How to simulate lex lookahead?

[petasis]

Hi all,

I am trying to develop a lexer that was previously written in lex. And I don't know how to handle the following case:

A8: 'Mr' '.' {self.LOOKAHEAD("[A-Z][a-z]+")}?
A27: [A-Za-z]+ '.' [A-Za-z]+

And I have the input string "Mr.George".

The correct tokens should be A8, followed by other lexer tokens. But my lexer grammar matches A27, just because it is a longer match.

The initial lex scanner matches A8, because lookahead is native to the scanner, and is used in the calculation of the match length, and A8 is before A27.

In antlr, I have simulated lookahead in a semantic predicate, but I don't know how to make antlr lexer to select the A8 rule over A27.

[KvanTTT]

Why can't you use something like this?

A8: 'Mr' '.' [A-Za-z]*
A27: [A-Za-z]+ '.' [A-Za-z]+