[3 of 4] Speedup GetRegistryLengthImpl() by seeding the DAFSA in reverse

net/tools/dafsa/make_dafsa.py

 #!/usr/bin/env python
 # Copyright 2014 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """
 A Deterministic acyclic finite state automaton (DAFSA) is a compact
 representation of an unordered word list (dictionary).
 
 http://en.wikipedia.org/wiki/Deterministic_acyclic_finite_state_automaton
@@ skipping 174 matching lines @@
 
  5: 0x61 <char>         label character 0x61 -> match "a"
  6: 0x61 <char>         label character 0x61 -> match "a"
  7: 0x81 <return_value> 0x81 & 0x0F -> return 1
 
  8: 0x62 <char>         label character 0x62 -> match "b"
  9: 0x62 <char>         label character 0x62 -> match "b"
 10: 0x82 <return_value> 0x82 & 0x0F -> return 2
 """
 
+import argparse
 import sys
 
+
+parser = argparse.ArgumentParser()
+parser.add_argument('infile', nargs='?', type=argparse.FileType('r'))
+parser.add_argument('outfile', nargs='?', type=argparse.FileType('w'))
+parser.add_argument('--reverse_inputs', action='store_true',
+                    help='Generates a DAFSA on reversed input strings; this is '
+                         'useful for suffix queries')
+args = parser.parse_args()
+
+
 class InputError(Exception):
   """Exception raised for errors in the input file."""
 
 
 def to_dafsa(words):
   """Generates a DAFSA from a word list and returns the source node.
 
   Each word is split into characters so that each character is represented by
   a unique node. It is assumed the word list is not empty.
   """
@@ skipping 239 matching lines @@
   end = lines.index('%%', begin)
   lines = lines[begin:end]
   for line in lines:
     if line[-3:-1] != ', ':
       raise InputError('Expected "domainname, <digit>", found "%s"' % line)
     # Technically the DAFSA format can support return values in the range
     # [0-31], but only the first three bits have any defined meaning.
     if not line.endswith(('0', '1', '2', '3', '4', '5', '6', '7')):
       raise InputError('Expected value to be in the range of 0-7, found "%s"' %
                        line[-1])
-  return [line[:-3] + line[-1] for line in lines]
-
+  result = []
+  for line in lines:
+    word, result_code = line[:-3], line[-1]
+    if args.reverse_inputs:
+      word = word[::-1]
+    result.append(word + result_code)
+  result.sort()
+  return result
 
 def main():
-  if len(sys.argv) != 3:
-    print('usage: %s infile outfile' % sys.argv[0])
-    return 1
-  with open(sys.argv[1], 'r') as infile, open(sys.argv[2], 'w') as outfile:
+  with args.infile as infile, args.outfile as outfile:
     outfile.write(words_to_cxx(parse_gperf(infile)))
   return 0
 
 
 if __name__ == '__main__':
   sys.exit(main())