diff --git a/lark/lexer.py b/lark/lexer.py index 5a8d21a..4d18704 100644 --- a/lark/lexer.py +++ b/lark/lexer.py @@ -186,6 +186,12 @@ class LineCounter: self.column = 1 self.line_start_pos = 0 + def __eq__(self, other): + if not isinstance(other, LineCounter): + return NotImplemented + + return self.char_pos == other.char_pos and self.newline_char == other.newline_char + def feed(self, token, test_newline=True): """Consume a token and calculate the new line & column. @@ -397,7 +403,7 @@ class TraditionalLexer(Lexer): raise EOFError(self) -class LexerState: +class LexerState(object): __slots__ = 'text', 'line_ctr', 'last_token' def __init__(self, text, line_ctr, last_token=None): @@ -405,6 +411,12 @@ class LexerState: self.line_ctr = line_ctr self.last_token = last_token + def __eq__(self, other): + if not isinstance(other, LexerState): + return NotImplemented + + return self.text is other.text and self.line_ctr == other.line_ctr and self.last_token == other.last_token + def __copy__(self): return type(self)(self.text, copy(self.line_ctr), self.last_token) @@ -456,7 +468,7 @@ class ContextualLexer(Lexer): except UnexpectedCharacters: raise e # Raise the original UnexpectedCharacters. The root lexer raises it with the wrong expected set. -class LexerThread: +class LexerThread(object): """A thread that ties a lexer instance and a lexer state, to be used by the parser""" def __init__(self, lexer, text): @@ -465,4 +477,10 @@ class LexerThread: def lex(self, parser_state): return self.lexer.lex(self.state, parser_state) + + def __copy__(self): + copied = object.__new__(LexerThread) + copied.lexer = self.lexer + copied.state = copy(self.state) + return copied ###} diff --git a/lark/parsers/lalr_parser.py b/lark/parsers/lalr_parser.py index 6271bd5..6fc76ea 100644 --- a/lark/parsers/lalr_parser.py +++ b/lark/parsers/lalr_parser.py @@ -94,8 +94,8 @@ class ParserState(object): # Necessary for match_examples() to work def __eq__(self, other): if not isinstance(other, ParserState): - return False - return self.position == other.position + return NotImplemented + return len(self.state_stack) == len(other.state_stack) and self.position == other.position def __copy__(self): return type(self)( diff --git a/tests/test_parser.py b/tests/test_parser.py index ef91a92..3d18b20 100644 --- a/tests/test_parser.py +++ b/tests/test_parser.py @@ -2408,9 +2408,18 @@ def _make_parser_test(LEXER, PARSER): # Skip comma return True elif e.token.type == 'SIGNED_NUMBER': + # Make a copy and ensure it is properly made + puppet_copy = e.puppet.copy() + assert puppet_copy.parser_state == e.puppet.parser_state + assert puppet_copy.lexer_state.state == e.puppet.lexer_state.state + assert puppet_copy.parser_state is not e.puppet.parser_state + assert puppet_copy.lexer_state.state is not e.puppet.lexer_state.state + assert puppet_copy.lexer_state.state.line_ctr is not e.puppet.lexer_state.state.line_ctr + # Try to feed a comma and retry the number e.puppet.feed_token(Token('COMMA', ',')) e.puppet.feed_token(e.token) + return True # Unhandled error. Will stop parse and raise exception