Add a message to the Exception in dense_sandwich

src/tabmat/ext/dense.pyx

@@ -38,7 +38,7 @@ def dense_sandwich(np.ndarray X, floating[:] d, int[:] rows, int[:] cols, int th
     elif X.flags["F_CONTIGUOUS"]:
         _denseF_sandwich(rowsp, colsp, Xp, dp, outp, in_n, out_m, m, n, thresh1d, kratio, innerblock)
     else:
-        raise Exception()
+        raise Exception("The matrix X is not contiguous.")
     return out
 
 

tests/test_fast_sandwich.py

@@ -2,10 +2,13 @@
 import pytest
 import scipy as sp
 import scipy.sparse
+from scipy.sparse import csc_matrix
 
 from tabmat.ext.dense import dense_sandwich
 from tabmat.ext.sparse import sparse_sandwich
 
+from tabmat import DenseMatrix, SparseMatrix, SplitMatrix
+
 
 @pytest.mark.parametrize("dtype", [np.float64, np.float32])
 def test_fast_sandwich_sparse(dtype):
@@ -62,6 +65,28 @@ def test_fast_sandwich_dense():
         check(A, d, cols)
 
 
+def test_dense_sandwich_on_non_contiguous():
+    """Non-regression test for #208"""
+    rng = np.random.default_rng(seed=123)
+    X = rng.standard_normal(size=(100, 20))
+
+    # Xd wraps a not-contiguous array.
+    Xd = DenseMatrix(X[:, :10])
+    Xs = SparseMatrix(csc_matrix(X[:, 10:]))
+    Xm = SplitMatrix([Xd, Xs])
+
+    # Making the sandwich product fail.
+    with pytest.raises(Exception, match="The matrix X is not contiguous"):
+        Xm.sandwich(np.ones(X.shape[0]))
+
+    # Xd wraps a copy, which makes the data contiguous.
+    Xd = DenseMatrix(X[:, :10].copy())
+    Xm = SplitMatrix([Xd, Xs])
+
+    # The sandwich product works without problem here.
+    Xm.sandwich(np.ones(X.shape[0]))
+
+
 def check(A, d, cols):
     Asub = A[:, cols]
     true = (Asub.T.multiply(d)).dot(Asub).toarray()