include/eigenpy/decompositions/sparse/cholmod/CholmodBase.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_base_hpp__
+#define __eigenpy_decomposition_sparse_cholmod_cholmod_base_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+#include "eigenpy/eigen/EigenBase.hpp"
+#include "eigenpy/decompositions/sparse/SparseSolverBase.hpp"
+
+#include <Eigen/CholmodSupport>
+
+namespace eigenpy {
+
+template <typename CholdmodDerived>
+struct CholmodBaseVisitor
+    : public boost::python::def_visitor<CholmodBaseVisitor<CholdmodDerived> > {
+  typedef CholdmodDerived Solver;
+
+  typedef typename CholdmodDerived::MatrixType MatrixType;
+  typedef typename MatrixType::Scalar Scalar;
+  typedef typename MatrixType::RealScalar RealScalar;
+  typedef MatrixType CholMatrixType;
+  typedef typename MatrixType::StorageIndex StorageIndex;
+
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl.def("analyzePattern", &Solver::analyzePattern,
+           bp::args("self", "matrix"),
+           "Performs a symbolic decomposition on the sparcity of matrix.\n"
+           "This function is particularly useful when solving for several "
+           "problems having the same structure.")
+
+        .def(EigenBaseVisitor<Solver>())
+        .def(SparseSolverBaseVisitor<Solver>())
+
+        .def("compute",
+             (Solver & (Solver::*)(const MatrixType &matrix)) & Solver::compute,
+             bp::args("self", "matrix"),
+             "Computes the sparse Cholesky decomposition of a given matrix.",
+             bp::return_self<>())
+
+        .def("determinant", &Solver::determinant, bp::arg("self"),
+             "Returns the determinant of the underlying matrix from the "
+             "current factorization.")
+
+        .def("factorize", &Solver::factorize, bp::args("self", "matrix"),
+             "Performs a numeric decomposition of a given matrix.\n"
+             "The given matrix must has the same sparcity than the matrix on "
+             "which the symbolic decomposition has been performed.\n"
+             "See also analyzePattern().")
+
+        .def("info", &Solver::info, bp::arg("self"),
+             "NumericalIssue if the input contains INF or NaN values or "
+             "overflow occured. Returns Success otherwise.")
+
+        .def("logDeterminant", &Solver::logDeterminant, bp::arg("self"),
+             "Returns the log determinant of the underlying matrix from the "
+             "current factorization.")
+
+        .def("setShift", &Solver::setShift, (bp::args("self", "offset")),
+             "Sets the shift parameters that will be used to adjust the "
+             "diagonal coefficients during the numerical factorization.\n"
+             "During the numerical factorization, the diagonal coefficients "
+             "are transformed by the following linear model: d_ii = offset + "
+             "d_ii.\n"
+             "The default is the identity transformation with offset=0.",
+             bp::return_self<>());
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_base_hpp__

include/eigenpy/decompositions/sparse/cholmod/CholmodDecomposition.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_decomposition_hpp__
+#define __eigenpy_decomposition_sparse_cholmod_cholmod_decomposition_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+#include "eigenpy/decompositions/sparse/cholmod/CholmodBase.hpp"
+
+namespace eigenpy {
+
+template <typename CholdmodDerived>
+struct CholmodDecompositionVisitor
+    : public boost::python::def_visitor<
+          CholmodDecompositionVisitor<CholdmodDerived> > {
+  typedef CholdmodDerived Solver;
+
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl
+
+        .def(CholmodBaseVisitor<Solver>())
+        .def("setMode", &Solver::setMode, bp::args("self", "mode"),
+             "Set the mode for the Cholesky decomposition.");
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef
+        // __eigenpy_decomposition_sparse_cholmod_cholmod_decomposition_hpp__

include/eigenpy/decompositions/sparse/cholmod/CholmodSimplicialLDLT.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_ldlt_hpp__
+#define __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_ldlt_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+#include "eigenpy/decompositions/sparse/cholmod/CholmodDecomposition.hpp"
+#include "eigenpy/utils/scalar-name.hpp"
+
+namespace eigenpy {
+
+template <typename MatrixType_, int UpLo_ = Eigen::Lower>
+struct CholmodSimplicialLDLTVisitor
+    : public boost::python::def_visitor<
+          CholmodSimplicialLDLTVisitor<MatrixType_, UpLo_> > {
+  typedef MatrixType_ MatrixType;
+  typedef typename MatrixType::Scalar Scalar;
+  typedef typename MatrixType::RealScalar RealScalar;
+
+  typedef Eigen::CholmodSimplicialLDLT<MatrixType_, UpLo_> Solver;
+
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl
+
+        .def(CholmodBaseVisitor<Solver>())
+        .def(bp::init<>(bp::arg("self"), "Default constructor"))
+        .def(bp::init<MatrixType>(bp::args("self", "matrix"),
+                                  "Constructs and performs the LDLT "
+                                  "factorization from a given matrix."))
+
+        ;
+  }
+
+  static void expose() {
+    static const std::string classname =
+        "CholmodSimplicialLDLT_" + scalar_name<Scalar>::shortname();
+    expose(classname);
+  }
+
+  static void expose(const std::string &name) {
+    bp::class_<Solver, boost::noncopyable>(
+        name.c_str(),
+        "A simplicial direct Cholesky (LDLT) factorization and solver based on "
+        "Cholmod.\n\n"
+        "This class allows to solve for A.X = B sparse linear problems via a "
+        "simplicial LL^T Cholesky factorization using the Cholmod library."
+        "This simplicial variant is equivalent to Eigen's built-in "
+        "SimplicialLDLT class."
+        "Therefore, it has little practical interest. The sparse matrix A must "
+        "be selfadjoint and positive definite."
+        "The vectors or matrices X and B can be either dense or sparse.",
+        bp::no_init)
+        .def(CholmodSimplicialLDLTVisitor());
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef
+        // __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_ldlt_hpp__

include/eigenpy/decompositions/sparse/cholmod/CholmodSimplicialLLT.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_llt_hpp__
+#define __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_llt_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+#include "eigenpy/decompositions/sparse/cholmod/CholmodDecomposition.hpp"
+#include "eigenpy/utils/scalar-name.hpp"
+
+namespace eigenpy {
+
+template <typename MatrixType_, int UpLo_ = Eigen::Lower>
+struct CholmodSimplicialLLTVisitor
+    : public boost::python::def_visitor<
+          CholmodSimplicialLLTVisitor<MatrixType_, UpLo_> > {
+  typedef MatrixType_ MatrixType;
+  typedef typename MatrixType::Scalar Scalar;
+  typedef typename MatrixType::RealScalar RealScalar;
+
+  typedef Eigen::CholmodSimplicialLLT<MatrixType_, UpLo_> Solver;
+
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl
+
+        .def(CholmodBaseVisitor<Solver>())
+        .def(bp::init<>(bp::arg("self"), "Default constructor"))
+        .def(bp::init<MatrixType>(bp::args("self", "matrix"),
+                                  "Constructs and performs the LLT "
+                                  "factorization from a given matrix."))
+
+        ;
+  }
+
+  static void expose() {
+    static const std::string classname =
+        "CholmodSimplicialLLT_" + scalar_name<Scalar>::shortname();
+    expose(classname);
+  }
+
+  static void expose(const std::string &name) {
+    bp::class_<Solver, boost::noncopyable>(
+        name.c_str(),
+        "A simplicial direct Cholesky (LLT) factorization and solver based on "
+        "Cholmod.\n\n"
+        "This class allows to solve for A.X = B sparse linear problems via a "
+        "simplicial LL^T Cholesky factorization using the Cholmod library."
+        "This simplicial variant is equivalent to Eigen's built-in "
+        "SimplicialLLT class."
+        "Therefore, it has little practical interest. The sparse matrix A must "
+        "be selfadjoint and positive definite."
+        "The vectors or matrices X and B can be either dense or sparse.",
+        bp::no_init)
+        .def(CholmodSimplicialLLTVisitor());
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef
+        // __eigenpy_decomposition_sparse_cholmod_cholmod_simplicial_llt_hpp__

