Optimize copying nalgebra matrices into NumPy arrays.

Author: adamreichold

src/convert.rs

@@ -183,10 +183,14 @@ where
 
     fn to_pyarray<'py>(&self, py: Python<'py>) -> &'py PyArray<Self::Item, Self::Dim> {
         unsafe {
-            let array = PyArray::new(py, (self.nrows(), self.ncols()), false);
-            for r in 0..self.nrows() {
-                for c in 0..self.ncols() {
-                    *array.uget_mut((r, c)) = self.get_unchecked((r, c)).clone();
+            let array = PyArray::<N, _>::new(py, (self.nrows(), self.ncols()), true);
+            let mut data_ptr = array.data();
+            if self.data.is_contiguous() {
+                ptr::copy_nonoverlapping(self.data.ptr(), data_ptr, self.len());
+            } else {
+                for item in self.iter() {
+                    data_ptr.write(item.clone());
+                    data_ptr = data_ptr.add(1);
                 }
             }
             array

tests/to_py.rs

@@ -303,6 +303,7 @@ fn slice_container_type_confusion() {
 #[test]
 fn matrix_to_numpy() {
     let matrix = nalgebra::Matrix3::<i32>::new(0, 1, 2, 3, 4, 5, 6, 7, 8);
+    assert!(nalgebra::storage::RawStorage::is_contiguous(&matrix.data));
 
     Python::with_gil(|py| {
         let array = matrix.to_pyarray(py);
@@ -312,4 +313,13 @@ fn matrix_to_numpy() {
             array![[0, 1, 2], [3, 4, 5], [6, 7, 8]],
         );
     });
+
+    let matrix = matrix.row(0);
+    assert!(!nalgebra::storage::RawStorage::is_contiguous(&matrix.data));
+
+    Python::with_gil(|py| {
+        let array = matrix.to_pyarray(py);
+
+        assert_eq!(array.readonly().as_array(), array![[0, 1, 2]]);
+    });
 }