coverage Diff @@
## main #752 +/- ##
==========================================
+ Coverage 92.02% 92.07% +0.05%
==========================================
Files 67 67
Lines 6873 6892 +19
==========================================
+ Hits 6325 6346 +21
+ Misses 548 546 -2
☔ View full report in Codecov by Sentry. |
|
It looks like this sort of change would need to be propagated through the entire stack all at once |
|
There was some additional change introduced in pydantic v1.10.12 that failed our set up, I haven't checked, but those changes probably also causing the failing tests here. |
|
It originally launched with a broken build (conda-forge/admin-requests#770 begins the paper trail) but was fixed. There could be other issues, but the one I ran into is not what I'm seeing in these logs |
|
Yeah the issue is just forcefield-utilities. I think we can quickly release get Pydantic 2 as a PR there and release it so these tests pass. |
|
I've done the absolute minimum work to get that going: mosdef-hub/forcefield-utilities#36 |
|
Is there a path to getting tests passing? The first thing I'd try, short of a release of |
|
@daico007 merged the PR into ffutils, so we can rerun the tests here and see if the bleeding edge tests pass. |
|
Ah, nice, I forgot you all set up those tests. Should be useful! |
|
Cool, looks like that run passed: https://github.com/mosdef-hub/gmso/actions/runs/5729710869/job/15560739904 |
|
Happy to help - could you take on the responsibility of merging this when appropriate? No rush on a release but will be happy to see it out |
|
Yes, for sure, I am planning to have this merged right before releasing (just so the tests on the main CI doesn't appear half-failed), but I could merge this now if you need this for other work. |
|
Not urgent at all, this won't cause us major problems for a few months |
Co-authored-by: Co Quach <43968221+daico007@users.noreply.github.com>
* Add fixed length bonds and angles to potential templates library. * Working to make MCF writer feature complete. Tests in progress. * Add fixed angle support * Fix charge sign error (gmso-wide) * Update ring identification to handle fused systems (mbuild #744) * Increase floating point accuracy * Increase element type length and atom type length * Correctly write atom indices for bonds/angles/dihedrals/impropers * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add dimensions to Fixed Bond and Angle templates * Update keys in OPLS dihedral parameters dict * Modify dihedral constant indexing from k0 to k3 to k1 to k4 in the OPLSTorsionPotential template. * Fix typed OPLS ethane test for MCF format * Updated from old Mie Xenon test to OPLS ethane. * Remove assertions related to n and m parameters of the Mie potential. * Fix floating point format issues in MCF writer. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add typed ethane test. This test is not complete yet. The 1/2 factor in the angles and OPLS dihedral is not accounted for. * Fix code style * Add fixture to parse mcf into sections * Support Pydantic v1 and v2 (#752) Co-authored-by: Co Quach <43968221+daico007@users.noreply.github.com> * Update MCF with GMSO builtin top site sorting * Use PotentialFilters to check for unique atomtypes * Add test to check charge neutrality * Add test for incompatible expressions * Test full MCF for Mie-Xe and LJ-Ar * Check charge neutrality in each molecule test * Add exception for not neutral system in MCF writer * Move parsing and neutrality check as utils * minor docstring/comment fixes, swap out simplify with symengine expand * [pre-commit.ci] pre-commit autoupdate (#763) updates: - [github.com/psf/black: 23.7.0 → 23.9.1](psf/black@23.7.0...23.9.1) Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> * Test for 0.5 factor of OPLS dihedral potential * Test to account for 0.5 factor in harmonic angles * Test for rigid angles using the TIP3P model * MCF writer test with topology with 10 argon mols * Tentative output of multiple MCF from one Topology * Change psi to phi and consts for ethylene xml test * Test for 0.5 factor in OPLS dihedrals * Add cassandra test for gmso * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add test for two atom fragment * Test for molecule with one ring * Change nitrogen test to ethane ua * Add ethane rigid reference xml * Use Fourier dihedrals as MCF standard dihedrals. * Fix dihedral type output fourier to opls * Match MCF GMSO Angle header to mb Angle header * Add a test comparing gmso and mbuild mcf writers * Use the Fourier converter for ethylene dihedrals * Write atom type masses instead of atom masses to account for UA beads * Output correct case for dihedral styles in MCFs * Run an energy calculation to compare GMSO and mBuild MCF writers * Update gmso/tests/test_mcf.py * Update gmso/tests/test_mcf.py --------- Co-authored-by: Ryan S. DeFever <r.s.defever@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Matt Thompson <mattwthompson@protonmail.com> Co-authored-by: Co Quach <43968221+daico007@users.noreply.github.com> Co-authored-by: Co Quach <daico007@gmail.com> Co-authored-by: CalCraven <nicholas.c.craven@vanderbilt.edu>
Stopgap bridge prior to #745, see https://docs.pydantic.dev/2.0/migration/#continue-using-pydantic-v1-features