[antlr-interest] strange bug (v 2.7.2)

[Lloyd Dupont]

(I join you a simple failing grammar)

basically I have a simple tokens definition like that:
//--------------------------- lexer.g ----------------
class FMSqlLexer extends Lexer;
options {
k = 3;
charVocabulary = '3' .. '177';
}

NOT_EQ: ("<>") | ("!=") | ("^=") ; 
LE: ("<=") ; 
LessT: "<" ; 
// ---------- end of lexer ----------

and ANTLR find it ambiguous !

what's wrong ?  
I mean which clues should I give to ANTLR to help it understand its errors ?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.antlr.org/pipermail/antlr-interest/attachments/20030702/95654983/attachment.html