openscvx 0.5.3.dev57__tar.gz → 0.5.3.dev58__tar.gz

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/.github/workflows/_docs.yml

@@ -65,7 +65,7 @@ jobs:
         uv pip install --system mike
 
     - name: Install package
-      run: uv pip install --system -e .
+      run: uv pip install --system -e ".[plotting]"
 
     - name: Build documentation (check only)
       if: ${{ !inputs.version }}

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/.github/workflows/tests-examples.yml

@@ -39,7 +39,9 @@ jobs:
         # cvxpygen omitted: its extra is unsatisfiable while cvxpy is pinned <1.9
         # (cvxpygen>=1.0.0 needs cvxpy>=1.9, #551); its tests skip until that resolves.
         # qpax omitted until its solves are stable on Linux runners (#550).
-        uv pip install --system -e ".[test,mjx,lie,frax,stl]"
+        # moreau omitted: examples that need it carry the moreau marker and skip
+        # here. Solver-level moreau tests run in tests-moreau.yml.
+        uv pip install --system -e ".[test,plotting,mjx,lie,frax,stl]"
 
     - name: Run examples sweep
       run: |

openscvx-0.5.3.dev58/.github/workflows/tests-moreau.yml

@@ -0,0 +1,20 @@
+name: moreau
+
+# Badge workflow: health of the moreau integration on main.
+# The weekly cron catches breakage from upstream releases between commits.
+# Kept off the main tests.yml extras job so a moreau release cannot fail
+# the mjx/lie checks.
+
+on:
+  push:
+    branches:
+      - main
+  schedule:
+    - cron: '0 6 * * 1'  # Weekly on Monday, offset from the Sunday examples sweep
+  workflow_dispatch:
+
+jobs:
+  moreau:
+    uses: ./.github/workflows/_extra.yml
+    with:
+      extra: moreau

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/.github/workflows/tests.yml

@@ -38,7 +38,7 @@ jobs:
 
     - name: Install dependencies
       run: |
-        uv pip install --system -e ".[test]"
+        uv pip install --system -e ".[test,plotting]"
 
     - name: Run unit tests
       run: |
@@ -82,7 +82,7 @@ jobs:
 
     - name: Install dependencies
       run: |
-        uv pip install --system -e ".[test]"
+        uv pip install --system -e ".[test,plotting]"
 
     - name: Run e2e tests
       run: |
@@ -123,6 +123,8 @@ jobs:
         # cvxpygen omitted: its extra is unsatisfiable while cvxpy is pinned <1.9
         # (cvxpygen>=1.0.0 needs cvxpy>=1.9, #551); its tests skip until that resolves.
         # qpax omitted until its solves are stable on Linux runners (#550).
+        # moreau omitted: tests-moreau.yml installs it and runs `-m moreau`, so an
+        # upstream moreau break stays on that badge.
         uv pip install --system -e ".[test,mjx,lie]"
 
     - name: Run extras tests

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/CONTRIBUTING.md

@@ -16,10 +16,11 @@ source venv/bin/activate  # On Windows: venv\Scripts\activate
 Then install the package in editable mode with development dependencies:
 
 ```bash
-pip install -e ".[test]"
+pip install -e ".[test,plotting]"
 ```
 
 For additional features, you can install optional dependencies:
+- `pip install -e ".[plotting]"` - Plotly, Matplotlib, and Viser visualization
 - `pip install -e ".[cvxpygen]"` - Code generation support
 - `pip install -e ".[stl]"` - STL (Signal Temporal Logic) constraints
 
