[antlr-interest] Problem with recursive rules

[Mark]

Hi,

Thank you very much for your reply!

I am still having some very strange behaviour though.  I have the ollowing
lexical rules:

N	:	('0'..'9');
L	:	(('A'..'Z')|('a'..'z'));
IDENTIFIER
:	L ((L|N)*);

However, whenever I generate code I am told that "token definition N is
unreachable" even though it is right there..

Also, AntlrWorks's "interpreter" and "debugger" do not seem to work properly
at all.  The debugger apparantly tries to compile a class called "__test__"
which fails to compile because it cannot find the lexer and parser classes
that Antlr generated.  It may also be because those classes also do not seem
to compile; the Java compiler gives errors like:

[00:07:14] \tmp\antlrworks\tParser.java:1525: orphaned default
[00:07:14]             default();
[00:07:14]             ^
[00:07:14] \tmp\antlrworks\tParser.java:1525: : expected
[00:07:14]             default();
[00:07:14]                    ^
[00:07:14] \tmp\antlrworks\tParser.java:1525: illegal start of expression
[00:07:14]             default();
[00:07:14]                     ^
[00:07:14] \tmp\antlrworks\tParser.java:1593: <identifier> expected
[00:07:14]     public final void default() throws RecognitionException {
[00:07:14]                      ^

Is there any solution for this?

Mark

________________________________________________
Message sent using UebiMiau 2.7.2