include/eigenpy/decompositions/sparse/cholmod/CholmodSupernodalLLT.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_decomposition_sparse_cholmod_cholmod_supernodal_llt_hpp__
+#define __eigenpy_decomposition_sparse_cholmod_cholmod_supernodal_llt_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+#include "eigenpy/decompositions/sparse/cholmod/CholmodDecomposition.hpp"
+#include "eigenpy/utils/scalar-name.hpp"
+
+namespace eigenpy {
+
+template <typename MatrixType_, int UpLo_ = Eigen::Lower>
+struct CholmodSupernodalLLTVisitor
+    : public boost::python::def_visitor<
+          CholmodSupernodalLLTVisitor<MatrixType_, UpLo_> > {
+  typedef MatrixType_ MatrixType;
+  typedef typename MatrixType::Scalar Scalar;
+  typedef typename MatrixType::RealScalar RealScalar;
+
+  typedef Eigen::CholmodSupernodalLLT<MatrixType_, UpLo_> Solver;
+
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl
+
+        .def(CholmodBaseVisitor<Solver>())
+        .def(bp::init<>(bp::arg("self"), "Default constructor"))
+        .def(bp::init<MatrixType>(bp::args("self", "matrix"),
+                                  "Constructs and performs the LLT "
+                                  "factorization from a given matrix."))
+
+        ;
+  }
+
+  static void expose() {
+    static const std::string classname =
+        "CholmodSupernodalLLT_" + scalar_name<Scalar>::shortname();
+    expose(classname);
+  }
+
+  static void expose(const std::string &name) {
+    bp::class_<Solver, boost::noncopyable>(
+        name.c_str(),
+        "A supernodal direct Cholesky (LLT) factorization and solver based on "
+        "Cholmod.\n\n"
+        "This class allows to solve for A.X = B sparse linear problems via a "
+        "supernodal LL^T Cholesky factorization using the Cholmod library."
+        "This supernodal variant performs best on dense enough problems, e.g., "
+        "3D FEM, or very high order 2D FEM."
+        "The sparse matrix A must be selfadjoint and positive definite. The "
+        "vectors or matrices X and B can be either dense or sparse.",
+        bp::no_init)
+        .def(CholmodSupernodalLLTVisitor());
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef
+        // __eigenpy_decomposition_sparse_cholmod_cholmod_supernodal_llt_hpp__

include/eigenpy/deprecation-policy.hpp

+//
+// Copyright (C) 2020 INRIA
+// Copyright (C) 2024 LAAS-CNRS, INRIA
+//
+#ifndef __eigenpy_deprecation_hpp__
+#define __eigenpy_deprecation_hpp__
+
+#include "eigenpy/fwd.hpp"
+
+namespace eigenpy {
+
+enum class DeprecationType { DEPRECATION, FUTURE };
+
+namespace detail {
+
+inline PyObject *deprecationTypeToPyObj(DeprecationType dep) {
+  switch (dep) {
+    case DeprecationType::DEPRECATION:
+      return PyExc_DeprecationWarning;
+    case DeprecationType::FUTURE:
+      return PyExc_FutureWarning;
+    default:  // The switch handles all cases explicitly, this should never be
+              // triggered.
+      throw std::invalid_argument(
+          "Undefined DeprecationType - this should never be triggered.");
+  }
+}
+
+}  // namespace detail
+
+/// @brief A Boost.Python call policy which triggers a Python warning on
+/// precall.
+template <DeprecationType deprecation_type = DeprecationType::DEPRECATION,
+          class BasePolicy = bp::default_call_policies>
+struct deprecation_warning_policy : BasePolicy {
+  using result_converter = typename BasePolicy::result_converter;
+  using argument_package = typename BasePolicy::argument_package;
+
+  deprecation_warning_policy(const std::string &warning_msg)
+      : BasePolicy(), m_what(warning_msg) {}
+
+  std::string what() const { return m_what; }
+
+  const BasePolicy *derived() const {
+    return static_cast<const BasePolicy *>(this);
+  }
+
+  template <class ArgPackage>
+  bool precall(const ArgPackage &args) const {
+    PyErr_WarnEx(detail::deprecationTypeToPyObj(deprecation_type),
+                 m_what.c_str(), 1);
+    return derived()->precall(args);
+  }
+
+ protected:
+  const std::string m_what;
+};
+
+template <DeprecationType deprecation_type = DeprecationType::DEPRECATION,
+          class BasePolicy = bp::default_call_policies>
+struct deprecated_function
+    : deprecation_warning_policy<deprecation_type, BasePolicy> {
+  deprecated_function(const std::string &msg =
+                          "This function has been marked as deprecated, and "
+                          "will be removed in the future.")
+      : deprecation_warning_policy<deprecation_type, BasePolicy>(msg) {}
+};
+
+template <DeprecationType deprecation_type = DeprecationType::DEPRECATION,
+          class BasePolicy = bp::default_call_policies>
+struct deprecated_member
+    : deprecation_warning_policy<deprecation_type, BasePolicy> {
+  deprecated_member(const std::string &msg =
+                        "This attribute or method has been marked as "
+                        "deprecated, and will be removed in the future.")
+      : deprecation_warning_policy<deprecation_type, BasePolicy>(msg) {}
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef __eigenpy_deprecation_hpp__

include/eigenpy/details.hpp

 /*
  * Copyright 2014-2019, CNRS
- * Copyright 2018-2023, INRIA
+ * Copyright 2018-2024, INRIA
  */
 
 #ifndef __eigenpy_details_hpp__
@@ -40,6 +40,24 @@ struct expose_eigen_type_impl<MatType, Eigen::MatrixBase<MatType>, Scalar> {
   }
 };
 
+template <typename MatType, typename Scalar>
+struct expose_eigen_type_impl<MatType, Eigen::SparseMatrixBase<MatType>,
+                              Scalar> {
+  static void run() {
+    if (check_registration<MatType>()) return;
+
+    // to-python
+    EigenToPyConverter<MatType>::registration();
+    // #if EIGEN_VERSION_AT_LEAST(3, 2, 0)
+    //     EigenToPyConverter<Eigen::Ref<MatType> >::registration();
+    //     EigenToPyConverter<const Eigen::Ref<const MatType> >::registration();
+    // #endif
+
+    // from-python
+    EigenFromPyConverter<MatType>::registration();
+  }
+};
+
 #ifdef EIGENPY_WITH_TENSOR_SUPPORT
 template <typename TensorType, typename Scalar>
 struct expose_eigen_type_impl<TensorType, Eigen::TensorBase<TensorType>,

include/eigenpy/eigen-allocator.hpp

@@ -198,6 +198,91 @@ struct cast<Scalar, NewScalar, EigenBase, false> {
       mat, NumpyMap<MatType, NewScalar>::map(                                 \
                pyArray, details::check_swap(pyArray, mat)))
 
+// Define specific cast for Windows and Mac
+#if defined _WIN32 || defined __CYGWIN__
+// Manage NPY_INT on Windows (NPY_INT32 is NPY_LONG).
+// See https://github.com/stack-of-tasks/eigenpy/pull/455
+#define EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH_OS_SPECIFIC( \
+    MatType, Scalar, pyArray, mat, CAST_MACRO)               \
+  case NPY_INT:                                              \
+    CAST_MACRO(MatType, int32_t, Scalar, pyArray, mat);      \
+    break;                                                   \
+  case NPY_UINT:                                             \
+    CAST_MACRO(MatType, uint32_t, Scalar, pyArray, mat);     \
+    break;
+#elif defined __APPLE__
+// Manage NPY_LONGLONG on Mac (NPY_INT64 is NPY_LONG).
+// long long and long are both the same type
+// but NPY_LONGLONG and NPY_LONG are different dtype.
+// See https://github.com/stack-of-tasks/eigenpy/pull/455
+#define EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH_OS_SPECIFIC( \
+    MatType, Scalar, pyArray, mat, CAST_MACRO)               \
+  case NPY_LONGLONG:                                         \
+    CAST_MACRO(MatType, int64_t, Scalar, pyArray, mat);      \
+    break;                                                   \
+  case NPY_ULONGLONG:                                        \
+    CAST_MACRO(MatType, uint64_t, Scalar, pyArray, mat);     \
+    break;
+#else
+#define EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH_OS_SPECIFIC( \
+    MatType, Scalar, pyArray, mat, CAST_MACRO)
+#endif
+
+/// Define casting between Numpy matrix type to Eigen type.
+#define EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH(                               \
+    pyArray_type_code, MatType, Scalar, pyArray, mat, CAST_MACRO)              \
+  switch (pyArray_type_code) {                                                 \
+    case NPY_BOOL:                                                             \
+      CAST_MACRO(MatType, bool, Scalar, pyArray, mat);                         \
+      break;                                                                   \
+    case NPY_INT8:                                                             \
+      CAST_MACRO(MatType, int8_t, Scalar, pyArray, mat);                       \
+      break;                                                                   \
+    case NPY_INT16:                                                            \
+      CAST_MACRO(MatType, int16_t, Scalar, pyArray, mat);                      \
+      break;                                                                   \
+    case NPY_INT32:                                                            \
+      CAST_MACRO(MatType, int32_t, Scalar, pyArray, mat);                      \
+      break;                                                                   \
+    case NPY_INT64:                                                            \
+      CAST_MACRO(MatType, int64_t, Scalar, pyArray, mat);                      \
+      break;                                                                   \
+    case NPY_UINT8:                                                            \
+      CAST_MACRO(MatType, uint8_t, Scalar, pyArray, mat);                      \
+      break;                                                                   \
+    case NPY_UINT16:                                                           \
+      CAST_MACRO(MatType, uint16_t, Scalar, pyArray, mat);                     \
+      break;                                                                   \
+    case NPY_UINT32:                                                           \
+      CAST_MACRO(MatType, uint32_t, Scalar, pyArray, mat);                     \
+      break;                                                                   \
+    case NPY_UINT64:                                                           \
+      CAST_MACRO(MatType, uint64_t, Scalar, pyArray, mat);                     \
+      break;                                                                   \
+    case NPY_FLOAT:                                                            \
+      CAST_MACRO(MatType, float, Scalar, pyArray, mat);                        \
+      break;                                                                   \
+    case NPY_CFLOAT:                                                           \
+      CAST_MACRO(MatType, std::complex<float>, Scalar, pyArray, mat);          \
+      break;                                                                   \
+    case NPY_DOUBLE:                                                           \
+      CAST_MACRO(MatType, double, Scalar, pyArray, mat);                       \
+      break;                                                                   \
+    case NPY_CDOUBLE:                                                          \
+      CAST_MACRO(MatType, std::complex<double>, Scalar, pyArray, mat);         \
+      break;                                                                   \
+    case NPY_LONGDOUBLE:                                                       \
+      CAST_MACRO(MatType, long double, Scalar, pyArray, mat);                  \
+      break;                                                                   \
+    case NPY_CLONGDOUBLE:                                                      \
+      CAST_MACRO(MatType, std::complex<long double>, Scalar, pyArray, mat);    \
+      break;                                                                   \
+      EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH_OS_SPECIFIC(                     \
+          MatType, Scalar, pyArray, mat, CAST_MACRO)                           \
+    default:                                                                   \
+      throw Exception("You asked for a conversion which is not implemented."); \
+  }
+
 template <typename EigenType>
 struct EigenAllocator;
 