@@ -32,7 +33,7 @@ git clone git@github.com:OpenSCvx/OpenSCvx.git
 cd OpenSCvx
 uv venv
 source .venv/bin/activate  # On Windows: .venv\Scripts\activate
-uv pip install -e ".[test]"
+uv pip install -e ".[test,plotting]"
 ```
 
 To contribute changes, you'll need to create a branch or fork. See [Forking and Merging](#forking-and-merging) below for details on our workflow and [Branch Naming Conventions](#branch-naming-conventions) for required naming patterns.

{openscvx-0.5.3.dev57/openscvx.egg-info → openscvx-0.5.3.dev58}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: openscvx
-Version: 0.5.3.dev57
+Version: 0.5.3.dev58
 Summary: A general Python-based successive convexification implementation which uses a JAX backend.
 Author-email: Chris Hayner and Griffin Norris <haynec@uw.edu>
 License: Apache Software License
@@ -14,13 +14,9 @@ Requires-Dist: cvxpy>=1.8.1
 Requires-Dist: qoco
 Requires-Dist: numpy
 Requires-Dist: jax
-Requires-Dist: plotly
 Requires-Dist: termcolor
 Requires-Dist: diffrax
-Requires-Dist: absl-py
 Requires-Dist: flatbuffers
-Requires-Dist: viser
-Requires-Dist: matplotlib
 Requires-Dist: pyyaml
 Requires-Dist: pydantic>=2.0
 Provides-Extra: cvxpygen
@@ -39,6 +35,10 @@ Requires-Dist: mujoco>=3.0; extra == "mjx"
 Requires-Dist: mujoco-mjx>=3.0; extra == "mjx"
 Provides-Extra: frax
 Requires-Dist: frax>=0.0.4; extra == "frax"
+Provides-Extra: plotting
+Requires-Dist: matplotlib; extra == "plotting"
+Requires-Dist: plotly; extra == "plotting"
+Requires-Dist: viser; extra == "plotting"
 Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: pytest-xdist; extra == "test"
@@ -156,9 +156,10 @@ uv pip install openscvx
 > [!TIP]
 > **Optional dependencies**
 >
-> For CVXPYGen code generation:
+> Plotting (Plotly, Matplotlib, and Viser) and CVXPYGen code generation:
 > ```bash
-> pip install openscvx[cvxpygen]
+> pip install "openscvx[plotting]"
+> pip install "openscvx[cvxpygen]"
 > ```
 
 > [!TIP]
@@ -221,7 +222,7 @@ Check out the OpenSCvx documentation:
 We also have a selection of problems in the `examples/` folder as well as on the [Examples page](https://openscvx.github.io/OpenSCvx/latest/Examples/abstract/brachistochrone/) of the documentation. The example trajectory optimization problems are grouped by application and represent some of the problem types that can be solved by OpenSCvx.
 
 > [!Note]
-> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode (`pip install -e .`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
+> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode with the plotting extra (`pip install -e ".[plotting]"`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
 
 To run a problem simply run any of the examples directly, for example:
 
@@ -260,7 +261,7 @@ OpenSCvx integrates with several optional third-party packages. Each installs as
 | [jaxlie](https://github.com/brentyi/jaxlie) | `lie` | SO(3)/SE(3) Lie-group operations and IK initialization | [![jaxlie](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml) |
 | [qpax](https://github.com/qpax-solver/qpax) | `qpax` | JAX-native QP solver backend | [![qpax](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-qpax.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/550) |
 | [CVXPYGen](https://github.com/cvxgrp/cvxpygen) | `cvxpygen` | Generated C solver code for the convex subproblem | [![cvxpygen](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-cvxpygen.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/551) |
-| [moreau](https://pypi.org/project/moreau/) | `moreau` | Licensed QP solver backend | ![moreau](https://img.shields.io/badge/moreau-license%20required-lightgrey) |
+| [moreau](https://pypi.org/project/moreau/) | `moreau` | Open Source SOCP solver backend | [![moreau](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml) |
 | [stljax](https://github.com/UW-CTRL/stljax) | `stl` | Signal Temporal Logic robustness bridge | ![stljax](https://img.shields.io/badge/stljax-no%20tests%20yet-lightgrey) |
 | [frax](https://github.com/danielpmorton/frax) | `frax` | Robot dynamics for the manipulator examples | [![frax](https://img.shields.io/badge/frax-weekly%20examples%20sweep-blue)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-examples.yml) |
 
@@ -275,7 +276,7 @@ This work was supported by a NASA Space Technology Graduate Research Opportunity
 Please cite the following works if you use the repository,
 
