Revision: 19174
Author: [email protected]
Date: Thu Feb 6 19:35:17 2014 UTC
Log: Experimental parser: unify parser construction
[email protected]
BUG=
Review URL: https://codereview.chromium.org/149113010
http://code.google.com/p/v8/source/detail?r=19174
Modified:
/branches/experimental/parser/tools/lexer_generator/regex_parser.py
/branches/experimental/parser/tools/lexer_generator/rule_parser.py
=======================================
--- /branches/experimental/parser/tools/lexer_generator/regex_parser.py Thu
Feb 6 16:38:00 2014 UTC
+++ /branches/experimental/parser/tools/lexer_generator/regex_parser.py Thu
Feb 6 19:35:17 2014 UTC
@@ -31,6 +31,47 @@
from regex_lexer import RegexLexer
from action import Term
+class ParserBuilder:
+
+ class Logger(object):
+ def debug(self,msg,*args,**kwargs):
+ pass
+
+ def info(self,msg,*args,**kwargs):
+ pass
+
+ def warning(self,msg,*args,**kwargs):
+ pass
+ # assert False, "warning: "+ (msg % args) + "\n"
+
+ def error(self,msg,*args,**kwargs):
+ assert False, "error: "+ (msg % args) + "\n"
+
+ __static_instances = {}
+ @staticmethod
+ def parse(
+ string, name, new_lexer, new_parser, preparse = None, postparse =
None):
+ if not name in ParserBuilder.__static_instances:
+ logger = ParserBuilder.Logger()
+ lexer_instance = new_lexer()
+ lexer_instance.lex = lex.lex(module=lexer_instance)
+ instance = new_parser()
+ instance.yacc = yacc.yacc(
+ module=instance, debug=True, write_tables=0,
+ debuglog=logger, errorlog=logger)
+ ParserBuilder.__static_instances[name] = (lexer_instance, instance)
+ (lexer_instance, instance) = ParserBuilder.__static_instances[name]
+ if preparse:
+ preparse(instance)
+ try:
+ return_value = instance.yacc.parse(string, lexer=lexer_instance.lex)
+ except Exception:
+ del ParserBuilder.__static_instances[name]
+ raise
+ if postparse:
+ postparse(instance)
+ return return_value
+
def build_escape_map(chars):
def add_escape(d, char):
d['\\' + char] = char
@@ -95,12 +136,12 @@
def t_CLASS_BEGIN(self, t):
r'\['
- self.lexer.push_state('class')
+ self.lex.push_state('class')
return t
def t_class_CLASS_END(self, t):
r'\]'
- self.lexer.pop_state()
+ self.lex.pop_state()
return t
t_class_RANGE = '-'
@@ -123,12 +164,12 @@
def t_REPEAT_BEGIN(self, t):
r'\{'
- self.lexer.push_state('repeat')
+ self.lex.push_state('repeat')
return t
def t_repeat_REPEAT_END(self, t):
r'\}'
- self.lexer.pop_state()
+ self.lex.pop_state()
return t
t_repeat_NUMBER = r'[0-9]+'
@@ -139,9 +180,6 @@
def t_ANY_error(self, t):
raise Exception("Illegal character '%s'" % t.value[0])
- def build(self, **kwargs):
- self.lexer = lex.lex(module=self, **kwargs)
-
class RegexParser:
tokens = RegexLexer.tokens
@@ -268,21 +306,8 @@
assert left
return left if not right else Term('CAT', left, right)
- def build(self, **kwargs):
- self.parser = yacc.yacc(module=self, debug=0, write_tables=0, **kwargs)
- self.lexer = RegexLexer()
- self.lexer.build(**kwargs)
-
- __static_instance = None
@staticmethod
- def parse(data):
- parser = RegexParser.__static_instance
- if not parser:
- parser = RegexParser()
- parser.build()
- RegexParser.__static_instance = parser
- try:
- return parser.parser.parse(data, lexer=parser.lexer.lexer)
- except Exception:
- RegexParser.__static_instance = None
- raise
+ def parse(string):
+ new_lexer = lambda: RegexLexer()
+ new_parser = lambda: RegexParser()
+ return ParserBuilder.parse(string, "RegexParser", new_lexer,
new_parser)
=======================================
--- /branches/experimental/parser/tools/lexer_generator/rule_parser.py Thu
Feb 6 11:40:10 2014 UTC
+++ /branches/experimental/parser/tools/lexer_generator/rule_parser.py Thu
Feb 6 19:35:17 2014 UTC
@@ -28,7 +28,7 @@
import ply.lex as lex
import ply.yacc as yacc
from action import Term, Action
-from regex_parser import RegexParser
+from regex_parser import RegexParser, ParserBuilder
from nfa_builder import NfaBuilder
from dfa import Dfa
from dfa_optimizer import DfaOptimizer
@@ -96,18 +96,9 @@
t_ACTION_CLOSE = '>'
t_COMMA = ','
- def t_LEFT_BRACKET(self, t):
- r'{'
- self.lexer.push_state('code')
- self.nesting = 1
- return t
-
def t_ANY_error(self, t):
raise Exception("Illegal character '%s'" % t.value[0])
- def build(self, **kwargs):
- self.lexer = lex.lex(module=self, **kwargs)
-
class RuleParserState:
def __init__(self, encoding):
@@ -297,6 +288,7 @@
escape_char = lambda string, char: string.replace(char, "\\" + char)
string = reduce(escape_char, "+?*|.[](){}",
string).replace("\\\"", "\"")
p[0] = RegexParser.parse(string)
+
def p_regex(self, p):
'regex : REGEX'
string = p[1][1:-1].replace("\\/", "/")
@@ -323,26 +315,16 @@
def p_error(self, p):
raise Exception("Syntax error in input '%s'" % str(p))
- def build(self, **kwargs):
- self.parser = yacc.yacc(module=self, debug=0, write_tables=0, **kwargs)
- self.lexer = RuleLexer()
- self.lexer.build(**kwargs)
-
- __static_instance = None
@staticmethod
- def parse(data, parser_state):
- parser = RuleParser.__static_instance
- if not parser:
- parser = RuleParser()
- parser.build()
- RuleParser.__static_instance = parser
- parser.__state = parser_state
- try:
- parser.parser.parse(data, lexer=parser.lexer.lexer)
- except Exception:
- RuleParser.__static_instance = None
- raise
- parser.__state = None
+ def parse(string, parser_state):
+ new_lexer = lambda: RuleLexer()
+ new_parser = lambda: RuleParser()
+ def preparse(parser):
+ parser.__state = parser_state
+ def postparse(parser):
+ parser.__state = None
+ return ParserBuilder.parse(
+ string, "RuleParser", new_lexer, new_parser, preparse, postparse)
class RuleProcessor(object):
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
