docs: clarify solver integration docs

[fanyahao1]

Description

Close #3015 without changing runtime behavior.

Fix two solver integration documentation inconsistencies.

• Update the solver overview table so SolverFeatherstone is documented as semi-implicit, matching its class docstring and symplectic-Euler integration kernel.
• Clarify the SolverMuJoCo overview entry as Explicit, Semi-implicit, Implicit-in-velocity, matching the MuJoCo integrator categories exposed by Newton.

Checklist

• New or existing tests cover these changes
• The documentation is up to date with these changes
• CHANGELOG.md has been updated (if user-facing change)

Test plan

• Manually compared the updated documentation against:
  • newton/_src/solvers/featherstone/solver_featherstone.py
  • newton/_src/solvers/featherstone/kernels.py
  • newton/_src/solvers/mujoco/solver_mujoco.py
• Verified that the diff is limited to the intended documentation updates.

Bug fix

Steps to reproduce:

1. Open docs/api/newton_solvers.rst.
2. Compare the SolverFeatherstone integration label against the SolverFeatherstone class docstring and integrate_generalized_joints implementation.
   Minimal reproduction:

# Documentation-only issue; no runtime reproduction.

Summary by CodeRabbit

• Documentation
  • Corrected the supported integration methods for SolverFeatherstone and SolverMuJoCo in the solver documentation to accurately reflect their capabilities.

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: fanyahao1 / name: fanyahao1 (970cf6c)

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro

Run ID: 38ab4f84-577b-43aa-b6ce-f161a9f30c19

📥 Commits

Reviewing files that changed from the base of the PR and between 970cf6c and f1adc15.

📒 Files selected for processing (1)

• newton/solvers.py

✅ Files skipped from review due to trivial changes (1)

• newton/solvers.py

---

📝 Walkthrough

Walkthrough

Documentation tables describing solver integration method support are corrected in both docs/api/newton_solvers.rst and newton/solvers.py. SolverFeatherstone integration method changed from "Explicit" to "Semi-implicit", and SolverMuJoCo changed from "Implicit" to "Implicit-in-velocity" to align with actual implementations.

Changes

Solver Integration Classification Documentation

Layer / File(s)	Summary
Solver integration method documentation corrections docs/api/newton_solvers.rst, newton/solvers.py	Updated the "Supported Features" table entries for SolverFeatherstone and SolverMuJoCo in both the RST API documentation and Python module docstring to reflect correct integration method classifications: SolverFeatherstone now lists "Semi-implicit" instead of "Explicit", and SolverMuJoCo now lists "Implicit-in-velocity" instead of "Implicit".

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• newton-physics/newton#2889: Both PRs update docs/api/newton_solvers.rst's "Supported Features" matrix with solver capability corrections.

Suggested labels

docs

Suggested reviewers

• adenzler-nvidia
• eric-heiden
• preist-nvidia

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Linked Issues check	⚠️ Warning	The PR partially addresses issue #3015 requirements: it corrects SolverFeatherstone to 'Semi-implicit' and updates SolverMuJoCo documentation, but omits the suggested docstring update for the missing 'implicit' integrator option.	Add the missing 'implicit' integrator option to the SolverMuJoCo constructor docstring in newton/_src/solvers/mujoco/solver_mujoco.py to fully resolve issue #3015.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The PR title 'docs: clarify solver integration docs' directly and specifically describes the main change: updating documentation to clarify solver integration classifications.
Out of Scope Changes check	✅ Passed	All changes are within scope: the PR updates solver documentation entries and table clarifications directly addressing issue #3015 without introducing unrelated modifications.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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 and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@newton/_src/solvers/mujoco/solver_mujoco.py`:
- Line 3135: Add a CHANGELOG entry under the [Unreleased] section (preferably in
the "Fixed" subsection) noting the user-facing documentation correction to
supported integrator values for solver_mujoco.py; reference the integrator
parameter documentation change (integrator: Integrator type... Can be "euler",
"rk4", "implicit", or "implicitfast", or their corresponding MuJoCo integer
constants) and include a short line like "Fix: document supported integrator
values for solver_mujoco.integrator" so the docs change is tracked.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro

Run ID: f6e621dc-396c-4613-bda9-8b5143fe3783

📥 Commits

Reviewing files that changed from the base of the PR and between baa3d3a and 970cf6c.

📒 Files selected for processing (2)

• docs/api/newton_solvers.rst
• newton/_src/solvers/mujoco/solver_mujoco.py

[adenzler-nvidia]

Thanks for the contribution — these solver docs have been inconsistent for a while and it's great to see them cleaned up.

Two of the three changes are spot on:

• SolverFeatherstone → "Semi-implicit" matches the class docstring ("a semi-implicit integrator using symplectic Euler"). ✅
• SolverMuJoCo → "Explicit, Semi-implicit, Implicit-in-velocity" is more accurate than the old "Implicit" — the default mujoco_warp backend's implicitfast is exactly "implicit in velocity." ✅

One request before merge (see inline comment): the constructor-docstring edit adds "implicit" as an integrator option, which pulls in the opposite direction from the table change. Could we revert that one line? With that, this is good to go.

[fanyahao1]

One request before merge (see inline comment): the constructor-docstring edit adds "implicit" as an integrator option, which pulls in the opposite direction from the table change. Could we revert that one line? With that, this is good to go.

You're right — that constructor docstring line should stay aligned with the table and the default mujoco_warp backend. I've reverted it. Thank you!

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ All tests successful. No failed tests found.

📢 Thoughts on this report? Let us know!