docs/concepts/articulations.rst (1)

14-18: Clarify fixed-base vs floating-base in the first example sentence.

As written, “two generalized coordinates … corresponding to the two joint angles and velocities” mixes positions and velocities and implicitly assumes a fixed base. Suggest being explicit and separating positions from velocities.

- Generalized (sometimes also called "reduced") coordinates describe the configuration of an articulation in terms of joint positions and velocities.
- That means, if an articulation has three bodies connected by two revolute joints, the articulation has two generalized coordinates corresponding to the two joint angles (:attr:`newton.State.joint_q`) and joint velocities (:attr:`newton.State.joint_qd`).
+ Generalized (sometimes also called "reduced") coordinates describe the configuration of an articulation in terms of joint positions (:attr:`newton.State.joint_q`) and joint velocities (:attr:`newton.State.joint_qd`).
+ For a fixed‑base articulation with three bodies connected by two revolute joints, there are two positional generalized coordinates (the two joint angles) and two velocity coordinates (the two joint angular velocities).

docs/concepts/articulations.rst (4)

32-48: Doctest robustness: add imports and tolerant float checks.

These doctests rely on newton, wp, and exact float equality. Make the block self‑contained and use np.isclose/allclose.

 .. testcode::
 
-  builder = newton.ModelBuilder()
+  import numpy as np
+  import warp as wp
+  import newton
+
+  builder = newton.ModelBuilder()
   builder.add_articulation()
   body = builder.add_body()
   builder.add_shape_box(body)  # add a shape to the body to add some inertia
   builder.add_joint_revolute(parent=-1, child=body, axis=wp.vec3(0.0, 0.0, 1.0))  # add a revolute joint to the body
   builder.joint_q[-1] = 0.5
   builder.joint_qd[-1] = 10.0
 
   model = builder.finalize()
   state = model.state()
 
   # The generalized coordinates have been initialized by the revolute joint:
-  assert all(state.joint_q.numpy() == [0.5])
-  assert all(state.joint_qd.numpy() == [10.0])
+  assert np.allclose(state.joint_q.numpy(), [0.5])
+  assert np.allclose(state.joint_qd.numpy(), [10.0])

---

101-107: Use tolerant comparison for joint_q vs transform.

Same flattening/equality concerns as above; compare pos/quat with tolerance.

-  assert len(state.joint_q) == 7
-  assert all(state.joint_q.numpy() == [*tf])
+  import numpy as np
+  assert len(state.joint_q) == 7
+  pos = wp.transform_get_translation(tf); quat = wp.transform_get_rotation(tf)
+  assert np.allclose(state.joint_q.numpy(), [pos[0], pos[1], pos[2], quat[0], quat[1], quat[2], quat[3]])

---

167-178: Minor consistency: qualify Model/State in examples or add a preface.

Elsewhere you use fully qualified names. Consider either from newton import Model, State in a setup block or qualify here for consistency.

---

83-98: Remove invalid slice assignment in free joint example.

• add_joint_free(body) already binds to the world by default (parent=-1), so no change needed.
• The commented

  # builder.joint_q[-7:] = *tf

  is invalid (and references non-existent helpers). Either remove it or replace it with a valid example using the actual API for extracting the translation and quaternion from the tf transform.

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro

newton/solvers.py (1)

16-18: LGTM! Consider optionally expanding the docstring.

The module docstring is clear and accurate. Since this module re-exports multiple solver types, you might optionally enhance it to briefly mention the available solvers or link to the documentation.

For example:

 """
-Solvers are used to integrate the dynamics of a Newton model.
+Solvers are used to integrate the dynamics of a Newton model.
+
+Available solvers include Featherstone, Semi-Implicit, XPBD, VBD, and others.
+See the documentation for details on each solver type and usage examples.
 """

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro