[mattwthompson]

Description

Closes #206

Todos

• Test with Pint 0.26
• Update dependencies
• Update code snippets in docs
• Run doctest on docs

AI disclosure

• Agentic tools or LLMs were used in this contribution - don't recall using any in writing, will be used in review

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.37%. Comparing base (49dd75d) to head (8978a71).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #207   +/-   ##
=======================================
  Coverage   91.37%   91.37%           
=======================================
  Files           6        6           
  Lines         232      232           
=======================================
  Hits          212      212           
  Misses         20       20           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Note

Copilot was unable to complete this review at the selected tier due to model policy or plan restrictions.

🟡 Changes recommended

Approval is blocked because the review evidence identifies unresolved compatibility and maintenance issues, including incomplete Pint-version coverage in CI and dependency range mismatches (unconstrained Pint in pixi.toml) alongside brittle doctests tied to exact repr output.

3 open issues

• CI misses supported Pint 0.25 lower-bound coverage · New
• Unconstrained Pint can drift past supported version range · New
• Doctest should avoid exact Quantity repr assertions · New

What changed in this PR

Updates dependency metadata, Pixi environments/lockfile, CI workflows, and doctest expectations to validate compatibility with Pint 0.26 (Closes #206).

Changes:

• Broaden Pint support in package metadata to include 0.26 (pint>=0.25,<0.27) and refresh Pixi lock data.
• Update Pixi environment definitions and CI matrices to run Pint 0.26-focused jobs.
• Adjust OpenMM-related doctest expected outputs to match Pint’s updated display behavior.

Key review notes to address before approval:

• CI coverage: restore at least one non-OpenMM Pixi environment run in .github/workflows/ci.yaml (OpenMM is optional).
• Lower-bound testing: either add Pint 0.25 to .github/workflows/ci_pypi.yaml or bump the declared lower bound if 0.25 is no longer supported.
• Doctest robustness: avoid asserting exact repr() for Quantity in openff/units/openmm.py (noted at lines ~165 and ~299); prefer explicit magnitude/unit assertions.
• Pixi dependency alignment: constrain pint in pixi.toml to the supported range (currently unconstrained as "*").

File	Summary
pyproject.toml	Updates declared Pint compatibility range to include 0.26 (>=0.25,<0.27).
pixi.toml	Updates Pixi environments/features for Pint 0.26; needs Pint version constraint aligned with pyproject.toml (currently "*").
pixi.lock	Regenerates lockfile to resolve against Pint 0.26.x and related dependency updates.
openff/​units/​openmm.py	Updates doctest expected outputs; needs less brittle doctests than exact repr() matching (noted at ~165 and ~299).
.github/​workflows/​ci.yaml	Updates Pixi CI matrix to Pint 0.26 OpenMM environments; needs at least one non-OpenMM Pixi job retained.
.github/​workflows/​ci_pypi.yaml	Updates PyPI-install CI to Pint 0.26; needs Pint 0.25 coverage or a lower-bound bump to 0.26.

This pull request requires changes.

_{Review effort: Balanced}

[j-wags]

Thanks @mattwthompson - This looks good. I left one blocking and two nonblocking comments.

(not blocking) My claude review points out:

  3. The reprs in docs/index.md are now out of date. The PR updates the two doctests in openmm.py to
  Pint 0.26's new repr, Quantity(9.0, "angstrom"). But docs/index.md still shows <Quantity(...)> at
  lines 38, 95, 111, 120 and 153, and the docs environment now builds with pint26. Users on 0.26 will
  see output that doesn't match the docs.

[j-wags]

(not blocking) Just checking whether this section is needed - it seems to be a verbatim copy of the pint dep above

[mattwthompson]

I think it is, because the conda packages have a different pin

I will confirm this before merging