[mattwthompson]

From a meeting with @lilyminium we decided #204 was planned incorrectly: #203 (comment)

[codecov-commenter]

Codecov Report

❌ Patch coverage is 16.66667% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.61%. Comparing base (6f3e337) to head (0bd185f).

Files with missing lines	Patch %	Lines
tyff/compute/_produce.py	0.00%	3 Missing ⚠️
tyff/compute/_equilibrate.py	0.00%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #205   +/-   ##
=======================================
  Coverage   83.61%   83.61%           
=======================================
  Files          58       58           
  Lines        5195     5195           
=======================================
  Hits         4344     4344           
  Misses        851      851           

☔ 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]

🟡 Changes recommended

The gas-phase path this PR enables now lets pressure=None reach the unchanged TensorReporter calls (pressure=pressure * openmm.unit.kilopascal) in _equilibrate.py and _produce.py, which raise a TypeError for None, so gas-phase runs still fail.

2 open issues

• The added comment says pressure will be reported "as None to TensorReporter later", but the… · New
• The added comment says pressure will be reported "as None to TensorReporter later", but the… · New

What changed in this PR

This PR reworks pressure handling so that gas-phase NVT "simulations" no longer require a pressure value. Previously the compute config required a non-null pressure and the equilibration/production steps raised PressureNotDefinedError whenever pressure was None, which caused gas-phase jobs (where pressure is irrelevant) to fail inside the Parsl machinery (issue #203). The change makes pressure nullable and restricts the "pressure required" guard to liquid NPT runs only.

Changes:

• BaseComputeConfig.pressure is now float | None, and gas configs are created with pressure=None.
• The PressureNotDefinedError guard is moved inside the tag == "liquid" branch in both _equilibrate.py and _produce.py.
• Documentation/tests updated to assert liquid configs carry a non-null pressure and gas configs carry None.

File	Description
tyff/​configs/​_compute.py	Makes pressure nullable (`float
tyff/​configs/​gas.py	Adds a docstring noting pressure should always be None for vacuum/gas.
tyff/​compute/​prep.py	Builds gas configs with pressure=None instead of copying the data entry pressure.
tyff/​compute/​_produce.py	Moves the pressure-required guard inside the liquid branch.
tyff/​compute/​_equilibrate.py	Moves the pressure-required guard inside the liquid branch.
tyff/​_tests/​compute/​test_prep.py	Asserts liquid configs have non-null pressure and gas configs have None.

_{Review effort: Balanced}

[Copilot]

Fixed by guarding the reporter pressure before the gas-phase TensorReporter call: pressure is converted to None when absent and is only multiplied by kPa inside the liquid-only path. The liquid PressureNotDefinedError guard remains in place. Commit: 1b9a02a.