[antlr-interest] Matching Substring In Lexer

[Kunal Sawlani]

Hi John,
I tried out your example and it worked just as expected.
The problem I was having was with the JavaScript target. For some reason,
the statement $type = TEXTNODE, kept reporting the error, TEXTNODE is not
defined. I changed this to use the actual type value(2 in this case) and it
worked!.
I am now confused as to why this is. I know the JavaScript target is new to
ANTLR, but this is something which I guess should be standard across all
platforms. I am not sure if there is any other way to set the type of the
token in the lexer.
Thanks

On Mon, Apr 26, 2010 at 7:20 PM, John B. Brodie <jbb at acm.org> wrote:

> Greetings!
>
> On Mon, 2010-04-26 at 19:02 -0400, Kunal Sawlani wrote:
> > Hi,
> > Thanks to both of you, I have something which works now. However, I am
> > running into a slight issue, where the parser complaints that ANYTHING
> > is not defined, in case of the grammar Jim suggested, and for John's
> > grammar, it complaints that TEXTNODE is not defined. I tried defining
> > these using Fragment, but to no avail. Also, I am using a combined
> > grammar, and will splitting it into a lexer and parser be necessary?
>
>
> In the complete and tested example that I posted there are 2 tokens:
> ANYTHING and TEXTINPUT. TEXTINPUT is defined in a tokens {} option block
> near the top of the file.
>
> As an aside, why did you expect a TEXTNODE token?
>
> In any case, you should be able to run my example through the
> org.antlr.Tool, compile the resultant .java files, execute the parser,
> and then observe the results.
>
>
>
>
>


-- 
Kunal Sawlani