/*******************************************************************************
 * Copyright (c) 2003 - 2005 University Of British Columbia and others.
 * All rights reserved. This program and the accompanying materials
 * are made available under the terms of the Eclipse Public License v1.0
 * which accompanies this distribution, and is available at
 * http://www.eclipse.org/legal/epl-v10.html
 *
 * Contributors:
 *     University Of British Columbia - initial API and implementation
 *******************************************************************************/
package org.eclipse.mylar.bugzilla.core.internal;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.Reader;
import java.net.URL;
import java.text.ParseException;
import java.util.HashMap;

public class HtmlStreamTokenizer {

	/** parser state */
	private State state;
	/** reader from which to parse the text */
	private BufferedReader in;
	/** base URL for resolving relative URLs */
	private URL base;
	/** buffer holding the text of the current token */
	private StringBuffer textBuffer;
	/** buffer holding whitespace preceding the current token */
	private StringBuffer whitespaceBuffer;
	/** holds a token that was read and then put back in the queue to be returned again on <code>nextToken</code> call */
	private Token pushbackToken;
	/** holds a character that was read and then determined not to be part of the current token */
	private int pushbackChar;
	/** current quote delimiter (single or double) */
	private int quoteChar;

	/**
	 * Constructor.
	 * @param in reader for the HTML document to tokenize
	 * @param base URL for resolving relative URLs
	 */
	public HtmlStreamTokenizer(Reader in, URL base) {
		textBuffer = new StringBuffer();
		whitespaceBuffer = new StringBuffer();
		pushbackChar = 0;
		state = State.TEXT;
		this.in = new BufferedReader(in);
		this.base = base;
	}

	/**
	 * Returns the next token from the stream.
	 */
	public Token nextToken() throws IOException, ParseException {
		if (pushbackToken != null) {
			Token token = pushbackToken;
			pushbackToken = null;
			return token;
		}
		
		int closingComment = 0;
		
		textBuffer.setLength(0);
		whitespaceBuffer.setLength(0);
		do {
			int ch;
			if (pushbackChar != 0) {
				ch = pushbackChar;
				pushbackChar = 0;
			}
			else {
				ch = in.read();
			}
			if (ch < 0) {
				State oldState = state;
				state = State.EOF;
				if (textBuffer.length() > 0 && oldState == State.TEXT) {
					return new Token(textBuffer, whitespaceBuffer, false);
				}
				else {
					return new Token();
				}
			}
			if (state == State.TEXT) {
				if (ch == '<') {
					state = State.TAG;
					if (textBuffer.length() > 0)
						return new Token(textBuffer, whitespaceBuffer, false);
				}
				else if (Character.isWhitespace((char)ch)) {
					pushbackChar = ch;
					state = State.WS;
					if (textBuffer.length() > 0)
						return new Token(textBuffer, whitespaceBuffer, false);
				}
				else {
					textBuffer.append((char) ch);
				}
			}
			else if (state == State.WS) {
				if (!Character.isWhitespace((char)ch)) {
					pushbackChar = ch;
					state = State.TEXT;
				}
				else {
					whitespaceBuffer.append((char) ch);
				}
			}
			else if (state == State.TAG) {
				if (ch == '>') {
					state = State.TEXT;
					HtmlTag tag = new HtmlTag(base);
					parseTag(textBuffer.toString(), tag);
					return new Token(tag, whitespaceBuffer);
				}
				if (ch == '<' && textBuffer.length() == 0) {
					textBuffer.append("<<");
					state = State.TEXT;
				}
				else if (ch == '-' && textBuffer.length() == 2 && textBuffer.charAt(1) == '-' 
				    	&& textBuffer.charAt(0) == '!') {
					textBuffer.setLength(0);
					state = State.COMMENT;
				}
				else if (ch == '\'' || ch == '"') {
					quoteChar = ch;
					textBuffer.append((char) ch);
					state = State.TAG_QUOTE;
				}
				else {
					textBuffer.append((char) ch);
				}
			}
			else if (state == State.TAG_QUOTE) {
				if (ch == '>') {
					pushbackChar = ch;
					state = State.TAG;
				}
				else {
					textBuffer.append((char) ch);
					if (ch == quoteChar)
						state = State.TAG;
				}
			}
			else if (state == State.COMMENT) {
				if (ch == '>' && closingComment >= 2) {
					textBuffer.setLength(textBuffer.length() - 2);
					closingComment = 0;
					state = State.TEXT;
					return new Token(textBuffer, whitespaceBuffer, true);
				}
				if (ch == '-') {
					closingComment++;
				}
				else {
					closingComment = 0;
				}
				textBuffer.append((char) ch);
			}
		} while (true);
	}

