Code optimization and UnitTest for performance

Author: babenek

tests/test_patch.py

@@ -4,6 +4,7 @@
 from whatthepatch.patch import Change, diffobj, header as headerobj
 
 
+import time
 import unittest
 import os
 
@@ -1420,6 +1421,27 @@ def test_svn_mixed_line_ends(self):
         results = list(wtp.patch.parse_patch(text))
         self.assertEqual(results[0].header, expected_header)
 
+    def test_huge_path(self):
+        start_time = time.time()
+        text = """diff --git a/huge.file b/huge.file
+index 0000000..1111111 100644
+--- a/huge.file
++++ a/huge.file
+@@ -3,13 +3,1000007 @@
+ 00000000
+ 11111111
+ 22222222
+-33333333
+-44444444
++55555555
++66666666
+"""
+        for n in range(0, 1000000):
+            text += "+" + hex(n) + "\n"
+        result = list(wtp.patch.parse_patch(text))
+        self.assertEqual(1, len(result))
+        self.assertEqual(1000007, len(result[0].changes))
+        self.assertGreater(10, time.time() - start_time)
 
 if __name__ == "__main__":
     unittest.main()

whatthepatch/patch.py

@@ -582,6 +582,8 @@ def parse_unified_diff(text):
     new = 0
     r = 0
     i = 0
+    old_len = 0
+    new_len = 0
 
     changes = list()
 
@@ -609,12 +611,11 @@ def parse_unified_diff(text):
                 h = None
                 break
 
-        while len(hunk) > 0:
-            c = unified_change.match(hunk[0])
+        for n in hunk:
+            c = unified_change.match(n)
             if c:
                 kind = c.group(1)
                 line = c.group(2)
-                c = None
 
                 if kind == "-" and (r != old_len or r == 0):
                     changes.append(Change(old + r, None, line, hunk_n))
@@ -628,8 +629,6 @@ def parse_unified_diff(text):
                     r += 1
                     i += 1
 
-            del hunk[0]
-
     if len(changes) > 0:
         return changes