[steven-murray]

Summary

Makes the mass-definition, concentration, conversion and profile modules of hmf.core call the same way as MassFunction and LinearPower: keyword-only arguments, one name per operation, one calibration-domain API. These modules are unreleased and experimental, so there are no deprecation aliases. Results are bit-identical.

Renames and signature changes (old → new)

hmf.core.mass_definitions.MassDefinition

Old	New
m_to_r(m, *, omega_m_z, rho_mean0)	radius_from_m(*, m, omega_m_z, rho_mean0)
r_to_m(r, *, omega_m_z, rho_mean0)	m_from_radius(*, r, omega_m_z, rho_mean0)
m_to_r_kernel(m, *, omega_m_z, rho_mean0)	radius_from_m_kernel(*, m, omega_m_z, rho_mean0)
r_to_m_kernel(r, *, omega_m_z, rho_mean0)	m_from_radius_kernel(*, r, omega_m_z, rho_mean0)
delta_halo_mean(omega_m_z)	delta_halo_mean(*, omega_m_z)
delta_halo_mean_kernel(omega_m_z)	delta_halo_mean_kernel(*, omega_m_z)

hmf.core.concentration.ConcentrationModel

Old	New
concentration(m, *, z)	concentration(*, m, z)
ln_concentration_and_slope_kernel(m, *, z)	ln_concentration_and_slope_kernel(*, m, z)
in_calibration_domain_kernel(m, *, z)	in_calibration_domain_kernel(*, m, z)

hmf.core.mass_conversion.MassConversion

Old	New
convert(m, *, z, omega_m_z, source, target)	convert(*, m, z, omega_m_z, source, target)
convert_kernel(m, *, z, omega_m_z, source, target)	convert_kernel(*, m, z, omega_m_z, source, target)

hmf.core.profiles.DensityProfile

Old	New
mu(x)	mu(*, x)
ln_mu_and_slope_kernel(x)	ln_mu_and_slope_kernel(*, x)
convert_concentration_kernel(c, *, slope, ln_q)	convert_concentration_kernel(*, c, slope, ln_q)

hmf.core.fits.FittingFunction

Old	New
in_calibration_domain(x)	in_calibration_domain_kernel(x). It now checks x against valid_domain (raising DomainError), like every other *_kernel. evaluate_fsigma doesn't repeat that check, because fsigma_kernel has already done it.

x stays positional here, to match fsigma_kernel(x): it's a FitInputs bundle, not an array.

The calibration-domain API

hmf.core.domain.Calibrated is a mixin with empty __slots__. Every model kind with calibration ranges now uses it: FittingFunction, ConcentrationModel, DensityProfile, Filter, and TransferModel / GrowthModel (through _CosmologyModel).

• model.calibration (instance property) is the API. It gives where this instance was calibrated, or None if it wasn't.
• cls.calibration_domain (class variable) is the calibration of the class's default parameters, and calibration returns it unless the model overrides it.
• Duffy08.calibration gives z = 0 for redshifts="0", while Duffy08.calibration_domain stays the z = 0–2 default.
• FittingFunction.in_calibration_domain_kernel, evaluate_fsigma, ConcentrationModel.in_calibration_domain_kernel and ProfileConversion all read calibration. A fit whose calibration depends on its parameters is masked by its own range (tested).
• The concentration, fits and domain module docs, docs/core.rst (one sentence in "Domains") and the model-class doc template now point to calibration.

Removed kernels

hmf.core._kernels.mass_conversion: nfw_ln_mu, nfw_ln_mu_slope, nfw_concentration_ratio_solve, nfw_convert. The library didn't use them. Their physical tests (closed form and series, ψ against a numerical derivative, round-off residual of the solve over 1e-3 < c < 1e4 and 1e-4 < q < 1e4, batch independence, the exact identity at q = 1, agreement with the v3 brentq solve, the Hu & Kravtsov accuracy, the Jacobian against a finite difference) now run through NFW().ln_mu_and_slope_kernel and NFW().convert_concentration_kernel. nfw_ln_mean_density and hu_kravtsov_inverse stay, because nfw_start uses them.

SOGeneric names round-trip

• MassDefinition.from_name parses "so(preferred <name>)".
• It raises ValueError when the preferred definition isn't an SOMean, SOCritical or SOVirial.
• Names now write a number in full when :g would round it. Before, SOMean(18 pi^2).name was "177.653m", which read back as a different definition. Names that were already exact ("200m", "500c", "fof(b=0.2)", …) don't change.
• A parametrised test round-trips every definition type, including those numbers.

Call sites

• mass_function.py: four one-line changes (m=m, omega_m_z=omega_m_z, and two in_calibration_domain_kernel calls).
• tests/regression/v4_providers.py: one keyword.

Checks

• uv run pytest tests/core tests/regression: 2319 passed, 11 skipped. Every touched hmf.core file is at 100% coverage. The global 95% gate counts v3 code, so it needs the full suite.
• uv run pytest benchmarks/test_gates.py: 25 passed.
• uv run mypy: clean.
• uv run prek run --files …: clean.
• sphinx-build -W --keep-going: succeeded.