@@ -247,43 +332,9 @@ struct eigen_allocator_impl_matrix {
           pyArray, details::check_swap(pyArray, mat));  // avoid useless cast
       return;
     }
-
-    switch (pyArray_type_code) {
-      case NPY_INT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, int, Scalar, pyArray,
-                                                  mat);
-        break;
-      case NPY_LONG:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, long, Scalar,
-                                                  pyArray, mat);
-        break;
-      case NPY_FLOAT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, float, Scalar,
-                                                  pyArray, mat);
-        break;
-      case NPY_CFLOAT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, std::complex<float>,
-                                                  Scalar, pyArray, mat);
-        break;
-      case NPY_DOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, double, Scalar,
-                                                  pyArray, mat);
-        break;
-      case NPY_CDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, std::complex<double>,
-                                                  Scalar, pyArray, mat);
-        break;
-      case NPY_LONGDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(MatType, long double, Scalar,
-                                                  pyArray, mat);
-        break;
-      case NPY_CLONGDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX(
-            MatType, std::complex<long double>, Scalar, pyArray, mat);
-        break;
-      default:
-        throw Exception("You asked for a conversion which is not implemented.");
-    }
+    EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH(
+        pyArray_type_code, MatType, Scalar, pyArray, mat,
+        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_MATRIX);
   }
 
   /// \brief Copy mat into the Python array using Eigen::Map
@@ -301,43 +352,8 @@ struct eigen_allocator_impl_matrix {
                                      details::check_swap(pyArray, mat)) = mat;
       return;
     }
-
-    switch (pyArray_type_code) {
-      case NPY_INT:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(MatType, Scalar, int, mat,
-                                                  pyArray);
-        break;
-      case NPY_LONG:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(MatType, Scalar, long, mat,
-                                                  pyArray);
-        break;
-      case NPY_FLOAT:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(MatType, Scalar, float, mat,
-                                                  pyArray);
-        break;
-      case NPY_CFLOAT:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(
-            MatType, Scalar, std::complex<float>, mat, pyArray);
-        break;
-      case NPY_DOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(MatType, Scalar, double, mat,
-                                                  pyArray);
-        break;
-      case NPY_CDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(
-            MatType, Scalar, std::complex<double>, mat, pyArray);
-        break;
-      case NPY_LONGDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(MatType, Scalar, long double,
-                                                  mat, pyArray);
-        break;
-      case NPY_CLONGDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_MATRIX_TO_PYARRAY(
-            MatType, Scalar, std::complex<long double>, mat, pyArray);
-        break;
-      default:
-        throw Exception("You asked for a conversion which is not implemented.");
-    }
+    throw Exception(
+        "Scalar conversion from Eigen to Numpy is not implemented.");
   }
 };
 
@@ -394,42 +410,9 @@ struct eigen_allocator_impl_tensor {
       return;
     }
 
-    switch (pyArray_type_code) {
-      case NPY_INT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(TensorType, int, Scalar,
-                                                  pyArray, tensor);
-        break;
-      case NPY_LONG:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(TensorType, long, Scalar,
-                                                  pyArray, tensor);
-        break;
-      case NPY_FLOAT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(TensorType, float, Scalar,
-                                                  pyArray, tensor);
-        break;
-      case NPY_CFLOAT:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(
-            TensorType, std::complex<float>, Scalar, pyArray, tensor);
-        break;
-      case NPY_DOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(TensorType, double, Scalar,
-                                                  pyArray, tensor);
-        break;
-      case NPY_CDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(
-            TensorType, std::complex<double>, Scalar, pyArray, tensor);
-        break;
-      case NPY_LONGDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(TensorType, long double,
-                                                  Scalar, pyArray, tensor);
-        break;
-      case NPY_CLONGDOUBLE:
-        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR(
-            TensorType, std::complex<long double>, Scalar, pyArray, tensor);
-        break;
-      default:
-        throw Exception("You asked for a conversion which is not implemented.");
-    }
+    EIGENPY_CAST_FROM_NUMPY_TO_EIGEN_SWITCH(
+        pyArray_type_code, TensorType, Scalar, pyArray, tensor,
+        EIGENPY_CAST_FROM_PYARRAY_TO_EIGEN_TENSOR);
   }
 
 #define EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar,         \
@@ -454,42 +437,8 @@ struct eigen_allocator_impl_tensor {
       return;
     }
 
-    switch (pyArray_type_code) {
-      case NPY_INT:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar, int,
-                                                  tensor, pyArray);
-        break;
-      case NPY_LONG:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar, long,
-                                                  tensor, pyArray);
-        break;
-      case NPY_FLOAT:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar, float,
-                                                  tensor, pyArray);
-        break;
-      case NPY_CFLOAT:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(
-            TensorType, Scalar, std::complex<float>, tensor, pyArray);
-        break;
-      case NPY_DOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar, double,
-                                                  tensor, pyArray);
-        break;
-      case NPY_CDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(
-            TensorType, Scalar, std::complex<double>, tensor, pyArray);
-        break;
-      case NPY_LONGDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(TensorType, Scalar,
-                                                  long double, tensor, pyArray);
-        break;
-      case NPY_CLONGDOUBLE:
-        EIGENPY_CAST_FROM_EIGEN_TENSOR_TO_PYARRAY(
-            TensorType, Scalar, std::complex<long double>, tensor, pyArray);
-        break;
-      default:
-        throw Exception("You asked for a conversion which is not implemented.");
-    }
+    throw Exception(
+        "Scalar conversion from Eigen to Numpy is not implemented.");
   }
 };
 #endif

include/eigenpy/eigen-from-python.hpp

@@ -62,7 +62,7 @@ struct copy_if_non_const<const MatType, true> {
 template <typename _RefType>
 struct referent_storage_eigen_ref {
   typedef _RefType RefType;
-  typedef typename get_eigen_ref_plain_type<RefType>::type PlainObjectType;
+  typedef typename get_eigen_plain_type<RefType>::type PlainObjectType;
   typedef typename ::eigenpy::aligned_storage<
       ::boost::python::detail::referent_size<RefType &>::value>::type
       AlignedStorage;
@@ -290,16 +290,10 @@ struct eigen_from_py_impl<MatType, Eigen::MatrixBase<MatType> > {
   static void registration();
 };
 
-#ifdef EIGENPY_MSVC_COMPILER
-template <typename EigenType>
-struct EigenFromPy<EigenType,
-                   typename boost::remove_reference<EigenType>::type::Scalar>
-#else
-template <typename EigenType, typename _Scalar>
-struct EigenFromPy
-#endif
-    : eigen_from_py_impl<EigenType> {
-};
+template <typename EigenType,
+          typename Scalar =
+              typename boost::remove_reference<EigenType>::type::Scalar>
+struct EigenFromPy : eigen_from_py_impl<EigenType> {};
 
 template <typename MatType>
 void *eigen_from_py_impl<MatType, Eigen::MatrixBase<MatType> >::convertible(
@@ -565,4 +559,6 @@ struct EigenFromPy<const Eigen::Ref<const MatType, Options, Stride> > {
 #include "eigenpy/tensor/eigen-from-python.hpp"
 #endif
 
+#include "eigenpy/sparse/eigen-from-python.hpp"
+
 #endif  // __eigenpy_eigen_from_python_hpp__

include/eigenpy/eigen-to-python.hpp

 //
-// Copyright (c) 2014-2023 CNRS INRIA
+// Copyright (c) 2014-2024 CNRS INRIA
 //
 
