[antlr-interest] Lexer question (Update on Input from Dominik)

[Micheal J]

> I was looking at ms sql grammar by Tomasz Jastrzebski (on 
> antlr.org) and this is what I found:
> 
> protected
> Integer :;
> 
> protected
> Real :; 
> 
> Number
>     :
>       ( (Digit)+ ('.' | 'e') ) => (Digit)+ ( '.' (Digit)* 
> (Exponent)? |
> Exponent) { _ttype = Real; }
>     | '.' { _ttype = DOT; } ( (Digit)+ (Exponent)? { _ttype = 
> Real; } )?
>     | (Digit)+ { _ttype = Integer; }
>     | "0x" ('a'..'f' | Digit)* { _ttype = HexLiteral; } // 
> "0x" is valid hex literal
>     ;
> 
> k is set to 2.
> 
> I haven't tested it so I don't know.
> Furthermore, since I'm a newbie, I am not even sure what the 
> two Integer and Real rule is supposed to do either.  What 
> does it do?  It's empty?
> 
> Jiho

Interger/Real:
--------------

I'm guessing that the author is trying to define those token types. He
should have used a tokens block e.g.
	tokens
	{
		INTEGER;
		REAL;
		......
	}


Examples of lexing numbers:
---------------------------

The KCSParse/csharp_v1 grammar has a more extensive set of rules for dealing
with numeric tokens.

Find it in the examples/csharp directory of a 2.7.6 distro or on the web
site in file sharing.


Micheal

-----------------------
The best way to contact me is via the list/forum. My time is very limited.