	/**
	 * Pushes the token back into the queue, to be returned by the subsequent
	 * call to <code>nextToken</code>
	 */
	public void pushback(Token token) {
		pushbackToken = token;
	}
	
	/**
	 * Parses an HTML tag out of a string of characters.
	 */
	private static void parseTag(String s, HtmlTag tag) throws ParseException {
		
		int i = 0;
		for (; i < s.length() && Character.isWhitespace(s.charAt(i)); i++){
			// just move forward
		}
		if (i == s.length())
			throw new ParseException("parse empty tag", 0);
			
		int start = i;
		for (; i < s.length() && !Character.isWhitespace(s.charAt(i)); i++){
			// just move forward
		}
		tag.setTagName(s.substring(start, i));
		
		for (; i < s.length() && Character.isWhitespace(s.charAt(i)); i++){
			// just move forward
		}
		if (i == s.length()) {
			return;
		}
		else {
			parseAttributes(tag, s, i);
			return;
		}
	}

	/**
	 * parses HTML tag attributes from a buffer and sets them in an HtmlTag
	 */
	private static void parseAttributes(HtmlTag tag, String s, int i) throws ParseException {
		while (i < s.length()) {
			// skip whitespace
			while (i < s.length()  && Character.isWhitespace(s.charAt(i)))
				i++;

			if (i == s.length())
				return;
			
			// read the attribute name -- the rule might be looser than the RFC specifies:
			// everything up to a space or an equal sign is included
			int start = i;
			for (; i < s.length() && !Character.isWhitespace(s.charAt(i)) && s.charAt(i) != '='; i++){
				// just move forward
			}
			String attributeName = s.substring(start, i).toLowerCase();
			
			for (; i < s.length() && Character.isWhitespace(s.charAt(i)); i++){
				// just move forward
			}
			if (i == s.length() || s.charAt(i) != '=') {
				// no attribute value
				tag.setAttribute(attributeName, "");
				continue;
			}
			
			// skip whitespace to the start of attribute value
			for (i = i+1; i < s.length() && Character.isWhitespace(s.charAt(i)); i++){
				// just move forward
			}
			if (i == s.length()) return;

			// read the attribute value -- the rule for unquoted attribute value is
			// looser than the one in Conolly's W3C 1996 lexical analyzer draft: everything
			// is included up to the next space
			String attributeValue;
			if (s.charAt(i) == '"') {
				start = ++i;
				for (; i < s.length() && s.charAt(i) != '"'; i++){
					// just move forward
				}
				if (i == s.length()) return; // shouldn't happen if input returned by nextToken
				attributeValue = unescape(s.substring(start, i));
				i++;
			}
			else if (s.charAt(i) == '\'') {
				start = ++i;
				for (; i < s.length() && s.charAt(i) != '\''; i++){
					// just move forward
				}
				if (i == s.length()) return; // shouldn't happen if input returned by nextToken
				attributeValue = unescape(s.substring(start, i));
				i++;
			}
			else {
				start = i;
				for (; i < s.length() && !Character.isWhitespace(s.charAt(i)); i++){
					// just move forward
				}
				attributeValue = s.substring(start, i);
			}
			tag.setAttribute(attributeName, attributeValue);
		}
	}

	/**
	 * Returns a string with HTML escapes changed into their corresponding characters.
	 */
	public static String unescape(String s) {
		if (s.indexOf('&') == -1) {
			return s;
		}
		else {
			StringBuffer sb = new StringBuffer(s);
			unescape(sb);
			return sb.toString();
		}
	}