Type of change

• Bug fix
• New feature
• Documentation
• Maintenance (refactoring, CI, dependencies, etc.)

Checklist

I have

• Added or updated tests covering this change.
• Updated docstrings and/or docs where relevant.

---

Generated by Claude Code

Summary by Sourcery

Standardize experimental core model interfaces and calibration-domain handling while consolidating profile kernels and preserving existing numerical results.

New Features:

• Add round-trip parsing and formatting for SOGeneric and spherical-overdensity mass-definition names.

Bug Fixes:

• Ensure mass-definition names preserve enough numeric precision to reconstruct equivalent definitions.

Enhancements:

• Standardize core model APIs around keyword-only arguments and consistent operation names.
• Introduce a shared instance-level calibration-domain API across calibrated model kinds.
• Route NFW profile calculations and conversion tests through DensityProfile kernels.

Documentation:

• Update core and model documentation to describe instance calibration domains and the standardized APIs.

Tests:

• Expand core tests for keyword-only interfaces, instance-dependent calibration domains, mass-definition name round trips, and profile-kernel behavior.

Chores:

• Remove unused mass-conversion kernels and their direct test usage while preserving bit-identical results.

[sourcery-ai]

Sorry @steven-murray, you've used your own review budget of 250,000 diff characters for the last 7 days.

You can request another review in 4 days and 5 hours by commenting @sourcery-ai review. Upgrade to get a review now.

[sourcery-ai]

Reviewer's Guide

Data structures

Introduce the Calibrated mixin as the shared instance-level calibration interface, defaulting to each model kind’s class calibration domain.

• src/hmf/core/_cosmology_models.py L6-7, L22-25, L41 (removed lines)
• src/hmf/core/domain.py L3-18, L413-436 (+2 small edits)

Make concentration, fits, filters and density profiles calibrated model kinds, and route instance calibration checks and fit evaluation through calibration.

• src/hmf/core/concentration.py L13-256 (10 small edits)
• src/hmf/core/filters.py L30, L68
• src/hmf/core/fits.py L87-92, L711-714, L731-735 (+5 small edits)
• src/hmf/core/profiles.py L41, L53

Make mass-definition names round-trip precisely, including parsing valid SOGeneric preferred definitions.

• src/hmf/core/mass_definitions.py L155-158, L176-184, L417-422 (+5 small edits)

API

Standardize public and kernel call signatures on keyword-only operation inputs, and rename mass-radius operations for clarity.

• src/hmf/core/mass_conversion.py L153, L189
• src/hmf/core/mass_definitions.py L266-270, L389-393 (+8 small edits)
• src/hmf/core/profiles.py L79-160 (4 small edits)

Rename the fit calibration check to its kernel API, validate inputs against the valid domain, and remove unused low-level NFW conversion entry points.

• src/hmf/core/_kernels/mass_conversion.py L49-295 (4 small edits)

Uses of the API

Update internal consumers to pass renamed operation arguments by keyword and to use the profile conversion API and instance calibration checks.

• src/hmf/core/mass_conversion.py L284-287, L292-300, L355-359 (+3 small edits)
• src/hmf/core/mass_function.py L525, L649, L972-976

Update the regression provider to call the mass-definition kernel with its keyword-only argument.

• tests/regression/v4_providers.py L166

Implementation

Implement the shared NFW profile calculations through combined log-mass-and-slope evaluation and adjust their kernel documentation and exports.

• src/hmf/core/_kernels/mass_conversion.py L79-83 (+5 small edits)

Documentation

Revise core and model documentation to describe instance calibration and the renamed mass-radius and profile conversion APIs.

• docs/core.rst L131-135
• docs/templates/core-model-class.rst L11
• src/hmf/core/mass_definitions.py L31

Tests

Exercise calibration inheritance and instance-specific behavior, keyword-only APIs, name round-trips, and profile-backed conversion behavior.

• tests/core/test_core_concentration.py L62-69, L85-92, L102-126 (+4 small edits)
• tests/core/test_core_domain.py L423-489
• tests/core/test_core_fits.py L872, L1021
• tests/core/test_core_kernel_safety.py L397-402, L418-443 (+4 small edits)
• tests/core/test_core_mass_conversion.py L49-74, L80-89, L274-282 (+25 small edits)
• tests/core/test_core_mass_definitions.py L92-107, L119-130, L278-301 (+10 small edits)
• tests/core/test_core_profiles.py L49-79, L107-115 (+5 small edits)

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.70%. Comparing base (a76c553) to head (26052ee).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #445      +/-   ##
==========================================
- Coverage   99.70%   99.70%   -0.01%     
==========================================
  Files          69       69              
  Lines        9537     9536       -1     
==========================================
- Hits         9509     9508       -1     
  Misses         28       28              

☔ 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.