 #ifndef __eigenpy_eigen_to_python_hpp__
@@ -11,54 +11,11 @@
 
 #include "eigenpy/eigen-allocator.hpp"
 #include "eigenpy/numpy-allocator.hpp"
+#include "eigenpy/scipy-allocator.hpp"
 #include "eigenpy/numpy-type.hpp"
+#include "eigenpy/scipy-type.hpp"
 #include "eigenpy/registration.hpp"
 
-namespace boost {
-namespace python {
-
-template <typename MatrixRef, class MakeHolder>
-struct to_python_indirect_eigen {
-  template <class U>
-  inline PyObject* operator()(U const& mat) const {
-    return eigenpy::EigenToPy<MatrixRef>::convert(const_cast<U&>(mat));
-  }
-
-#ifndef BOOST_PYTHON_NO_PY_SIGNATURES
-  inline PyTypeObject const* get_pytype() const {
-    return converter::registered_pytype<MatrixRef>::get_pytype();
-  }
-#endif
-};
-
-template <typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime,
-          int Options, int MaxRowsAtCompileTime, int MaxColsAtCompileTime,
-          class MakeHolder>
-struct to_python_indirect<
-    Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
-                  MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
-    MakeHolder>
-    : to_python_indirect_eigen<
-          Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
-                        MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
-          MakeHolder> {};
-
-template <typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime,
-          int Options, int MaxRowsAtCompileTime, int MaxColsAtCompileTime,
-          class MakeHolder>
-struct to_python_indirect<
-    const Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
-                        MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
-    MakeHolder>
-    : to_python_indirect_eigen<
-          const Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime,
-                              Options, MaxRowsAtCompileTime,
-                              MaxColsAtCompileTime>&,
-          MakeHolder> {};
-
-}  // namespace python
-}  // namespace boost
-
 namespace eigenpy {
 
 EIGENPY_DOCUMENTATION_START_IGNORE
@@ -101,9 +58,8 @@ struct eigen_to_py_impl_matrix {
     PyArrayObject* pyArray;
     // Allocate Python memory
     if ((((!(C == 1) != !(R == 1)) && !MatrixDerived::IsVectorAtCompileTime) ||
-         MatrixDerived::IsVectorAtCompileTime) &&
-        NumpyType::getType() ==
-            ARRAY_TYPE)  // Handle array with a single dimension
+         MatrixDerived::IsVectorAtCompileTime))  // Handle array with a single
+                                                 // dimension
     {
       npy_intp shape[1] = {C == 1 ? R : C};
       pyArray = NumpyAllocator<MatType>::allocate(
@@ -117,6 +73,50 @@ struct eigen_to_py_impl_matrix {
     // Create an instance (either np.array or np.matrix)
     return NumpyType::make(pyArray).ptr();
   }
+
+  static PyTypeObject const* get_pytype() { return getPyArrayType(); }
+};
+
+template <typename MatType>
+struct eigen_to_py_impl_sparse_matrix;
+
+template <typename MatType>
+struct eigen_to_py_impl<MatType, Eigen::SparseMatrixBase<MatType> >
+    : eigen_to_py_impl_sparse_matrix<MatType> {};
+
+template <typename MatType>
+struct eigen_to_py_impl<MatType&, Eigen::SparseMatrixBase<MatType> >
+    : eigen_to_py_impl_sparse_matrix<MatType&> {};
+
+template <typename MatType>
+struct eigen_to_py_impl<const MatType, const Eigen::SparseMatrixBase<MatType> >
+    : eigen_to_py_impl_sparse_matrix<const MatType> {};
+
+template <typename MatType>
+struct eigen_to_py_impl<const MatType&, const Eigen::SparseMatrixBase<MatType> >
+    : eigen_to_py_impl_sparse_matrix<const MatType&> {};
+
+template <typename MatType>
+struct eigen_to_py_impl_sparse_matrix {
+  enum { IsRowMajor = MatType::IsRowMajor };
+
+  static PyObject* convert(
+      typename boost::add_reference<
+          typename boost::add_const<MatType>::type>::type mat) {
+    typedef typename boost::remove_const<
+        typename boost::remove_reference<MatType>::type>::type MatrixDerived;
+
+    // Allocate and perform the copy
+    PyObject* pyArray =
+        ScipyAllocator<MatType>::allocate(const_cast<MatrixDerived&>(mat));
+
+    return pyArray;
+  }
+
+  static PyTypeObject const* get_pytype() {
+    return IsRowMajor ? ScipyType::getScipyCSRMatrixType()
+                      : ScipyType::getScipyCSCMatrixType();
+  }
 };
 
 #ifdef EIGENPY_WITH_TENSOR_SUPPORT
@@ -150,22 +150,17 @@ struct eigen_to_py_impl_tensor {
     // Create an instance (either np.array or np.matrix)
     return NumpyType::make(pyArray).ptr();
   }
+
+  static PyTypeObject const* get_pytype() { return getPyArrayType(); }
 };
 #endif
 
 EIGENPY_DOCUMENTATION_END_IGNORE
 
-#ifdef EIGENPY_MSVC_COMPILER
-template <typename EigenType>
-struct EigenToPy<EigenType,
-                 typename boost::remove_reference<EigenType>::type::Scalar>
-#else
-template <typename EigenType, typename _Scalar>
-struct EigenToPy
-#endif
-    : eigen_to_py_impl<EigenType> {
-  static PyTypeObject const* get_pytype() { return getPyArrayType(); }
-};
+template <typename EigenType,
+          typename Scalar =
+              typename boost::remove_reference<EigenType>::type::Scalar>
+struct EigenToPy : eigen_to_py_impl<EigenType> {};
 
 template <typename MatType>
 struct EigenToPyConverter {
@@ -173,6 +168,52 @@ struct EigenToPyConverter {
     bp::to_python_converter<MatType, EigenToPy<MatType>, true>();
   }
 };
+
 }  // namespace eigenpy
 
+namespace boost {
+namespace python {
+
+template <typename MatrixRef, class MakeHolder>
+struct to_python_indirect_eigen {
+  template <class U>
+  inline PyObject* operator()(U const& mat) const {
+    return eigenpy::EigenToPy<MatrixRef>::convert(const_cast<U&>(mat));
+  }
+
+#ifndef BOOST_PYTHON_NO_PY_SIGNATURES
+  inline PyTypeObject const* get_pytype() const {
+    return converter::registered_pytype<MatrixRef>::get_pytype();
+  }
+#endif
+};
+
+template <typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime,
+          int Options, int MaxRowsAtCompileTime, int MaxColsAtCompileTime,
+          class MakeHolder>
+struct to_python_indirect<
+    Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
+                  MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
+    MakeHolder>
+    : to_python_indirect_eigen<
+          Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
+                        MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
+          MakeHolder> {};
+
+template <typename Scalar, int RowsAtCompileTime, int ColsAtCompileTime,
+          int Options, int MaxRowsAtCompileTime, int MaxColsAtCompileTime,
+          class MakeHolder>
+struct to_python_indirect<
+    const Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime, Options,
+                        MaxRowsAtCompileTime, MaxColsAtCompileTime>&,
+    MakeHolder>
+    : to_python_indirect_eigen<
+          const Eigen::Matrix<Scalar, RowsAtCompileTime, ColsAtCompileTime,
+                              Options, MaxRowsAtCompileTime,
+                              MaxColsAtCompileTime>&,
+          MakeHolder> {};
+
+}  // namespace python
+}  // namespace boost
+
 #endif  // __eigenpy_eigen_to_python_hpp__

include/eigenpy/eigen-typedef.hpp

@@ -32,6 +32,7 @@
   EIGENPY_MAKE_FIXED_TYPEDEFS(Type, Options, TypeSuffix, 2)           \
   EIGENPY_MAKE_FIXED_TYPEDEFS(Type, Options, TypeSuffix, 3)           \
   EIGENPY_MAKE_FIXED_TYPEDEFS(Type, Options, TypeSuffix, 4)           \
-  EIGENPY_MAKE_TYPEDEFS(Type, Options, TypeSuffix, 1, 1)
+  EIGENPY_MAKE_TYPEDEFS(Type, Options, TypeSuffix, 1, 1)              \
+  typedef Eigen::SparseMatrix<Scalar, Options> SparseMatrixX##TypeSuffix
 