	/**
	 * Replaces (in-place) HTML escapes in a StringBuffer with their corresponding characters.
	 */
	public static StringBuffer unescape(StringBuffer sb) {
		int i = 0; // index into the unprocessed section of the buffer
		int j = 0; // index into the processed section of the buffer

		while (i < sb.length()) {
			char ch = sb.charAt(i);
			if (ch == '&') {
				int start = i;
				String escape = null;
				for (i = i+1; i < sb.length(); i++) {
					ch = sb.charAt(i);
					if (!Character.isLetterOrDigit(ch) && !(ch == '#' && i == (start+1))) {
						escape = sb.substring(start+1, i);
						break;
					}
				}
				if (i == sb.length() && i != (start+1)) {
					escape = sb.substring(start + 1);
				}
				if (escape != null) {
					Character character = parseReference(escape);
					if (character != null) {
						ch = character.charValue();
					}
					else {
						// not an HTML escape; rewind
						i = start;
						ch = '&';
					}
				}
			}
			sb.setCharAt(j, ch);
			i++;
			j++;
		}

		sb.setLength(j);
		return sb;
	}

	/**
	 * Parses HTML character and entity references and returns the
	 * corresponding character.
	 */
	private static Character parseReference(String s) {
		if (s.length() == 0)
			return null;
			
		if (s.charAt(0) == '#') {
			// character reference
			if (s.length() == 1)
				return null;

			try {
			    int value;
			    if (s.charAt(1) == 'x') {
			        // Hex reference
			        value = Integer.parseInt(s.substring(2), 16);
			    }
				else {
				    // Decimal reference
				    value = Integer.parseInt(s.substring(1));
				}
				return new Character((char)value);
			} catch (NumberFormatException e) {
			    return null;
			}
		}
		else {
			return entities.get(s);
		}
	}

	/**
	 * Class for current token.
	 */
	public static class Token {
		public static final Type EOF = new Type();
		public static final Type TEXT = new Type();
		public static final Type TAG = new Type();
		public static final Type COMMENT = new Type();
		
		/** token's type */
		private Type type;
		/** token's value */
		private Object value;
		/** whitespace preceding the token */
		private StringBuffer whitespace;
		
		/**
		 * Constructor for the EOF token.
		 */
		protected Token() {
			type = EOF;
			value = null;
			whitespace = null;
		}
		
		/**
		 * Constructor for the HTML tag tokens.
		 */
		protected Token(HtmlTag tag, StringBuffer whitespace) {
			type = TAG;
			value = tag;
			this.whitespace = whitespace;
		}
		
		/**
		 * Constructor for regular text and comments.
		 */
		protected Token(StringBuffer text, StringBuffer whitespace, boolean comment) {
			if (comment) {
				type = COMMENT;
			}
			else {
				type = TEXT;
			}
			this.value = text;
			this.whitespace = whitespace;
		}
		
		/**
		 * Returns the token's type.
		 */
		public Type getType() {
			return type;
		}
		
		/**
		 * Returns the whitespace preceding the token.
		 */
		public StringBuffer getWhitespace() {
			return whitespace;
		}
		
		/**
		 * Returns the token's value. This is an HtmlTag for tokens of type <code>TAG</code>
		 * and a StringBuffer for tokens of type <code>TEXT</code> and <code>COMMENT</code>.
		 * For tokens of type <code>EOF</code>, the value is <code>null</code>. 
		 */
		public Object getValue() {
			return value;
		}
		
		/**
		 * Returns the string representation of the token, including the preceding whitespace.
		 */
		@Override
		public String toString() {
			StringBuffer sb = new StringBuffer();
			if (whitespace != null) {
				sb.append(whitespace);
			}
			if (value != null) {
				if (type == TAG) {
					sb.append('<');
				}
				else if (type == COMMENT) {
					sb.append("<!");
				}
				sb.append(value);
				if (type == TAG) {
					sb.append('>');
				}
				else if (type == COMMENT) {
					sb.append("-->");
				}
			}
			return sb.toString();
		}
		
		/**
		 * Private enum class for token type.
		 */
		private static class Type {
			private Type() {
				// don't need to do anything
			}
		}
	}
	
