Experimental lexer generator: random refactoring + comments.

tools/lexer_generator/nfa.py

 # Copyright 2013 the V8 project authors. All rights reserved.
 # Redistribution and use in source and binary forms, with or without
 # modification, are permitted provided that the following conditions are
 # met:
 #
 #     * Redistributions of source code must retain the above copyright
 #       notice, this list of conditions and the following disclaimer.
 #     * Redistributions in binary form must reproduce the above
 #       copyright notice, this list of conditions and the following
 #       disclaimer in the documentation and/or other materials provided
@@ skipping 80 matching lines @@
     unclosed, self.__unclosed = self.__unclosed, None
     if end == None:
       assert not unclosed
       return
     for key in unclosed:
       self.__add_transition(key, end)
     if not unclosed:
       self.__add_transition(TransitionKey.epsilon(), end)
 
   def __matches(self, match_func, value):
-    f = lambda acc, (k, vs): acc | vs if match_func(k, value) else acc
+    # f collects states whose corresponding TransitionKey matches 'value'.
+    f = (lambda acc, (key, states):
+           acc | states if match_func(key, value) else acc)
     return reduce(f, self.__transitions.items(), set())
 
-  def char_matches(self, value):
+  def next_states_with_char(self, value):
     return self.__matches(lambda k, v : k.matches_char(v), value)
 
   def key_matches(self, value):
     return self.__matches(lambda k, v : k.is_superset_of_key(v), value)
 
 class Nfa(Automaton):
 
   def __init__(self, start, end, nodes_created):
     super(Nfa, self).__init__()
     self.__start = start
@@ skipping 14 matching lines @@
     assert count == nodes_created
 
   @staticmethod
   def __close(states):
     f = lambda acc, node: acc | node.epsilon_closure()
     return reduce(f, states, set(states))
 
   def matches(self, string):
     valid_states = Nfa.__close(set([self.__start]))
     for c in string:
-      f = lambda acc, state: acc | state.char_matches(c)
+      f = lambda acc, state: acc | state.next_states_with_char(c)
       transitions = reduce(f, valid_states, set())
       if not transitions:
         return False
       valid_states = Nfa.__close(transitions)
     return self.__end in valid_states
 
   @staticmethod
   def __gather_transition_keys(state_set):
     keys = set(chain(*map(lambda state: state.key_iter(), state_set)))
     keys.discard(TransitionKey.epsilon())
@@ skipping 31 matching lines @@
       for state in reduce(f, nfa_state_set, set()):
         match_states.add(state)
       transition_state = Nfa.__to_dfa(match_states, dfa_nodes, end_node)
       dfa_nodes[name]['transitions'][key] = transition_state
     return name
 
   def compute_dfa(self):
     dfa_nodes = {}
     start_name = self.__to_dfa(set([self.__start]), dfa_nodes, self.__end)
     return (start_name, dfa_nodes)