 #endif  // ifndef __eigenpy_eigen_typedef_hpp__

include/eigenpy/eigen/EigenBase.hpp

+/*
+ * Copyright 2024 INRIA
+ */
+
+#ifndef __eigenpy_eigen_eigen_base_hpp__
+#define __eigenpy_eigen_eigen_base_hpp__
+
+#include "eigenpy/eigenpy.hpp"
+
+namespace eigenpy {
+
+template <typename Derived>
+struct EigenBaseVisitor
+    : public boost::python::def_visitor<EigenBaseVisitor<Derived> > {
+  template <class PyClass>
+  void visit(PyClass &cl) const {
+    cl.def("cols", &Derived::cols, bp::arg("self"),
+           "Returns the number of columns.")
+        .def("rows", &Derived::rows, bp::arg("self"),
+             "Returns the number of rows.")
+        .def("size", &Derived::rows, bp::arg("self"),
+             "Returns the number of coefficients, which is rows()*cols().");
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef __eigenpy_eigen_eigen_base_hpp__

include/eigenpy/eigenpy.hpp

 /*
  * Copyright 2014-2019, CNRS
- * Copyright 2018-2023, INRIA
+ * Copyright 2018-2024, INRIA
  */
 
 #ifndef __eigenpy_eigenpy_hpp__
@@ -10,6 +10,9 @@
 #include "eigenpy/eigen-typedef.hpp"
 #include "eigenpy/expose.hpp"
 
+/// Custom CallPolicies
+#include "eigenpy/std-unique-ptr.hpp"
+
 #define ENABLE_SPECIFIC_MATRIX_TYPE(TYPE) \
   ::eigenpy::enableEigenPySpecific<TYPE>();
 
@@ -19,6 +22,8 @@ namespace eigenpy {
  */
 void EIGENPY_DLLAPI enableEigenPy();
 
+bool EIGENPY_DLLAPI withTensorSupport();
+
 /* Enable the Eigen--Numpy serialization for the templated MatType class.*/
 template <typename MatType>
 void enableEigenPySpecific();
@@ -52,6 +57,8 @@ EIGEN_DONT_INLINE void exposeType() {
   ENABLE_SPECIFIC_MATRIX_TYPE(VectorXs);
   ENABLE_SPECIFIC_MATRIX_TYPE(RowVectorXs);
   ENABLE_SPECIFIC_MATRIX_TYPE(MatrixXs);
+
+  enableEigenPySpecific<SparseMatrixXs>();
 }
 
 template <typename Scalar>

include/eigenpy/fwd.hpp

 /*
- * Copyright 2014-2023 CNRS INRIA
+ * Copyright 2014-2024 CNRS INRIA
  */
 
 #ifndef __eigenpy_fwd_hpp__
@@ -43,9 +43,8 @@
   EIGENPY_PRAGMA_WARNING(Deprecated : the_message)
 #define EIGENPY_PRAGMA_DEPRECATED_HEADER(old_header, new_header) \
   EIGENPY_PRAGMA_WARNING(                                        \
-      Deprecated header file                                     \
-      : #old_header has been replaced                            \
-            by #new_header.\n Please use #new_header instead of #old_header.)
+      Deprecated header file : #old_header has been replaced     \
+          by #new_header.\n Please use #new_header instead of #old_header.)
 #elif defined(WIN32)
 #define EIGENPY_PRAGMA(x) __pragma(#x)
 #define EIGENPY_PRAGMA_MESSAGE(the_message) \
@@ -65,6 +64,7 @@
 #define EIGENPY_DOCUMENTATION_END_IGNORE    /// \endcond
 
 #include "eigenpy/config.hpp"
+#include <boost/type_traits/is_base_of.hpp>
 
 // Silence a warning about a deprecated use of boost bind by boost python
 // at least fo boost 1.73 to 1.75
@@ -73,6 +73,9 @@
 #include <boost/python.hpp>
 #include <boost/python/scope.hpp>
 
+#include <type_traits>
+#include <utility>
+
 namespace eigenpy {
 
 namespace bp = boost::python;
@@ -86,6 +89,7 @@ namespace bp = boost::python;
 #undef BOOST_BIND_GLOBAL_PLACEHOLDERS
 
 #include <Eigen/Core>
+#include <Eigen/Sparse>
 #include <Eigen/Geometry>
 
 #ifdef EIGENPY_WITH_CXX11_SUPPORT
@@ -105,6 +109,12 @@ namespace bp = boost::python;
 
 #define EIGENPY_UNUSED_VARIABLE(var) (void)(var)
 #define EIGENPY_UNUSED_TYPE(type) EIGENPY_UNUSED_VARIABLE((type *)(NULL))
+#ifndef NDEBUG
+#define EIGENPY_USED_VARIABLE_ONLY_IN_DEBUG_MODE(var)
+#else
+#define EIGENPY_USED_VARIABLE_ONLY_IN_DEBUG_MODE(var) \
+  EIGENPY_UNUSED_VARIABLE(var)
+#endif
 
 #ifdef EIGENPY_WITH_CXX11_SUPPORT
 #include <memory>
@@ -115,13 +125,13 @@ namespace bp = boost::python;
 #endif
 
 namespace eigenpy {
-template <typename MatType,
-          typename Scalar =
-              typename boost::remove_reference<MatType>::type::Scalar>
+
+// Default Scalar value can't be defined in the declaration
+// because of a CL bug.
+// See https://github.com/stack-of-tasks/eigenpy/pull/462
+template <typename MatType, typename Scalar>
 struct EigenToPy;
-template <typename MatType,
-          typename Scalar =
-              typename boost::remove_reference<MatType>::type::Scalar>
+template <typename MatType, typename Scalar>
 struct EigenFromPy;
 
 template <typename T>
@@ -135,17 +145,20 @@ struct get_eigen_base_type {
   typedef typename remove_const_reference<EigenType>::type EigenType_;
   typedef typename boost::mpl::if_<
       boost::is_base_of<Eigen::MatrixBase<EigenType_>, EigenType_>,
-      Eigen::MatrixBase<EigenType_>
-#ifdef EIGENPY_WITH_TENSOR_SUPPORT
-      ,
+      Eigen::MatrixBase<EigenType_>,
       typename boost::mpl::if_<
-          boost::is_base_of<Eigen::TensorBase<EigenType_>, EigenType_>,
-          Eigen::TensorBase<EigenType_>, void>::type
+          boost::is_base_of<Eigen::SparseMatrixBase<EigenType_>, EigenType_>,
+          Eigen::SparseMatrixBase<EigenType_>
+#ifdef EIGENPY_WITH_TENSOR_SUPPORT
+          ,
+          typename boost::mpl::if_<
+              boost::is_base_of<Eigen::TensorBase<EigenType_>, EigenType_>,
+              Eigen::TensorBase<EigenType_>, void>::type
 #else
-      ,
-      void
+          ,
+          void
 #endif
-      >::type _type;
+          >::type>::type _type;
 
   typedef typename boost::mpl::if_<
       boost::is_const<typename boost::remove_reference<EigenType>::type>,
@@ -153,22 +166,39 @@ struct get_eigen_base_type {
 };
 
 template <typename EigenType>
-struct get_eigen_ref_plain_type;
+struct get_eigen_plain_type;
 
 template <typename MatType, int Options, typename Stride>
-struct get_eigen_ref_plain_type<Eigen::Ref<MatType, Options, Stride> > {
+struct get_eigen_plain_type<Eigen::Ref<MatType, Options, Stride> > {
   typedef typename Eigen::internal::traits<
       Eigen::Ref<MatType, Options, Stride> >::PlainObjectType type;
 };
 
 #ifdef EIGENPY_WITH_TENSOR_SUPPORT
 template <typename TensorType>
-struct get_eigen_ref_plain_type<Eigen::TensorRef<TensorType> > {
+struct get_eigen_plain_type<Eigen::TensorRef<TensorType> > {
   typedef TensorType type;
 };
 #endif
+
+namespace internal {
+template <class T1, class T2>
+struct has_operator_equal_impl {
+  template <class U, class V>
+  static auto check(U *) -> decltype(std::declval<U>() == std::declval<V>());
+  template <typename, typename>
+  static auto check(...) -> std::false_type;
+
+  using type = typename std::is_same<bool, decltype(check<T1, T2>(0))>::type;
+};
+}  // namespace internal
+
+template <class T1, class T2 = T1>
+struct has_operator_equal : internal::has_operator_equal_impl<T1, T2>::type {};
+
 }  // namespace eigenpy
 
 #include "eigenpy/alignment.hpp"
+#include "eigenpy/id.hpp"
 
