[antlr-interest] plugin, ITokenScanner, nextToken --> SWT BUG

[jabon]

Gavin Lambert a écrit :
> At 19:57 10/04/2008, elekis wrote:
> >Yes I run under windows.  but is it in eclipse or in antlr that I
> >have to patch??
>
> You don't have to patch anything.  Having CRLF line terminators is 
> perfectly normal, and your lexer deals with it properly anyway.  So 
> what's the problem?
>
I found a other thing.
I have my grammar like that

WS                          :(' '|'\r'|'\t'|'\u000C')+; 
RETURN                        :('\n')+;

(I separed the return and space).


on the code
FROM fslkdfjl  fsdlfkjflk IMPORT lfskjfd ;
FROM lfskjd IMPORT dlskjdq;

after the ; I have that

TOKEN -->
<--
14
SIZE: 1
--------------------------------
TOKEN -->
<--
15
******
SIZE: 1


this mean that the first return is considered by the WS lexer and the 
second with RETURN lexer. why I have two return and 2return differrent?????

thanks


a++