 ```tex
-@misc{hayner2026openscvxopensourcemodularextensible,
+@misc{hayner2026openscvx,
       title={OpenSCvx: An Open-Source Modular and Extensible Nonlinear Trajectory Planning Package}, 
       author={Christopher R. Hayner and Griffin J. Norris and Fabio Spada and Samet Uzun and Avi Mittal and Behcet Acıkmese and Karen Leung},
       year={2026},
@@ -315,4 +316,4 @@ keywords = {Trajectory optimization, Optimal control, Continuous-time constraint
 }
 ```
 
-Additionally please run `problem.citation()` to generate the appropriate bibtex citations for the corresponding components used within your problem.
+Additionally please run `problem.citation()` after `problem.intitialize()` to generate the appropriate bibtex citations for the corresponding components used within your problem.

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/README.md

@@ -109,9 +109,10 @@ uv pip install openscvx
 > [!TIP]
 > **Optional dependencies**
 >
-> For CVXPYGen code generation:
+> Plotting (Plotly, Matplotlib, and Viser) and CVXPYGen code generation:
 > ```bash
-> pip install openscvx[cvxpygen]
+> pip install "openscvx[plotting]"
+> pip install "openscvx[cvxpygen]"
 > ```
 
 > [!TIP]
@@ -174,7 +175,7 @@ Check out the OpenSCvx documentation:
 We also have a selection of problems in the `examples/` folder as well as on the [Examples page](https://openscvx.github.io/OpenSCvx/latest/Examples/abstract/brachistochrone/) of the documentation. The example trajectory optimization problems are grouped by application and represent some of the problem types that can be solved by OpenSCvx.
 
 > [!Note]
-> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode (`pip install -e .`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
+> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode with the plotting extra (`pip install -e ".[plotting]"`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
 
 To run a problem simply run any of the examples directly, for example:
 
@@ -213,7 +214,7 @@ OpenSCvx integrates with several optional third-party packages. Each installs as
 | [jaxlie](https://github.com/brentyi/jaxlie) | `lie` | SO(3)/SE(3) Lie-group operations and IK initialization | [![jaxlie](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml) |
 | [qpax](https://github.com/qpax-solver/qpax) | `qpax` | JAX-native QP solver backend | [![qpax](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-qpax.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/550) |
 | [CVXPYGen](https://github.com/cvxgrp/cvxpygen) | `cvxpygen` | Generated C solver code for the convex subproblem | [![cvxpygen](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-cvxpygen.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/551) |
-| [moreau](https://pypi.org/project/moreau/) | `moreau` | Licensed QP solver backend | ![moreau](https://img.shields.io/badge/moreau-license%20required-lightgrey) |
+| [moreau](https://pypi.org/project/moreau/) | `moreau` | Open Source SOCP solver backend | [![moreau](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml) |
 | [stljax](https://github.com/UW-CTRL/stljax) | `stl` | Signal Temporal Logic robustness bridge | ![stljax](https://img.shields.io/badge/stljax-no%20tests%20yet-lightgrey) |
 | [frax](https://github.com/danielpmorton/frax) | `frax` | Robot dynamics for the manipulator examples | [![frax](https://img.shields.io/badge/frax-weekly%20examples%20sweep-blue)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-examples.yml) |
 
@@ -228,7 +229,7 @@ This work was supported by a NASA Space Technology Graduate Research Opportunity
 Please cite the following works if you use the repository,
 
 ```tex
-@misc{hayner2026openscvxopensourcemodularextensible,
+@misc{hayner2026openscvx,
       title={OpenSCvx: An Open-Source Modular and Extensible Nonlinear Trajectory Planning Package}, 
       author={Christopher R. Hayner and Griffin J. Norris and Fabio Spada and Samet Uzun and Avi Mittal and Behcet Acıkmese and Karen Leung},
       year={2026},
@@ -268,4 +269,4 @@ keywords = {Trajectory optimization, Optimal control, Continuous-time constraint
 }
 ```
 
-Additionally please run `problem.citation()` to generate the appropriate bibtex citations for the corresponding components used within your problem.
+Additionally please run `problem.citation()` after `problem.intitialize()` to generate the appropriate bibtex citations for the corresponding components used within your problem.

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/docs/UsersGuide/01_hello_world_brachistochrone.md

@@ -316,7 +316,7 @@ theta_traj = results.trajectory["theta"]
 
 ### Visualizing the Results
 
-OpenSCvx provides built-in plotting utilities for quick visualization of your results. The simplest way to see your solution is with `plot_states()` and `plot_controls()`:
+OpenSCvx provides plotting utilities for quick visualization of your results. Install them with `pip install "openscvx[plotting]"`. The simplest way to see your solution is with `plot_states()` and `plot_controls()`:
 
 ```python
 from openscvx.plotting import plot_states, plot_controls

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/docs/UsersGuide/05_visualization.md

@@ -22,7 +22,11 @@ This tutorial covers:
 
 ## 2D Plots with Plotly
 
-The `openscvx.plotting` module provides high-level functions for common visualization tasks. These work with any problem and require no additional setup.
+The `openscvx.plotting` module provides high-level functions for common visualization tasks. These work with any problem once the plotting extra is installed:
+
+```sh
+pip install "openscvx[plotting]"
+```
 
 ### Time Series: States and Controls
 

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/docs/installation.md

@@ -9,8 +9,8 @@ description: >-
 
 OpenSCvx is published on PyPI as [`openscvx`](https://pypi.org/project/openscvx/)
 and requires **Python 3.11 or newer**. Its core depends on JAX and CVXPY; the
-optional extras below pull in additional solvers, dynamics backends, and modeling
-features only when you need them.
+optional extras below pull in plotting, additional solvers, dynamics backends,
+and modeling features only when you need them.
 
 ## How to install OpenSCvx
 
@@ -19,8 +19,8 @@ features only when you need them.
     ```sh
     uv pip install openscvx
 
-    # Optional: cvxpygen, stljax, jaxlie, mujoco-mjx extras
-    uv pip install "openscvx[cvxpygen,stl,lie,mjx]"
+    # Optional: plotting, cvxpygen, stljax, jaxlie, mujoco-mjx extras
+    uv pip install "openscvx[plotting,cvxpygen,stl,lie,mjx]"
     ```
 
 === "pip"
@@ -28,8 +28,8 @@ features only when you need them.
     ```sh
     pip install openscvx
 
-    # Optional: cvxpygen, stljax, jaxlie, mujoco-mjx extras
-    pip install "openscvx[cvxpygen,stl,lie,mjx]"
+    # Optional: plotting, cvxpygen, stljax, jaxlie, mujoco-mjx extras
+    pip install "openscvx[plotting,cvxpygen,stl,lie,mjx]"
     ```
 
 === "From source (editable)"
@@ -39,12 +39,13 @@ features only when you need them.
     cd OpenSCvx
     uv pip install -e .
 
-    # Optional: cvxpygen, stljax, jaxlie, mujoco-mjx extras
-    uv pip install -e ".[cvxpygen,stl,lie,mjx]"
+    # Optional: plotting, cvxpygen, stljax, jaxlie, mujoco-mjx extras
+    uv pip install -e ".[plotting,cvxpygen,stl,lie,mjx]"
     ```
 
-Install from source if you want to run the [examples](examples.md), modify the
-library, or track the development branch.
+Install from source if you want to modify the library or track the development
+branch. Running the [examples](examples.md) needs the `plotting` extra as well,
+because those scripts import Plotly and Viser at startup.
 
 ## Optional extras
 
@@ -52,9 +53,10 @@ Extras are installed with the `openscvx[extra1,extra2]` syntax shown above.
 
 | Extra | Enables |
 |-------|---------|
+| `plotting` | 2D plots and 3D scenes (`matplotlib`, `plotly`, `viser`). |
 | `cvxpygen` | Generated C solvers via `cvxpygen` + `qocogen`. |
 | `qpax` | The `qpax` differentiable QP backend. |
-| `moreau` | The licensed `moreau` conic solver. |
+| `moreau` | The `moreau` conic solver backend. |
 | `stl` | Signal Temporal Logic constraints via `stljax`. |
 | `lie` | Lie-group operators via `jaxlie`. |
 | `mjx` | MuJoCo MJX dynamics (`mujoco`, `mujoco-mjx`). |

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/examples/animations/_sensor_view.py

@@ -6,7 +6,8 @@ frame-perfectly aligned with the viser recording and can be composited with
 ffmpeg afterwards.
 