	/**
	 * Enum class for parser state.
	 */
	private static class State {
		static final State EOF = new State();
		static final State COMMENT = new State();
		static final State TEXT = new State();
		static final State TAG = new State();
		static final State WS = new State();
		static final State TAG_QUOTE = new State();
		
		private State() {
			// don't need to do anything
		}
	}
	
	/** names and values of HTML entity references */
	private static HashMap<String, Character> entities;

	/*
     * Based on ISO 8879.
     * 
     * Portions © International Organization for Standardization 1986 Permission
     * to copy in any form is granted for use with conforming SGML systems and
     * applications as defined in ISO 8879, provided this notice is included in
     * all copies.
     *  
     */
	static {
		entities = new HashMap<String, Character>();
		entities.put(new String("nbsp"), new Character('\240')); // no-break space = non-breaking space 
		entities.put(new String("iexcl"), new Character('\241')); // inverted exclamation mark
		entities.put(new String("cent"), new Character('\242')); // cent sign
		entities.put(new String("pound"), new Character('\243')); // pound sign
		entities.put(new String("curren"), new Character('\244')); // currency sign
		entities.put(new String("yen"), new Character('\245')); // yen sign = yuan sign
		entities.put(new String("brvbar"), new Character('\246')); // broken bar = broken vertical bar
		entities.put(new String("sect"), new Character('\247')); // section sign
		entities.put(new String("uml"), new Character('\250')); // diaeresis = spacing diaeresis
		entities.put(new String("copy"), new Character('\251')); // copyright sign
		entities.put(new String("ordf"), new Character('\252')); // feminine ordinal indicator
		entities.put(new String("laquo"), new Character('\253')); // left-pointing double angle quotation mark = left pointing guillemet
		entities.put(new String("not"), new Character('\254')); // not sign
		entities.put(new String("shy"), new Character('\255')); // soft hyphen = discretionary hyphen
		entities.put(new String("reg"), new Character('\256')); // registered sign = registered trade mark sign
		entities.put(new String("macr"), new Character('\257')); // macron = spacing macron = overline = APL overbar
		entities.put(new String("deg"), new Character('\260')); // degree sign
		entities.put(new String("plusmn"), new Character('\261')); // plus-minus sign = plus-or-minus sign
		entities.put(new String("sup2"), new Character('\262')); // superscript two = superscript digit two = squared
		entities.put(new String("sup3"), new Character('\263')); // superscript three = superscript digit three = cubed
		entities.put(new String("acute"), new Character('\264')); // acute accent = spacing acute
		entities.put(new String("micro"), new Character('\265')); // micro sign
		entities.put(new String("para"), new Character('\266')); // pilcrow sign = paragraph sign
		entities.put(new String("middot"), new Character('\267')); // middle dot = Georgian comma = Greek middle dot
		entities.put(new String("cedil"), new Character('\270')); // cedilla = spacing cedilla
		entities.put(new String("sup1"), new Character('\271')); // superscript one = superscript digit one
		entities.put(new String("ordm"), new Character('\272')); // masculine ordinal indicator
		entities.put(new String("raquo"), new Character('\273')); // right-pointing double angle quotation mark = right pointing guillemet
		entities.put(new String("frac14"), new Character('\274')); // vulgar fraction one quarter = fraction one quarter
		entities.put(new String("frac12"), new Character('\275')); // vulgar fraction one half = fraction one half
		entities.put(new String("frac34"), new Character('\276')); // vulgar fraction three quarters = fraction three quarters
		entities.put(new String("iquest"), new Character('\277')); // inverted question mark = turned question mark
		entities.put(new String("Agrave"), new Character('\300')); // latin capital letter A with grave = latin capital letter A grave
		entities.put(new String("Aacute"), new Character('\301')); // latin capital letter A with acute
		entities.put(new String("Acirc"), new Character('\302')); // latin capital letter A with circumflex
		entities.put(new String("Atilde"), new Character('\303')); // latin capital letter A with tilde
		entities.put(new String("Auml"), new Character('\304')); // latin capital letter A with diaeresis
