Extended StringMatcher with Jaro-Winkler Distance algorithm.

fuzzywuzzy/StringMatcher.py

@@ -18,6 +18,7 @@ class StringMatcher:
     def _reset_cache(self):
         self._ratio = self._distance = None
         self._opcodes = self._editops = self._matching_blocks = None
+        self._jaro_ratio = self._prefix_length = None
 
     def __init__(self, isjunk=None, seq1='', seq2=''):
         if isjunk:
@@ -78,3 +79,19 @@ def distance(self):
         if not self._distance:
             self._distance = distance(self._str1, self._str2)
         return self._distance
+
+    def jaro_winkler_ratio(self):
+        if not self._jaro_ratio:
+            if self._prefix_length:
+                prefix_weight = 1.0 / self._prefix_length
+                self._jaro_ratio = jaro_winkler(self._str1, self._str2, prefix_weight)
+                return self._jaro_ratio
+            self._jaro_ratio = jaro_winkler(self._str1, self._str2)
+        return self._jaro_ratio
+
+    def set_prefix_length(self, common_prefix_length):
+        if common_prefix_length:
+            if isinstance(common_prefix_length, int) and common_prefix_length > 0:
+                self._prefix_length = common_prefix_length
+                return
+        self._prefix_length = None

fuzzywuzzy/fuzz.py

@@ -28,6 +28,17 @@ def ratio(s1, s2):
     return utils.intr(100 * m.ratio())
 
 
+@utils.check_for_none
+@utils.check_for_equivalence
+@utils.check_empty_string
+def jaro_winkler_ratio(s1, s2, common_prefix_length=None):
+    s1, s2 = utils.make_type_consistent(s1, s2)
+
+    m = SequenceMatcher(None, s1, s2)
+    m.set_prefix_length(common_prefix_length)
+    return utils.intr(100 * m.jaro_winkler_ratio())
+
+
 @utils.check_for_none
 @utils.check_for_equivalence
 @utils.check_empty_string

test_fuzzywuzzy.py

@@ -91,6 +91,17 @@ def setUp(self):
         self.s9a = '{a'
         self.s10 = 'a{'
         self.s10a = '{b'
+        # Jaro tests
+        self.s11 = 'initial 10test'
+        self.s11a = 'initial 10 test'
+        self.s12 = 'five test'
+        self.s12a = 'five case'
+        self.s13 = 'seven t'
+        self.s13a = 'seven test'
+        self.s14 = 'Thorkel'
+        self.s14a = 'Thorgier'
+        self.s15 = 'D'
+        self.s15a = 'Dinsdale'
 
         self.cirque_strings = [
             "cirque du soleil - zarkana - las vegas",
@@ -123,6 +134,24 @@ def testCaseInsensitive(self):
     def testPartialRatio(self):
         self.assertEqual(fuzz.partial_ratio(self.s1, self.s3), 100)
 
+    def testJaroWinkler(self):
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s1, self.s1a), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s8, self.s8a), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s9, self.s9a), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s11, self.s11a), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s12, self.s12a, common_prefix_length=5), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s12, self.s12a), 90)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s13, self.s13a, common_prefix_length=7), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s14, self.s14a), 87)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s14, self.s14a, common_prefix_length=4), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s14, self.s14a, common_prefix_length=0), 87)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s14, self.s14a, common_prefix_length=4.2), 87)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s15, self.s15a), 74)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s15, self.s15a, common_prefix_length=1), 100)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s15, self.s15a, common_prefix_length=-1), 74)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s15, self.s15a, common_prefix_length="1"), 74)
+        self.assertEqual(fuzz.jaro_winkler_ratio(self.s15, self.s15a, common_prefix_length=1000), 71)
+
     def testTokenSortRatio(self):
         self.assertEqual(fuzz.token_sort_ratio(self.s1, self.s1a), 100)