If items are iterable, clustering breaks.

This commit will package tuples into lists before creating the matrix.

Fixes #20

Author: exhuma

cluster/matrix.py

@@ -57,9 +57,9 @@ def worker(self):
         tasks_completed = 0
         for task in iter(self.task_queue.get, 'STOP'):
             col_index, item, item2 = task
-            if not hasattr(item, '__iter__'):
+            if not hasattr(item, '__iter__') or isinstance(item, tuple):
                 item = [item]
-            if not hasattr(item2, '__iter__'):
+            if not hasattr(item2, '__iter__') or isinstance(item2, tuple):
                 item2 = [item2]
             result = (col_index, self.combinfunc(item, item2))
             self.done_queue.put(result)
@@ -123,9 +123,9 @@ def genmatrix(self, num_processes=1):
                         num_tasks_completed += 1
                 else:
                     # Otherwise do it here, in line
-                    if not hasattr(item, '__iter__'):
+                    if not hasattr(item, '__iter__') or isinstance(item, tuple):
                         item = [item]
-                    if not hasattr(item2, '__iter__'):
+                    if not hasattr(item2, '__iter__') or isinstance(item2, tuple):
                         item2 = [item2]
                     row[col_index] = self.combinfunc(item, item2)
 

cluster/test/test_hierarchical.py

@@ -26,6 +26,7 @@
 """
 
 from difflib import SequenceMatcher
+from math import sqrt
 from sys import hexversion
 import unittest
 
@@ -219,6 +220,25 @@ def testUnmodifiedData(self):
         self.assertEqual(sorted(new_data), sorted(self.__data))
 
 
+class HClusterTuplesTestCase(Py23TestCase):
+    '''
+    Test case to cover the case where the data contains tuple-items
+
+    See Github issue #20
+    '''
+
+    def testSingleLinkage(self):
+        "Basic Hierarchical Clustering test with integers"
+
+        def euclidian_distance(a, b):
+            return sqrt(sum([pow(z[0] - z[1], 2) for z in zip(a, b)]))
+
+        self.__data = [(1, 1), (1, 2), (1, 3)]
+        cl = HierarchicalClustering(self.__data, euclidian_distance)
+        result = cl.getlevel(40)
+        self.assertIsNotNone(result)
+
+
 if __name__ == '__main__':
     suite = unittest.TestSuite((
         unittest.makeSuite(HClusterIntegerTestCase),