Merge pull request #3 from larsmans/master

simple benchmark script

benchmark.py

@@ -0,0 +1,16 @@
+import sys
+from timeit import timeit
+
+from leven import levenshtein
+
+n_runs = 30
+
+with open("/usr/share/dict/words") as f:
+    strs = [ln.strip() for ln in f]
+print("Running %d times on %d reference strings" % (n_runs, len(strs)))
+
+arg = sys.argv[1]
+t = timeit(stmt='for s in strs: levenshtein(arg, s)',
+           setup='from __main__ import arg, levenshtein, strs',
+           number=n_runs)
+print("%.3f seconds" % t)