 Uses matplotlib's Agg canvas rather than plotly + kaleido so there's no
-headless-Chrome dependency — matplotlib is already a project dep. Raw RGB
+headless-Chrome dependency. Matplotlib comes with the plotting extra
+(``pip install 'openscvx[plotting]'``). Raw RGB
 frames are piped straight into ``ffmpeg -f rawvideo`` (same pattern as
 ``_render.py``), so there's no intermediate PNGs and no Python-side decode.
 """

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '0.5.3.dev57'
-__version_tuple__ = version_tuple = (0, 5, 3, 'dev57')
+__version__ = version = '0.5.3.dev58'
+__version_tuple__ = version_tuple = (0, 5, 3, 'dev58')
 
-__commit_id__ = commit_id = 'g9f23a62a0'
+__commit_id__ = commit_id = 'gb7be593bb'

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx/loader.py

@@ -89,7 +89,7 @@ def load_yaml(path: Union[str, Path]) -> dict:
     Returns:
         Dict of keyword arguments suitable for ``Problem(**result)``.
     """
-    import yaml  # optional dependency
+    import yaml
 
     with open(path) as f:
         data = yaml.safe_load(f)

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx/plotting/__init__.py

@@ -1,5 +1,9 @@
 """Trajectory visualization and plotting utilities.
 
+Requires the plotting extra::
+
+    pip install 'openscvx[plotting]'
+
 This module provides reusable building blocks for visualizing trajectory
 optimization results. It is intentionally minimal - we provide common utilities
 that can be composed together, not a complete solution that tries to do
@@ -41,27 +45,10 @@ For problem-specific visualization examples (drones, rockets, etc.), see
 ``examples/plotting_viser.py``.
 """
 
-try:
-    from . import viser
-except ModuleNotFoundError as e:
-    # Make viser an optional dependency so 2D plotting works without it.
-    # Accessing openscvx.plotting.viser will raise a helpful error message.
-    class _MissingOptionalDependency:
-        def __init__(self, *, package: str, exc: Exception):
-            self._package = package
-            self._exc = exc
-
-        def __getattr__(self, name: str):
-            raise ModuleNotFoundError(
-                f"Optional dependency '{self._package}' is required for 3D visualization.\n"
-                f"Install it with: pip install {self._package}\n"
-                f"Original error: {self._exc}"
-            ) from self._exc
-
-        def __repr__(self) -> str:  # pragma: no cover
-            return f"<missing optional dependency '{self._package}'>"
-
-    viser = _MissingOptionalDependency(package="viser", exc=e)  # type: ignore[assignment]
+from . import (
+    _deps,  # noqa: F401  # raises if the plotting extra is missing
+    viser,
+)
 from .plotting import (
     plot_control_component,
     plot_controls,

openscvx-0.5.3.dev58/openscvx/plotting/_deps.py

@@ -0,0 +1,18 @@
+"""Fail fast when the plotting extra is not installed.
+
+Imported by :mod:`openscvx.plotting` before Plotly, Matplotlib, or Viser.
+``import openscvx`` does not import this module, so solving a problem does not
+require the plotting stack.
+"""
+
+from importlib.util import find_spec
+
+_REQUIRED = ("matplotlib", "plotly", "viser")
+
+_missing = [name for name in _REQUIRED if find_spec(name) is None]
+if _missing:
+    missing = ", ".join(_missing)
+    raise ImportError(
+        "openscvx.plotting requires the optional plotting extra "
+        f"(missing {missing}). Install it with: pip install 'openscvx[plotting]'"
+    )

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx/solvers/moreau_ptr_solver.py

@@ -77,13 +77,11 @@ try:
     import jax.numpy as jnp
     import moreau
     from moreau.jax import Solver as _MoreauJaxSolver
-    from moreau.jax import solver as _moreau_jax_solver_fn  # functional factory
 
     _MOREAU_AVAILABLE = True
 except ImportError:  # pragma: no cover — exercised by the install-error test
     moreau = None  # type: ignore[assignment]
     _MoreauJaxSolver = None  # type: ignore[assignment,misc]
-    _moreau_jax_solver_fn = None  # type: ignore[assignment]
     jax = None  # type: ignore[assignment]
     jnp = None  # type: ignore[assignment]
     _MOREAU_AVAILABLE = False
@@ -269,12 +267,14 @@ class MoreauPTRSolver(PTRSolver):
     iterations on successful solves, and natively handles user
     :class:`~openscvx.solvers.cones.SOCConstraint` convex constraints.
 