 #endif  // ifndef __eigenpy_fwd_hpp__

include/eigenpy/id.hpp

+//
+// Copyright (c) 2024 INRIA
+//
+
+#ifndef __eigenpy_id_hpp__
+#define __eigenpy_id_hpp__
+
+#include <boost/python.hpp>
+#include <boost/cstdint.hpp>
+
+namespace eigenpy {
+
+///
+/// \brief Add the Python method id to retrieving a unique id for a given object
+/// exposed with Boost.Python
+///
+template <class C>
+struct IdVisitor : public bp::def_visitor<IdVisitor<C> > {
+  template <class PyClass>
+  void visit(PyClass& cl) const {
+    cl.def("id", &id, bp::arg("self"),
+           "Returns the unique identity of an object.\n"
+           "For object held in C++, it corresponds to its memory address.");
+  }
+
+ private:
+  static boost::int64_t id(const C& self) {
+    return boost::int64_t(reinterpret_cast<const void*>(&self));
+  }
+};
+}  // namespace eigenpy
+
+#endif  // ifndef __eigenpy_id_hpp__

include/eigenpy/map.hpp

+/// Copyright (c) 2016-2024 CNRS INRIA
+/// This file was originally taken from Pinocchio (header
+/// <pinocchio/bindings/python/utils/std-vector.hpp>)
+///
+
+#ifndef __eigenpy_map_hpp__
+#define __eigenpy_map_hpp__
+
+#include "eigenpy/pickle-vector.hpp"
+#include "eigenpy/registration.hpp"
+#include "eigenpy/utils/empty-visitor.hpp"
+
+#include <boost/python/suite/indexing/map_indexing_suite.hpp>
+#include <boost/python/stl_iterator.hpp>
+#include <boost/python/to_python_converter.hpp>
+
+namespace eigenpy {
+
+/// \brief Change the behavior of indexing (method __getitem__ in Python).
+/// This is suitable e.g. for container of Eigen matrix objects if you want to
+/// mutate them.
+/// \sa overload_base_get_item_for_std_vector
+template <typename Container>
+struct overload_base_get_item_for_map
+    : public boost::python::def_visitor<
+          overload_base_get_item_for_map<Container> > {
+  typedef typename Container::value_type value_type;
+  typedef typename Container::value_type::second_type data_type;
+  typedef typename Container::key_type key_type;
+  typedef typename Container::key_type index_type;
+
+  template <class Class>
+  void visit(Class& cl) const {
+    cl.def("__getitem__", &base_get_item);
+  }
+
+ private:
+  static boost::python::object base_get_item(
+      boost::python::back_reference<Container&> container, PyObject* i_) {
+    index_type idx = convert_index(container.get(), i_);
+    typename Container::iterator i = container.get().find(idx);
+    if (i == container.get().end()) {
+      PyErr_SetString(PyExc_KeyError, "Invalid key");
+      boost::python::throw_error_already_set();
+    }
+
+    typename boost::python::to_python_indirect<
+        data_type&, boost::python::detail::make_reference_holder>
+        convert;
+    return boost::python::object(boost::python::handle<>(convert(i->second)));
+  }
+
+  static index_type convert_index(Container& /*container*/, PyObject* i_) {
+    boost::python::extract<key_type const&> i(i_);
+    if (i.check()) {
+      return i();
+    } else {
+      boost::python::extract<key_type> i(i_);
+      if (i.check()) return i();
+    }
+
+    PyErr_SetString(PyExc_TypeError, "Invalid index type");
+    boost::python::throw_error_already_set();
+    return index_type();
+  }
+};
+
+///////////////////////////////////////////////////////////////////////////////
+// The following snippet of code has been taken from the header
+// https://github.com/loco-3d/crocoddyl/blob/v2.1.0/bindings/python/crocoddyl/utils/map-converter.hpp
+// The Crocoddyl library is written by Carlos Mastalli, Nicolas Mansard and
+// Rohan Budhiraja.
+///////////////////////////////////////////////////////////////////////////////
+
+namespace bp = boost::python;
+
+/**
+ * @brief Create a pickle interface for the map type
+ *
+ * @param[in] Container  Map type to be pickled
+ * \sa Pickle
+ */
+template <typename Container>
+struct PickleMap : public PickleVector<Container> {
+  static void setstate(bp::object op, bp::tuple tup) {
+    Container& o = bp::extract<Container&>(op)();
+    bp::stl_input_iterator<typename Container::value_type> begin(tup[0]), end;
+    o.insert(begin, end);
+  }
+};
+
+/// Conversion from dict to map solution proposed in
+/// https://stackoverflow.com/questions/6116345/boostpython-possible-to-automatically-convert-from-dict-stdmap
+/// This template encapsulates the conversion machinery.
+template <typename Container>
+struct dict_to_map {
+  static void register_converter() {
+    bp::converter::registry::push_back(&dict_to_map::convertible,
+                                       &dict_to_map::construct,
+                                       bp::type_id<Container>());
+  }
+
+  /// Check if conversion is possible
+  static void* convertible(PyObject* object) {
+    // Check if it is a list
+    if (!PyObject_GetIter(object)) return 0;
+    return object;
+  }
+
+  /// Perform the conversion
+  static void construct(PyObject* object,
+                        bp::converter::rvalue_from_python_stage1_data* data) {
+    // convert the PyObject pointed to by `object` to a bp::dict
+    bp::handle<> handle(bp::borrowed(object));  // "smart ptr"
+    bp::dict dict(handle);
+
+    // get a pointer to memory into which we construct the map
+    // this is provided by the Python runtime
+    typedef bp::converter::rvalue_from_python_storage<Container> storage_type;
+    void* storage = reinterpret_cast<storage_type*>(data)->storage.bytes;
+
+    // placement-new allocate the result
+    new (storage) Container();
+
+    // iterate over the dictionary `dict`, fill up the map `map`
+    Container& map(*(static_cast<Container*>(storage)));
+    bp::list keys(dict.keys());
+    int keycount(static_cast<int>(bp::len(keys)));
+    for (int i = 0; i < keycount; ++i) {
+      // get the key
+      bp::object keyobj(keys[i]);
+      bp::extract<typename Container::key_type> keyproxy(keyobj);
+      if (!keyproxy.check()) {
+        PyErr_SetString(PyExc_KeyError, "Bad key type");
+        bp::throw_error_already_set();
+      }
+      typename Container::key_type key = keyproxy();
+
+      // get the corresponding value
+      bp::object valobj(dict[keyobj]);
+      bp::extract<typename Container::mapped_type> valproxy(valobj);
+      if (!valproxy.check()) {
+        PyErr_SetString(PyExc_ValueError, "Bad value type");
+        bp::throw_error_already_set();
+      }
+      typename Container::mapped_type val = valproxy();
+      map.emplace(key, val);
+    }
+
+    // remember the location for later
+    data->convertible = storage;
+  }
+
+  static bp::dict todict(Container& self) {
+    bp::dict dict;
+    typename Container::const_iterator it;
+    for (it = self.begin(); it != self.end(); ++it) {
+      dict.setdefault(it->first, it->second);
+    }
+    return dict;
+  }
+};
+
+/// Policies which handle the non-default constructible case
+/// and set_item() using emplace().
+template <class Container, bool NoProxy>
+struct emplace_set_derived_policies
+    : bp::map_indexing_suite<
+          Container, NoProxy,
+          emplace_set_derived_policies<Container, NoProxy> > {
+  typedef typename Container::key_type index_type;
+  typedef typename Container::value_type::second_type data_type;
+  typedef typename Container::value_type value_type;
+  using DerivedPolicies =
+      bp::detail::final_map_derived_policies<Container, NoProxy>;
+
+  template <class Class>
+  static void extension_def(Class& cl) {
+    //  Wrap the map's element (value_type)
+    std::string elem_name = "map_indexing_suite_";
+    bp::object class_name(cl.attr("__name__"));
+    bp::extract<std::string> class_name_extractor(class_name);
+    elem_name += class_name_extractor();
+    elem_name += "_entry";
+    namespace mpl = boost::mpl;
+
+    typedef typename mpl::if_<
+        mpl::and_<boost::is_class<data_type>, mpl::bool_<!NoProxy> >,
+        bp::return_internal_reference<>, bp::default_call_policies>::type
+        get_data_return_policy;
+
+    bp::class_<value_type>(elem_name.c_str(), bp::no_init)
+        .def("__repr__", &DerivedPolicies::print_elem)
+        .def("data", &DerivedPolicies::get_data, get_data_return_policy())
+        .def("key", &DerivedPolicies::get_key);
+  }
+
+  static void set_item(Container& container, index_type i, data_type const& v) {
+    container.emplace(i, v);
+  }
+};
+
+/**
+ * @brief Expose the map-like container, e.g. (std::map).
+ *
+ * @param[in] Container  Container to expose.
+ * @param[in] NoProxy    When set to false, the elements will be copied when
+ * returned to Python.
+ */
+template <class Container, bool NoProxy = false>
+struct GenericMapVisitor
+    : public emplace_set_derived_policies<Container, NoProxy>,
+      public dict_to_map<Container> {
+  typedef dict_to_map<Container> FromPythonDictConverter;
+
+  template <typename DerivedVisitor>
+  static void expose(const std::string& class_name,
+                     const std::string& doc_string,
+                     const bp::def_visitor<DerivedVisitor>& visitor) {
+    namespace bp = bp;
+
+    if (!register_symbolic_link_to_registered_type<Container>()) {
+      bp::class_<Container>(class_name.c_str(), doc_string.c_str())
+          .def(GenericMapVisitor())
+          .def("todict", &FromPythonDictConverter::todict, bp::arg("self"),
+               "Returns the map type as a Python dictionary.")
+          .def_pickle(PickleMap<Container>())
+          .def(visitor);
+      // Register conversion
+      FromPythonDictConverter::register_converter();
+    }
+  }
+
+  static void expose(const std::string& class_name,
+                     const std::string& doc_string = "") {
+    expose(class_name, doc_string, EmptyPythonVisitor());
+  }
+
+  template <typename DerivedVisitor>
+  static void expose(const std::string& class_name,
+                     const bp::def_visitor<DerivedVisitor>& visitor) {
+    expose(class_name, "", visitor);
+  }
+};
+
+}  // namespace eigenpy
+
+#endif  // ifndef __eigenpy_map_hpp__

include/eigenpy/numpy-allocator.hpp

@@ -138,7 +138,12 @@ struct numpy_allocator_impl_matrix<Eigen::Ref<MatType, Options, Stride> > {
                         outer_stride = reverse_strides ? mat.innerStride()
                                                        : mat.outerStride();
 
