PARSER_BEGIN(p) 

import java.util.*;
import lib.*;
import IRUtilities.*;

public class Parser {

  public static void main(String args[]) throws ParseException {
    p parser = new p(System.in);
    parser.Input();
  }

}

PARSER_END(p)

SKIP :
{
  " "
| "\t"
| "\n"
| "\r"
}

// LEXICAL SPECIFICATIONS BEGIN HERE
// Parsing rules
// Stuff still needed -- handle commas in numbers
TOKEN :
{
  <DOC_BEGIN: "<DOC>"  >
|
  <DOC_END:  "</DOC>"  >
| 
  <DATE_BEGIN: "<DD>"  >
|
  <DATE_END:   "</DD>" >
|
  <HL_BEGIN:   "<HL>"  >
|
  <HL_END:     "</HL>" >
|
  <TEXT_BEGIN:  "<TEXT>" >
|
  <TEXT_END:  "</TEXT>" >
|
  < Acronym: (["A"-"Z"]) (["A"-"Z"])* >
|
  // Handle U.S
  < Abbreviation: (["A"-"Z"] ".")* >
|
  // Handle single letter followed by dash followed by one or more
  // numbers -- so we get F-16 as a single token
  < Model: ["a"-"z","A"-"Z"] "-" (["0"-"9"])* >
| 
  < Word: ["a"-"z","A"-"Z"] ( ["a"-"z","A"-"Z" ])* >
|
  < Integer:  ["0"-"9"] (["0"-"9"])* >
|
  < Decimal:  (["0"-"9"])* "." (["0"-"9"])+ >
|
  < Other:  ~[] >
}


// Initializes a internally generated docid -- needs work here to 
// allow user to specify a start docid from the command prompt
// Also, we don't want to reinit docid at start of each file

void Input() :
{ 
	int DocId = 0; 
	Document d = new Document(0);
}
{
  	(parseDocument(d)
        {
		d.output();
	  	++DocId;
		d = new Document(DocId);
	}
	)* <EOF>
}

void parseDocument(Document d) :
{
}
{
	(<DOC_BEGIN> 
	  	{
			System.out.println(" ");
			System.out.println("Now Processing a new Document:");
			System.out.println(" ");
		}
	
        header(d) text(d) <DOC_END>)
	{}
}

void text(Document d) :
{
	String t;
	Porter p;
}
{
	(t = term()
	{
		// Lets find the stem of this word 
		p = new Porter(t);
		System.out.println("Stem   --> "+p.getStem());
	        System.out.println();
   		
		// Here we could check for stop words, stem and then
		// add to the document
	}
        )*  <TEXT_END>
	
}


String term():
{  	
	Token t;
}
{
	(
		(t = <Acronym>  
			{
				System.out.println("Acronym--> " +t.image);
			}
	        ) |

		(t = <Abbreviation>  
			{	
				System.out.println("Abbrev --> "+t.image);
			}
                ) |
		(t = <Word>   
			{
				System.out.println("Word   --> "+t.image);
			}
		) |
		(t = <Integer> 
			{
				System.out.println("Integer--> "+t.image);
			}
		) |
		(t = <Decimal> 
			{
				System.out.println("Decimal--> "+t.image);
			}
		) |
		(t = <Model> 
			{
				System.out.println("Model  --> "+t.image);
			}
		) |
		(t = <Other>
			{
				// Nothing do -- just skip it
			}
		) 

	)
	{return(t.image);}

}

void header(Document d):
{
	String hl = "";
	String dl = "";
	String t;
}
{
       (dl = dateline() | hl = headline() | term())* <TEXT_BEGIN> 
       {d.addDateLine(dl);
	d.addHeadLine(hl);
       }

}


String  dateline():
{
	String t;
	String dl = "";
}
{
	<DATE_BEGIN> (t = term() 
			{
				dl = dl + " " + t ;
			}
                     )* <DATE_END>
	{return(dl);}
}

String headline():
{
	String t;
	String hl = "";
}
{
	<HL_BEGIN> (t = term()
			{
				hl = hl + " " + t;
			}
		    )* <HL_END>
	{return(hl);}
}