Experimental parser: dfa minimization

tools/lexer_generator/rule_parser.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 199 matching lines @@
     try:
       parser.parser.parse(data, lexer=parser.lexer.lexer)
     except Exception:
       RuleParser.__static_instance = None
       raise
     parser.__state = None
     assert parser_state.transitions <= set(parser_state.rules.keys())
 
 class RuleProcessor(object):
 
-  def __init__(self, parser_state):
+  def __init__(self, parser_state, minimize_dfa):
     self.__automata = {}
-    self.__process_parser_state(parser_state)
+    self.__process_parser_state(parser_state, minimize_dfa)
 
   @staticmethod
-  def parse(string):
+  def parse(string, minimize_dfa = True):
     parser_state = RuleParserState()
     RuleParser.parse(string, parser_state)
-    return RuleProcessor(parser_state)
+    return RuleProcessor(parser_state, minimize_dfa)
 
   def automata_iter(self):
     return iter(self.__automata.items())
 
   def default_automata(self):
     return self.__automata['default']
 
   def lex(self, string):
     dfa = self.default_automata().dfa()
     return dfa.lex(string)
 
   class Automata(object):
 
-    def __init__(self, nfa):
+    def __init__(self, nfa, minimize_dfa):
       (start, dfa_nodes) = nfa.compute_dfa()
       self.__nfa = nfa
       self.__dfa = Dfa(start, dfa_nodes)
-      self.__minimial_dfa = self.__dfa.minimize()
+      self.__minimial_dfa = self.__dfa.minimize() if minimize_dfa else None
 
     def nfa(self):
       return self.__nfa
 
     def dfa(self):
       return self.__dfa
 
     def minimal_dfa(self):
       return self.__minimial_dfa
 
-  def __process_parser_state(self, parser_state):
+  def __process_parser_state(self, parser_state, minimize_dfa):
     rule_map = {}
     builder = NfaBuilder()
     builder.set_character_classes(parser_state.character_classes)
     assert 'default' in parser_state.rules
     def process(k, v):
       graphs = []
       continues = 0
       for (graph, precedence, code, transition) in v['regex']:
         default_code = v['default_action']
         if code or default_code:
@@ skipping 23 matching lines @@
       graph = NfaBuilder.or_graphs(graphs)
       rule_map[k] = graph
     # process first the subgraphs, then the default graph
     for k, v in parser_state.rules.items():
       if k == 'default': continue
       process(k, v)
     process('default', parser_state.rules['default'])
     # build the automata
     for rule_name, graph in rule_map.items():
       nfa = builder.nfa(graph)
-      self.__automata[rule_name] = RuleProcessor.Automata(nfa)
+      self.__automata[rule_name] = RuleProcessor.Automata(nfa, minimize_dfa)
     self.default_action = parser_state.rules['default']['default_action']