+#if NPY_ABI_VERSION < 0x02000000
       const int elsize = call_PyArray_DescrFromType(Scalar_type_code)->elsize;
+#else
+      const int elsize =
+          PyDataType_ELSIZE(call_PyArray_DescrFromType(Scalar_type_code));
+#endif
       npy_intp strides[2] = {elsize * inner_stride, elsize * outer_stride};
 
       PyArrayObject *pyArray = (PyArrayObject *)call_PyArray_New(
@@ -204,7 +209,12 @@ struct numpy_allocator_impl_matrix<
                         outer_stride = reverse_strides ? mat.innerStride()
                                                        : mat.outerStride();
 
+#if NPY_ABI_VERSION < 0x02000000
       const int elsize = call_PyArray_DescrFromType(Scalar_type_code)->elsize;
+#else
+      const int elsize =
+          PyDataType_ELSIZE(call_PyArray_DescrFromType(Scalar_type_code));
+#endif
       npy_intp strides[2] = {elsize * inner_stride, elsize * outer_stride};
 
       PyArrayObject *pyArray = (PyArrayObject *)call_PyArray_New(

include/eigenpy/numpy-type.hpp

 /*
- * Copyright 2018-2020 INRIA
+ * Copyright 2018-2023 INRIA
  */
 
 #ifndef __eigenpy_numpy_type_hpp__
@@ -17,17 +17,54 @@ namespace eigenpy {
 
 template <typename Scalar>
 bool np_type_is_convertible_into_scalar(const int np_type) {
-  if (static_cast<NPY_TYPES>(NumpyEquivalentType<Scalar>::type_code) >=
-      NPY_USERDEF)
+  const auto scalar_np_code =
+      static_cast<NPY_TYPES>(NumpyEquivalentType<Scalar>::type_code);
+
+  if (scalar_np_code >= NPY_USERDEF)
     return np_type == Register::getTypeCode<Scalar>();
 
-  if (NumpyEquivalentType<Scalar>::type_code == np_type) return true;
+  if (scalar_np_code == np_type) return true;
 
+  // Manage type promotion
   switch (np_type) {
+    case NPY_BOOL:
+      return FromTypeToType<bool, Scalar>::value;
+    case NPY_INT8:
+      return FromTypeToType<int8_t, Scalar>::value;
+    case NPY_INT16:
+      return FromTypeToType<int16_t, Scalar>::value;
+    case NPY_INT32:
+      return FromTypeToType<int32_t, Scalar>::value;
+    case NPY_INT64:
+      return FromTypeToType<int64_t, Scalar>::value;
+    case NPY_UINT8:
+      return FromTypeToType<uint8_t, Scalar>::value;
+    case NPY_UINT16:
+      return FromTypeToType<uint16_t, Scalar>::value;
+    case NPY_UINT32:
+      return FromTypeToType<uint32_t, Scalar>::value;
+    case NPY_UINT64:
+      return FromTypeToType<uint64_t, Scalar>::value;
+
+#if defined _WIN32 || defined __CYGWIN__
+    // Manage NPY_INT on Windows (NPY_INT32 is NPY_LONG).
+    // See https://github.com/stack-of-tasks/eigenpy/pull/455
     case NPY_INT:
-      return FromTypeToType<int, Scalar>::value;
-    case NPY_LONG:
-      return FromTypeToType<long, Scalar>::value;
+      return FromTypeToType<int32_t, Scalar>::value;
+    case NPY_UINT:
+      return FromTypeToType<uint32_t, Scalar>::value;
+#endif  // WIN32
+
+#if defined __APPLE__
+    // Manage NPY_LONGLONG on Mac (NPY_INT64 is NPY_LONG)..
+    // long long and long are both the same type
+    // but NPY_LONGLONG and NPY_LONG are different dtype.
+    // See https://github.com/stack-of-tasks/eigenpy/pull/455
+    case NPY_LONGLONG:
+      return FromTypeToType<int64_t, Scalar>::value;
+    case NPY_ULONGLONG:
+      return FromTypeToType<uint64_t, Scalar>::value;
+#endif  // MAC
     case NPY_FLOAT:
       return FromTypeToType<float, Scalar>::value;
     case NPY_CFLOAT:
@@ -45,54 +82,28 @@ bool np_type_is_convertible_into_scalar(const int np_type) {
   }
 }
 
-enum NP_TYPE { MATRIX_TYPE, ARRAY_TYPE };
-
 struct EIGENPY_DLLAPI NumpyType {
   static NumpyType& getInstance();
 
-  operator bp::object() { return getInstance().CurrentNumpyType; }
-
   static bp::object make(PyArrayObject* pyArray, bool copy = false);
 
   static bp::object make(PyObject* pyObj, bool copy = false);
 
-  static void setNumpyType(bp::object& obj);
-
   static void sharedMemory(const bool value);
 
   static bool sharedMemory();
 
-  static void switchToNumpyArray();
-
-  static void switchToNumpyMatrix();
-
-  static NP_TYPE& getType();
-
-  static bp::object getNumpyType();
-
-  static const PyTypeObject* getNumpyMatrixType();
-
   static const PyTypeObject* getNumpyArrayType();
 
-  static bool isMatrix();
-
-  static bool isArray();
-
  protected:
   NumpyType();
 
-  bp::object CurrentNumpyType;
   bp::object pyModule;
 
   // Numpy types
-  bp::object NumpyMatrixObject;
-  PyTypeObject* NumpyMatrixType;
-  // bp::object NumpyAsMatrixObject; PyTypeObject * NumpyAsMatrixType;
   bp::object NumpyArrayObject;
   PyTypeObject* NumpyArrayType;
 
-  NP_TYPE np_type;
-
   bool shared_memory;
 };
 }  // namespace eigenpy

include/eigenpy/numpy.hpp

 /*
- * Copyright 2020-2022 INRIA
+ * Copyright 2020-2024 INRIA
  */
 
 #ifndef __eigenpy_numpy_hpp__
 #define __eigenpy_numpy_hpp__
 
-#include "eigenpy/fwd.hpp"
+#include "eigenpy/config.hpp"
 
 #ifndef PY_ARRAY_UNIQUE_SYMBOL
 #define PY_ARRAY_UNIQUE_SYMBOL EIGENPY_ARRAY_API
 #endif
 
+// For compatibility with Numpy 2.x. See:
+// https://numpy.org/devdocs/reference/c-api/array.html#c.NPY_API_SYMBOL_ATTRIBUTE
+#define NPY_API_SYMBOL_ATTRIBUTE EIGENPY_DLLAPI
+
+// When building with MSVC, Python headers use some pragma operator to link
+// against the Python DLL.
+// Unfortunately, it can link against the wrong build type of the library
+// leading to some linking issue.
+// Boost::Python provides a helper specifically dedicated to selecting the right
+// Python library depending on build type, so let's make use of it.
+// Numpy headers drags Python with them. As a result, it
+// is necessary to include this helper before including Numpy.
+// See: https://github.com/stack-of-tasks/eigenpy/pull/514
+#include <boost/python/detail/wrap_python.hpp>
+
 #include <numpy/numpyconfig.h>
 #ifdef NPY_1_8_API_VERSION
 #define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
 #endif
 
+// Allow compiling against NumPy 1.x and 2.x. See:
+// https://github.com/numpy/numpy/blob/afea8fd66f6bdbde855f5aff0b4e73eb0213c646/doc/source/reference/c-api/array.rst#L1224
+#if NPY_ABI_VERSION < 0x02000000
+#define PyArray_DescrProto PyArray_Descr
+#endif
+
 #include <numpy/ndarrayobject.h>
 #include <numpy/ufuncobject.h>
 