-    The JAX-pure entry point :meth:`iteration_callback` builds Moreau's
-    functional factory ``moreau.jax.solver(...)`` once at
-    :meth:`initialize` and calls it as
-    ``(P_data, A_data, q, b) -> (JaxSolution, JaxSolveInfo)`` so the backend
-    composes with ``jax.jit`` and ``jax.vmap``. The functional API does not
-    expose a warm-start hook, so every call is a cold start: both
+    The JAX-pure entry point :meth:`iteration_callback` calls the jitted
+    kernel on the :class:`moreau.jax.Solver` built in :meth:`initialize`,
+    as ``(P_data, A_data, q, b) -> (JaxSolution, JaxSolveInfo)``, so the
+    backend composes with ``jax.jit`` and ``jax.vmap``. Moreau 0.4 removed
+    the public ``moreau.jax.solver`` factory. ``Solver.solve`` returns only
+    the solution and stores status on ``solver.info``, which is not batched
+    under ``vmap``. The kernel has no warm-start argument, so every call is
+    a cold start: both
     :meth:`~openscvx.problem.Problem.solve` and
     :meth:`~openscvx.problem.Problem.solve_jax` drive
     ``iteration_callback`` for the SCP body, so neither thread Moreau's
@@ -607,22 +607,6 @@ class MoreauPTRSolver(PTRSolver):
             settings=moreau_settings,
         )
 
-        # Functional-API solve function for ``iteration_callback``. Returns
-        # ``(JaxSolution, JaxSolveInfo)`` as registered pytrees and does NOT
-        # accept warm-start — see the 2026-05-17 Decision Log entry in
-        # ``plans/solver-iteration-callbacks.md``. The OO ``self._moreau``
-        # path stays in use for the NumPy ``solve()`` route (which warm-starts).
-        self._moreau_solve_fn = _moreau_jax_solver_fn(
-            n=L.n_z,
-            m=self._n_con,
-            P_row_offsets=jnp.array(P_indptr),
-            P_col_indices=jnp.array(P_indices),
-            A_row_offsets=jnp.array(A_indptr),
-            A_col_indices=jnp.array(A_indices),
-            cones=cones,
-            settings=moreau_settings,
-        )
-
     # ------------------------------------------------------------------
     # Per-iteration update hooks
     # ------------------------------------------------------------------
@@ -1386,16 +1370,19 @@ class MoreauPTRSolver(PTRSolver):
     def iteration_callback(self):
         """JAX-pure ``(state, SubproblemData) -> SubproblemSolution``.
 
-        Composes :meth:`_assemble_conic_jax` + Moreau's functional
-        ``solve_fn`` + :meth:`_build_solution_jax` into a single
-        ``@jax.jit``-decorated closure built once at :meth:`initialize`.
-
-        Moreau's functional API (``moreau.jax.solver()``) is the JIT/vmap-friendly
-        path but does **not** accept a warm-start (see the 2026-05-17 Decision
-        Log entry in ``plans/solver-iteration-callbacks.md``). Every call is a
-        cold start; ``state`` is accepted only for cross-backend signature
-        uniformity. Both :meth:`~openscvx.problem.Problem.solve` and
-        :meth:`~openscvx.problem.Problem.solve_jax` route through this
+        Composes :meth:`_assemble_conic_jax` + Moreau's jitted kernel
+        + :meth:`_build_solution_jax` into a single ``@jax.jit``-decorated
+        closure.
+
+        The kernel (``moreau.jax.Solver._solve_raw``) returns
+        ``(JaxSolution, JaxSolveInfo)`` and does **not** accept a warm-start
+        (see the 2026-05-17 Decision Log entry in
+        ``plans/solver-iteration-callbacks.md``). Moreau 0.4 removed the
+        public ``moreau.jax.solver`` factory; ``Solver.solve`` drops status
+        onto ``solver.info``, which does not batch under ``vmap``. Every call
+        is a cold start; ``state`` is accepted only for cross-backend
+        signature uniformity. Both :meth:`~openscvx.problem.Problem.solve`
+        and :meth:`~openscvx.problem.Problem.solve_jax` route through this
         callback for the SCP body, so the warm-start carry isn't threaded
         on either path — threading it would require an
         :class:`AlgorithmState.moreau_carry` field (future extension in
@@ -1403,7 +1390,7 @@ class MoreauPTRSolver(PTRSolver):
 
