Merge pull request #39 from sbinq/implement-group-by

Implement traditional group-by as found in most other languages like clojure/scala/ruby/etc, thanks to @sbinq

fn/iters.py

@@ -100,6 +100,16 @@ def grouper(n, iterable, fillvalue=None):
     args = [iter(iterable)] * n
     return zip_longest(*args, fillvalue=fillvalue)
 
+def group_by(keyfunc, iterable):
+    """Returns a dict of the elements from given iterable keyed by result
+    of keyfunc on each element. The value at each key will be a list of
+    the corresponding elements, in the order they appeared in the iterable.
+    """
+    grouped = {}
+    for item in iterable:
+        grouped.setdefault(keyfunc(item), []).append(item)
+    return grouped
+
 def roundrobin(*iterables):
     """roundrobin('ABC', 'D', 'EF') --> A D E B F C
     Recipe originally credited to George Sakkis.

tests.py

@@ -490,6 +490,19 @@ def test_grouper(self):
         self.assertEqual(["D","E","F"], list(b))
         self.assertEqual(["G","x","x"], list(c))
 
+    def test_group_by(self):
+        # verify grouping logic
+        grouped = iters.group_by(len, ['1', '12', 'a', '123', 'ab'])
+        self.assertEqual({1: ['1', 'a'], 2: ['12', 'ab'], 3: ['123']}, grouped)
+
+        # verify it works with any iterable - not only lists
+        def gen():
+            yield '1'
+            yield '12'
+
+        grouped = iters.group_by(len, gen())
+        self.assertEqual({1: ['1'], 2: ['12']}, grouped)
+
     def test_roundrobin(self):
         r = iters.roundrobin('ABC', 'D', 'EF')
         self.assertEqual(["A","D","E","B","F","C"], list(r))