+#if NPY_ABI_VERSION < 0x02000000
+static inline PyArray_ArrFuncs* PyDataType_GetArrFuncs(PyArray_Descr* descr) {
+  return descr->f;
+}
+#endif
+
+/* PEP 674 disallow using macros as l-values
+   see : https://peps.python.org/pep-0674/
+*/
+#if PY_VERSION_HEX < 0x030900A4 && !defined(Py_SET_TYPE)
+static inline void _Py_SET_TYPE(PyObject* o, PyTypeObject* type) {
+  Py_TYPE(o) = type;
+}
+#define Py_SET_TYPE(o, type) _Py_SET_TYPE((PyObject*)(o), type)
+#endif
+
 #if defined _WIN32 || defined __CYGWIN__
 #define EIGENPY_GET_PY_ARRAY_TYPE(array) \
   call_PyArray_MinScalarType(array)->type_num
@@ -26,68 +63,140 @@
 #define EIGENPY_GET_PY_ARRAY_TYPE(array) PyArray_MinScalarType(array)->type_num
 #endif
 
+#include <complex>
+
 namespace eigenpy {
 void EIGENPY_DLLAPI import_numpy();
 int EIGENPY_DLLAPI PyArray_TypeNum(PyTypeObject* type);
 
 // By default, the Scalar is considered as a Python object
-template <typename Scalar>
+template <typename Scalar, typename Enable = void>
 struct NumpyEquivalentType {
   enum { type_code = NPY_USERDEF };
 };
 
 template <>
-struct NumpyEquivalentType<float> {
-  enum { type_code = NPY_FLOAT };
+struct NumpyEquivalentType<bool> {
+  enum { type_code = NPY_BOOL };
 };
+
 template <>
-struct NumpyEquivalentType<std::complex<float> > {
-  enum { type_code = NPY_CFLOAT };
+struct NumpyEquivalentType<char> {
+  enum { type_code = NPY_INT8 };
 };
 template <>
-struct NumpyEquivalentType<double> {
-  enum { type_code = NPY_DOUBLE };
+struct NumpyEquivalentType<unsigned char> {
+  enum { type_code = NPY_UINT8 };
 };
 template <>
-struct NumpyEquivalentType<std::complex<double> > {
-  enum { type_code = NPY_CDOUBLE };
+struct NumpyEquivalentType<int8_t> {
+  enum { type_code = NPY_INT8 };
 };
+
 template <>
-struct NumpyEquivalentType<long double> {
-  enum { type_code = NPY_LONGDOUBLE };
+struct NumpyEquivalentType<int16_t> {
+  enum { type_code = NPY_INT16 };
 };
 template <>
-struct NumpyEquivalentType<std::complex<long double> > {
-  enum { type_code = NPY_CLONGDOUBLE };
+struct NumpyEquivalentType<uint16_t> {
+  enum { type_code = NPY_UINT16 };
 };
+
 template <>
-struct NumpyEquivalentType<bool> {
-  enum { type_code = NPY_BOOL };
+struct NumpyEquivalentType<int32_t> {
+  enum { type_code = NPY_INT32 };
+};
+template <>
+struct NumpyEquivalentType<uint32_t> {
+  enum { type_code = NPY_UINT32 };
+};
+
+// On Windows, long is a 32 bytes type but it's a different type than int
+// See https://github.com/stack-of-tasks/eigenpy/pull/455
+#if defined _WIN32 || defined __CYGWIN__
+
+template <>
+struct NumpyEquivalentType<long> {
+  enum { type_code = NPY_INT32 };
+};
+template <>
+struct NumpyEquivalentType<unsigned long> {
+  enum { type_code = NPY_UINT32 };
 };
+
+#endif  // WIN32
+
 template <>
-struct NumpyEquivalentType<int> {
-  enum { type_code = NPY_INT };
+struct NumpyEquivalentType<int64_t> {
+  enum { type_code = NPY_INT64 };
 };
 template <>
-struct NumpyEquivalentType<unsigned int> {
-  enum { type_code = NPY_UINT };
+struct NumpyEquivalentType<uint64_t> {
+  enum { type_code = NPY_UINT64 };
 };
+
+// On Mac, long is a 64 bytes type but it's a different type than int64_t
+// See https://github.com/stack-of-tasks/eigenpy/pull/455
+#if defined __APPLE__
+
 template <>
 struct NumpyEquivalentType<long> {
-  enum { type_code = NPY_LONG };
+  enum { type_code = NPY_INT64 };
 };
-// #if defined _WIN32 || defined __CYGWIN__
 template <>
-struct NumpyEquivalentType<long long> {
+struct NumpyEquivalentType<unsigned long> {
+  enum { type_code = NPY_UINT64 };
+};
+
+#endif  // MAC
+
+// On Linux, long long is a 64 bytes type but it's a different type than int64_t
+// See https://github.com/stack-of-tasks/eigenpy/pull/455
+#if defined __linux__
+
+#include <type_traits>
+
+template <typename Scalar>
+struct NumpyEquivalentType<
+    Scalar,
+    typename std::enable_if<!std::is_same<int64_t, long long>::value &&
+                            std::is_same<Scalar, long long>::value>::type> {
   enum { type_code = NPY_LONGLONG };
 };
-// #else
-//   template <> struct NumpyEquivalentType<long long>    { enum { type_code =
-//   NPY_LONGLONG    };};
-// #endif
+template <typename Scalar>
+struct NumpyEquivalentType<
+    Scalar, typename std::enable_if<
+                !std::is_same<uint64_t, unsigned long long>::value &&
+                std::is_same<Scalar, unsigned long long>::value>::type> {
+  enum { type_code = NPY_ULONGLONG };
+};
+
+#endif  // Linux
+
 template <>
-struct NumpyEquivalentType<unsigned long> {
-  enum { type_code = NPY_ULONG };
+struct NumpyEquivalentType<float> {
+  enum { type_code = NPY_FLOAT };
+};
+template <>
+struct NumpyEquivalentType<double> {
+  enum { type_code = NPY_DOUBLE };
+};
+template <>
+struct NumpyEquivalentType<long double> {
+  enum { type_code = NPY_LONGDOUBLE };
+};
+
+template <>
+struct NumpyEquivalentType<std::complex<float> > {
+  enum { type_code = NPY_CFLOAT };
+};
+template <>
+struct NumpyEquivalentType<std::complex<double> > {
+  enum { type_code = NPY_CDOUBLE };
+};
+template <>
+struct NumpyEquivalentType<std::complex<long double> > {
+  enum { type_code = NPY_CLONGDOUBLE };
 };
 
 template <typename Scalar>
@@ -122,7 +231,7 @@ EIGENPY_DLLAPI PyArray_Descr* call_PyArray_DescrFromType(int typenum);
 
 EIGENPY_DLLAPI void call_PyArray_InitArrFuncs(PyArray_ArrFuncs* funcs);
 
-EIGENPY_DLLAPI int call_PyArray_RegisterDataType(PyArray_Descr* dtype);
+EIGENPY_DLLAPI int call_PyArray_RegisterDataType(PyArray_DescrProto* dtype);
 
 EIGENPY_DLLAPI int call_PyArray_RegisterCanCast(PyArray_Descr* descr,
                                                 int totype,
@@ -170,7 +279,7 @@ inline void call_PyArray_InitArrFuncs(PyArray_ArrFuncs* funcs) {
   PyArray_InitArrFuncs(funcs);
 }
 
-inline int call_PyArray_RegisterDataType(PyArray_Descr* dtype) {
+inline int call_PyArray_RegisterDataType(PyArray_DescrProto* dtype) {
   return PyArray_RegisterDataType(dtype);
 }
 

include/eigenpy/optional.hpp

+///
 /// Copyright (c) 2023 CNRS INRIA
+///
 /// Definitions for exposing boost::optional<T> types.
 /// Also works with std::optional.
 
@@ -7,6 +9,8 @@
 
 #include "eigenpy/fwd.hpp"
 #include "eigenpy/eigen-from-python.hpp"
+#include "eigenpy/registration.hpp"
+
 #include <boost/optional.hpp>
 #ifdef EIGENPY_WITH_CXX17_SUPPORT
 #include <optional>
@@ -35,6 +39,7 @@ struct expected_pytype_for_arg<std::optional<T> > : expected_pytype_for_arg<T> {
 }  // namespace boost
 
 namespace eigenpy {
+
 namespace detail {
 
 /// Helper struct to decide which type is the "none" type for a specific
@@ -56,6 +61,17 @@ struct nullopt_helper<std::optional> {
 };
 #endif
 
+template <typename NoneType>
+struct NoneToPython {
+  static PyObject *convert(const NoneType &) { Py_RETURN_NONE; }
+
+  static void registration() {
+    if (!check_registration<NoneType>()) {
+      bp::to_python_converter<NoneType, NoneToPython, false>();
+    }
+  }
+};
+
 template <typename T,
           template <typename> class OptionalTpl = EIGENPY_DEFAULT_OPTIONAL>
 struct OptionalToPython {
@@ -72,7 +88,9 @@ struct OptionalToPython {
   }
 
   static void registration() {
-    bp::to_python_converter<OptionalTpl<T>, OptionalToPython, true>();
+    if (!check_registration<OptionalTpl<T> >()) {
+      bp::to_python_converter<OptionalTpl<T>, OptionalToPython, true>();
+    }
   }
 };