[antlr-interest] Python Parse problem

[Michael O'Keeffe]

Hi all,
I'm fairly new to ANTLR (about  half a day of playing). I've been
mucking around with the CSV example test driven development but as a
python version. The generated lexer works fine but the parse just dies
when asked to parse. It produces the following error message:

---------------------------------------------
Traceback (most recent call last):
  File "CSVParser.py", line 272, in <module>
    main(sys.argv)
  File "CSVParser.py", line 268, in main
    main.execute(argv)
  File "c:\apps\Python27\lib\site-packages\antlr_python_runtime-3.1.3-py2.7.egg\
antlr3\main.py", line 163, in execute
    self.parseStream(options, inStream)
  File "c:\apps\Python27\lib\site-packages\antlr_python_runtime-3.1.3-py2.7.egg\
antlr3\main.py", line 236, in parseStream
    result = getattr(parser, options.parserRule)()
TypeError: getattr(): attribute name must be string
---------------------------------------------

My grammar is short so here it is...
---------------------------------------------
grammar CSV;
options {
    language = Python;
}
line returns [result]
scope { fields }
@init { $line::fields = []; }
     : ( (NEWLINE) => NEWLINE
         | field (SEPARATOR field)* NEWLINE
       )
       { $result = $line::fields; }
    ;
field     :
  ( f=QUOTED
  | f=UNQUOTED
  | // nothing
  )
  { $line::fields.append( $f.getText() ); }
  ;
QUOTE :    '"';
QUOTED :     ( QUOTE ( options{ greedy=false; }: . )* QUOTE );
UNQUOTED :    NONBREAKING*;
SEPARATOR :    ',';

NEWLINE    :    '\r'? '\n';
fragment NONBREAKING
    :    ~( '\r' | '\n' | SEPARATOR );
---------------------------------------------

Hope I've just made a n00b mistake. Any advice you can provide will be
greatly appreciated.

Regards,

Michael