Now ATIDLAS is standalone. Everything dynamic....

python/pyatidlas/external/boost/libs/numpy/test/ndarray.py

@@ -0,0 +1,79 @@
+#!/usr/bin/env python
+
+# Copyright Jim Bosch & Ankit Daftery 2010-2012.
+# Distributed under the Boost Software License, Version 1.0.
+#    (See accompanying file LICENSE_1_0.txt or copy at
+#          http://www.boost.org/LICENSE_1_0.txt)
+
+import ndarray_mod
+import unittest
+import numpy
+
+class TestNdarray(unittest.TestCase):
+    
+    def testNdzeros(self):    
+        for dtp in (numpy.int16, numpy.int32, numpy.float32, numpy.complex128):
+            v = numpy.zeros(60, dtype=dtp)
+            dt = numpy.dtype(dtp)
+            for shape in ((60,),(6,10),(4,3,5),(2,2,3,5)):
+                a1 = ndarray_mod.zeros(shape,dt)
+                a2 = v.reshape(a1.shape)
+                self.assertEqual(shape,a1.shape)
+                self.assert_((a1 == a2).all())
+
+    def testNdzeros_matrix(self):    
+        for dtp in (numpy.int16, numpy.int32, numpy.float32, numpy.complex128):
+            dt = numpy.dtype(dtp)
+            shape = (6, 10)
+            a1 = ndarray_mod.zeros_matrix(shape, dt)
+            a2 = numpy.matrix(numpy.zeros(shape, dtype=dtp))
+            self.assertEqual(shape,a1.shape)
+            self.assert_((a1 == a2).all())
+            self.assertEqual(type(a1), type(a2))
+
+    def testNdarray(self):    
+        a = range(0,60)
+        for dtp in (numpy.int16, numpy.int32, numpy.float32, numpy.complex128):
+            v = numpy.array(a, dtype=dtp)
+            dt = numpy.dtype(dtp)
+            a1 = ndarray_mod.array(a)
+            a2 = ndarray_mod.array(a,dt)
+            self.assert_((a1 == v).all())
+            self.assert_((a2 == v).all())
+            for shape in ((60,),(6,10),(4,3,5),(2,2,3,5)):
+                a1 = a1.reshape(shape)
+                self.assertEqual(shape,a1.shape)
+                a2 = a2.reshape(shape)
+                self.assertEqual(shape,a2.shape)
+
+    def testNdempty(self):
+        for dtp in (numpy.int16, numpy.int32, numpy.float32, numpy.complex128):
+            dt = numpy.dtype(dtp)
+            for shape in ((60,),(6,10),(4,3,5),(2,2,3,5)):
+                a1 = ndarray_mod.empty(shape,dt)
+                a2 = ndarray_mod.c_empty(shape,dt)
+                self.assertEqual(shape,a1.shape)
+                self.assertEqual(shape,a2.shape)
+
+    def testTranspose(self):
+        for dtp in (numpy.int16, numpy.int32, numpy.float32, numpy.complex128):
+            dt = numpy.dtype(dtp)
+            for shape in ((6,10),(4,3,5),(2,2,3,5)):
+                a1 = numpy.empty(shape,dt)
+                a2 = a1.transpose()
+                a1 = ndarray_mod.transpose(a1)    
+                self.assertEqual(a1.shape,a2.shape)
+
+    def testSqueeze(self):
+        a1 = numpy.array([[[3,4,5]]])
+        a2 = a1.squeeze()
+        a1 = ndarray_mod.squeeze(a1)
+        self.assertEqual(a1.shape,a2.shape)
+
+    def testReshape(self):
+        a1 = numpy.empty((2,2))
+        a2 = ndarray_mod.reshape(a1,(1,4))
+        self.assertEqual(a2.shape,(1,4))
+
+if __name__=="__main__":
+    unittest.main()