[antlr-interest] Simple ASP-like grammar

[Dimitrios Kolovos]

Hi,

I am trying to crate a grammar for a simple language that contains two 
kinds of things: Text inside [% %] and text outside it (something like a 
simple version of ASP or JSP but inside or outside the brackets any type 
of text can exist). An example file that I would like to parse would be:

Text1 [%Text2%] Text3
[%Text4%]

The tree I would expect to have is:
-Program
---Text1 (anyText)
---Text2 (tagged)
---Text3 (anyText)
---Text4 (tagged)

The grammar I have developed is compiled without errors or warnings. 
However the parser returns the whole text as a single child of the root 
AST node. I have tried to find similar grammars on the web but no luck. 
My grammar follows:

header {
package org.xtgl.parse;
import java.util.*;
}

class XtglParser extends Parser;

options {
    k = 2;
    exportVocab = XtglParser;
    buildAST = true;
}

tokens {
    TAGGED;
    ANYTEXT;
    PROGRAM;
}

program
    : (tagged|anyText)*
    {#program = #(#[PROGRAM,"Program"], program);}
    ;

anyText
    : AnyTextBlock^
    {#anyText.setType(ANYTEXT);}
    ;

tagged
    : "[%" AnyTextBlock^ "%]"
    {#tagged.setType(TAGGED);}
    ;

class XtglLexer extends Lexer;

options {
    k = 2;
    exportVocab = XtglParser;
    charVocabulary = '\0' .. '\u00FF';
}

AnyTextBlock
    :
    ('\0' .. '\u00FF')+
    ;

Any ideas?

Cheers,
Dimitrios

-- 
Dimitrios S. Kolovos,
Research Associate,
Department of Computer Science,
The University of York,
Url: http://www-users.cs.york.ac.uk/~dkolovos
Phone: 01904434333