         The returned callable takes ``(state, data)``: ``state`` is the
         :class:`AlgorithmState` pytree, accepted for cross-backend signature
-        uniformity but unused (Moreau's functional API takes no warm-start);
+        uniformity but unused (the kernel takes no warm-start);
         ``data`` is the :class:`SubproblemData` pytree carrying the
         per-iteration linearization arrays, penalty weights, and boundary
         conditions.
@@ -1414,11 +1401,14 @@ class MoreauPTRSolver(PTRSolver):
                 "have been called first."
             )
 
-        solve_fn = self._moreau_solve_fn
+        # Public ``Solver.solve`` returns only the solution. The jitted kernel
+        # it calls still returns ``(JaxSolution, JaxSolveInfo)``, which is
+        # what ``vmap`` can carry as a pytree.
+        solve_fn = self._moreau._solve_raw
         csr_to_coo_perm = jnp.asarray(self._csr_to_coo_perm)
 
         def step(state, data: SubproblemData) -> SubproblemSolution:
-            del state  # Moreau's functional API takes no warm-start.
+            del state  # The jitted kernel takes no warm-start.
             P_data, coo_vals, q, b = self._assemble_conic_jax(data)
             A_data = coo_vals[csr_to_coo_perm]
             sol, info = solve_fn(P_data, A_data, q, b)

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx/solvers/ptr_solver.py

@@ -30,9 +30,10 @@ Backends:
         JAX-differentiable SCP loop in follow-up work.
     :class:`openscvx.solvers.moreau_ptr_solver.MoreauPTRSolver`
         Sparse conic program assembled as CSR JAX arrays and solved with
-        ``moreau.jax.Solver`` (NumPy path, warm-started between SCP
-        iterations) or the functional ``moreau.jax.solver(...)`` factory
-        (JAX-pure path, no warm-start). Uses SOC epigraphs for the L1 / pos
+        ``moreau.jax.Solver``. The NumPy path warm-starts between SCP
+        iterations. The JAX-pure path calls that solver's jitted kernel,
+        which returns ``(solution, info)`` and does not warm-start. Uses SOC
+        epigraphs for the L1 / pos
         PTR penalties instead of QPAX-style slack expansion.
 """
 

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58/openscvx.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: openscvx
-Version: 0.5.3.dev57
+Version: 0.5.3.dev58
 Summary: A general Python-based successive convexification implementation which uses a JAX backend.
 Author-email: Chris Hayner and Griffin Norris <haynec@uw.edu>
 License: Apache Software License
@@ -14,13 +14,9 @@ Requires-Dist: cvxpy>=1.8.1
 Requires-Dist: qoco
 Requires-Dist: numpy
 Requires-Dist: jax
-Requires-Dist: plotly
 Requires-Dist: termcolor
 Requires-Dist: diffrax
-Requires-Dist: absl-py
 Requires-Dist: flatbuffers
-Requires-Dist: viser
-Requires-Dist: matplotlib
 Requires-Dist: pyyaml
 Requires-Dist: pydantic>=2.0
 Provides-Extra: cvxpygen
@@ -39,6 +35,10 @@ Requires-Dist: mujoco>=3.0; extra == "mjx"
 Requires-Dist: mujoco-mjx>=3.0; extra == "mjx"
 Provides-Extra: frax
 Requires-Dist: frax>=0.0.4; extra == "frax"
+Provides-Extra: plotting
+Requires-Dist: matplotlib; extra == "plotting"
+Requires-Dist: plotly; extra == "plotting"
+Requires-Dist: viser; extra == "plotting"
 Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: pytest-xdist; extra == "test"
@@ -156,9 +156,10 @@ uv pip install openscvx
 > [!TIP]
 > **Optional dependencies**
 >
-> For CVXPYGen code generation:
+> Plotting (Plotly, Matplotlib, and Viser) and CVXPYGen code generation:
 > ```bash
-> pip install openscvx[cvxpygen]
+> pip install "openscvx[plotting]"
+> pip install "openscvx[cvxpygen]"
 > ```
 
 > [!TIP]
@@ -221,7 +222,7 @@ Check out the OpenSCvx documentation:
 We also have a selection of problems in the `examples/` folder as well as on the [Examples page](https://openscvx.github.io/OpenSCvx/latest/Examples/abstract/brachistochrone/) of the documentation. The example trajectory optimization problems are grouped by application and represent some of the problem types that can be solved by OpenSCvx.
 
