Bump pyscal3 from 3.3.2 to 4.1.0

[dependabot]

Bumps pyscal3 from 3.3.2 to 4.0.0.

Release notes

Sourced from pyscal3's releases.

4.0.0

What's Changed

• Feat/wl parameters by @​srmnitc in pyscal/pyscal3#63
• Feat/ace by @​srmnitc in pyscal/pyscal3#62
• feat: add Minkowski structure metrics (Voronoi-weighted q_l) by @​srmnitc in pyscal/pyscal3#64
• feat: add Ackland-Jones structure classifier by @​srmnitc in pyscal/pyscal3#65
• feat: add coordination number variants (CN, ECoN, GCN, local density) by @​srmnitc in pyscal/pyscal3#66
• Feat/adf by @​srmnitc in pyscal/pyscal3#67
• Feat/deformation by @​srmnitc in pyscal/pyscal3#68
• Feat/wigner seitz by @​srmnitc in pyscal/pyscal3#69
• Refactor/ase first by @​srmnitc in pyscal/pyscal3#61
• add more examples by @​srmnitc in pyscal/pyscal3#70
• Fixes by @​srmnitc in pyscal/pyscal3#71

Full Changelog: pyscal/pyscal3@v3.3.2...4.0.0

Changelog

Sourced from pyscal3's changelog.

4.0.0 (unreleased)

A new API

pyscal 4 is a rewrite around two ideas: ASE Atoms is the data structure, and every descriptor is a top-level function. The System / Atoms classes of pyscal 3 are gone. Neighbor lists are computed once with find_neighbors(atoms, ...) and reused by all descriptors; per-atom results are written to atoms.arrays["pyscal_*"] (uniform shapes) or atoms.info["pyscal_*"] (ragged data and arrays with more than two dimensions). import pyscal and import pyscal3 are synonyms.

New descriptors: Wigner W_l, Minkowski structure metrics, Ackland-Jones classification, effective and generalized coordination numbers, local density, angular and bond-length distribution functions, atomic strain, von Mises strain, D^2_min, slip vector, Wigner-Seitz defect analysis, and ACE descriptors up to body order four.

Behaviour changes relative to pyscal 3.x

• radial_distribution_function now returns a properly normalised g(r) (it tends to 1 for an ideal gas and its integral counts neighbors); the previous values were scaled by an arbitrary constant.
• short_range_order computes the Warren-Cowley parameter alpha_AB = 1 - p_AB / c_B for atoms of the reference species and averages over those atoms; species may be given as symbols or atomic numbers and default to the two most abundant species. The old mean over all atoms was identically zero.
• SANN neighbor lists now contain the correct m nearest atoms (an off-by-one skipped the 4th nearest neighbor).
• The disorder parameter compares each atom with its actual neighbors; values on disordered structures differ from 3.x.
• find_solids honours the bonds criterion (a dangling else had made it inert).
• W_l values are now rotation invariant (a missing (-1)^m phase made them depend on the crystal orientation).
• entropy(local=True) uses each atom's own local density, the trapezoidal integration includes all grid points, and a warning is issued when rm exceeds the neighbor cutoff.
• make_crystal without element assigns the lattice type as atomic number (type 1 -> Z=1, ...) instead of Z=0, so sublattices remain distinguishable.
• Neighbor vectors (pyscal_diff) are stored in atoms.info, so ase.io.write keeps working after a pyscal calculation.

Fixes

• Cell lists are built on fractional coordinates: triclinic, hexagonal and rotated cells (primitive fcc, hcp, ASE fcc111 slabs, LAMMPS triclinic boxes) gave wrong neighbors or hung above 250 atoms.

... (truncated)

Commits

• 4329f04 Merge pull request #71 from pyscal/fixes
• 0b707cb trajectory: write dump files with LF newlines on every platform
• c703b54 tests: read source files as UTF-8 in the bindings test
• e2f4672 ci: move to Node 24 action majors, macOS 15 runner, least privilege
• 24804c0 docs: document stored s_ij arrays and how to read structure files
• 4b4ad0f docs: add changelog for 4.0.0
• 4220a63 docs: add method pages for common neighbor analysis and short-range order
• 986ca49 build: single source for the version number
• 8df8b55 fix: Wigner-Seitz analysis for atoms outside the reference cell
• 301a3c2 fix: merge Voronoi vertices into unique sites over all atoms
• Additional commits viewable in compare view

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

• Configuration used: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: d7ad2b5e-c096-42ae-b13f-1c7a1e34c7b5

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[codecov]

Codecov Report

❌ Patch coverage is 65.90909% with 15 lines in your changes missing coverage. Please review.
✅ Project coverage is 97.44%. Comparing base (afe2af5) to head (2ff73ac).
⚠️ Report is 6 commits behind head on main.

Files with missing lines	Patch %	Lines
src/structuretoolkit/analyse/pyscal.py	65.85%	14 Missing ⚠️
src/structuretoolkit/common/pyscal.py	66.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #551      +/-   ##
==========================================
- Coverage   97.90%   97.44%   -0.47%     
==========================================
  Files          29       29              
  Lines        2004     2033      +29     
==========================================
+ Hits         1962     1981      +19     
- Misses         42       52      +10     

☔ 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.