 > [!Note]
-> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode (`pip install -e .`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
+> To run the examples, you'll need to clone this repository and install OpenSCvx in editable mode with the plotting extra (`pip install -e ".[plotting]"`). See the [Installing From Source](#installing-from-source) section above for detailed installation instructions.
 
 To run a problem simply run any of the examples directly, for example:
 
@@ -260,7 +261,7 @@ OpenSCvx integrates with several optional third-party packages. Each installs as
 | [jaxlie](https://github.com/brentyi/jaxlie) | `lie` | SO(3)/SE(3) Lie-group operations and IK initialization | [![jaxlie](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-lie.yml) |
 | [qpax](https://github.com/qpax-solver/qpax) | `qpax` | JAX-native QP solver backend | [![qpax](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-qpax.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/550) |
 | [CVXPYGen](https://github.com/cvxgrp/cvxpygen) | `cvxpygen` | Generated C solver code for the convex subproblem | [![cvxpygen](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-cvxpygen.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/issues/551) |
-| [moreau](https://pypi.org/project/moreau/) | `moreau` | Licensed QP solver backend | ![moreau](https://img.shields.io/badge/moreau-license%20required-lightgrey) |
+| [moreau](https://pypi.org/project/moreau/) | `moreau` | Open Source SOCP solver backend | [![moreau](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml/badge.svg?branch=main)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-moreau.yml) |
 | [stljax](https://github.com/UW-CTRL/stljax) | `stl` | Signal Temporal Logic robustness bridge | ![stljax](https://img.shields.io/badge/stljax-no%20tests%20yet-lightgrey) |
 | [frax](https://github.com/danielpmorton/frax) | `frax` | Robot dynamics for the manipulator examples | [![frax](https://img.shields.io/badge/frax-weekly%20examples%20sweep-blue)](https://github.com/OpenSCvx/OpenSCvx/actions/workflows/tests-examples.yml) |
 
@@ -275,7 +276,7 @@ This work was supported by a NASA Space Technology Graduate Research Opportunity
 Please cite the following works if you use the repository,
 
 ```tex
-@misc{hayner2026openscvxopensourcemodularextensible,
+@misc{hayner2026openscvx,
       title={OpenSCvx: An Open-Source Modular and Extensible Nonlinear Trajectory Planning Package}, 
       author={Christopher R. Hayner and Griffin J. Norris and Fabio Spada and Samet Uzun and Avi Mittal and Behcet Acıkmese and Karen Leung},
       year={2026},
@@ -315,4 +316,4 @@ keywords = {Trajectory optimization, Optimal control, Continuous-time constraint
 }
 ```
 
-Additionally please run `problem.citation()` to generate the appropriate bibtex citations for the corresponding components used within your problem.
+Additionally please run `problem.citation()` after `problem.intitialize()` to generate the appropriate bibtex citations for the corresponding components used within your problem.

{openscvx-0.5.3.dev57 → openscvx-0.5.3.dev58}/openscvx.egg-info/SOURCES.txt

@@ -20,6 +20,7 @@ pyproject.toml
 .github/workflows/tests-examples.yml
 .github/workflows/tests-lie.yml
 .github/workflows/tests-mjx.yml
+.github/workflows/tests-moreau.yml
 .github/workflows/tests-qpax.yml
 .github/workflows/tests.yml
 docs/changelog.md
@@ -301,6 +302,7 @@ openscvx/lowered/parameters.py
 openscvx/lowered/problem.py
 openscvx/lowered/unified.py
 openscvx/plotting/__init__.py
+openscvx/plotting/_deps.py
 openscvx/plotting/plotting.py
 openscvx/plotting/publication.py
 openscvx/plotting/scp_iteration.py