		entities.put(new String("Aring"), new Character('\305')); // latin capital letter A with ring above = latin capital letter A ring
		entities.put(new String("AElig"), new Character('\306')); // latin capital letter AE = latin capital ligature AE
		entities.put(new String("Ccedil"), new Character('\307')); // latin capital letter C with cedilla
		entities.put(new String("Egrave"), new Character('\310')); // latin capital letter E with grave
		entities.put(new String("Eacute"), new Character('\311')); // latin capital letter E with acute
		entities.put(new String("Ecirc"), new Character('\312')); // latin capital letter E with circumflex
		entities.put(new String("Euml"), new Character('\313')); // latin capital letter E with diaeresis
		entities.put(new String("Igrave"), new Character('\314')); // latin capital letter I with grave
		entities.put(new String("Iacute"), new Character('\315')); // latin capital letter I with acute
		entities.put(new String("Icirc"), new Character('\316')); // latin capital letter I with circumflex
		entities.put(new String("Iuml"), new Character('\317')); // latin capital letter I with diaeresis
		entities.put(new String("ETH"), new Character('\320')); // latin capital letter ETH
		entities.put(new String("Ntilde"), new Character('\321')); // latin capital letter N with tilde
		entities.put(new String("Ograve"), new Character('\322')); // latin capital letter O with grave
		entities.put(new String("Oacute"), new Character('\323')); // latin capital letter O with acute
		entities.put(new String("Ocirc"), new Character('\324')); // latin capital letter O with circumflex
		entities.put(new String("Otilde"), new Character('\325')); // latin capital letter O with tilde
		entities.put(new String("Ouml"), new Character('\326')); // latin capital letter O with diaeresis
		entities.put(new String("times"), new Character('\327')); // multiplication sign
		entities.put(new String("Oslash"), new Character('\330')); // latin capital letter O with stroke = latin capital letter O slash
		entities.put(new String("Ugrave"), new Character('\331')); // latin capital letter U with grave
		entities.put(new String("Uacute"), new Character('\332')); // latin capital letter U with acute
		entities.put(new String("Ucirc"), new Character('\333')); // latin capital letter U with circumflex
		entities.put(new String("Uuml"), new Character('\334')); // latin capital letter U with diaeresis
		entities.put(new String("Yacute"), new Character('\335')); // latin capital letter Y with acute
		entities.put(new String("THORN"), new Character('\336')); // latin capital letter THORN
		entities.put(new String("szlig"), new Character('\337')); // latin small letter sharp s = ess-zed
		entities.put(new String("agrave"), new Character('\340')); // latin small letter a with grave = latin small letter a grave
		entities.put(new String("aacute"), new Character('\341')); // latin small letter a with acute
		entities.put(new String("acirc"), new Character('\342')); // latin small letter a with circumflex
		entities.put(new String("atilde"), new Character('\343')); // latin small letter a with tilde
		entities.put(new String("auml"), new Character('\344')); // latin small letter a with diaeresis
		entities.put(new String("aring"), new Character('\345')); // latin small letter a with ring above = latin small letter a ring
		entities.put(new String("aelig"), new Character('\346')); // latin small letter ae = latin small ligature ae
		entities.put(new String("ccedil"), new Character('\347')); // latin small letter c with cedilla
		entities.put(new String("egrave"), new Character('\350')); // latin small letter e with grave
		entities.put(new String("eacute"), new Character('\351')); // latin small letter e with acute
		entities.put(new String("ecirc"), new Character('\352')); // latin small letter e with circumflex
		entities.put(new String("euml"), new Character('\353')); // latin small letter e with diaeresis
		entities.put(new String("igrave"), new Character('\354')); // latin small letter i with grave
		entities.put(new String("iacute"), new Character('\355')); // latin small letter i with acute
		entities.put(new String("icirc"), new Character('\356')); // latin small letter i with circumflex
		entities.put(new String("iuml"), new Character('\357')); // latin small letter i with diaeresis
		entities.put(new String("eth"), new Character('\360')); // latin small letter eth
		entities.put(new String("ntilde"), new Character('\361')); // latin small letter n with tilde
		entities.put(new String("ograve"), new Character('\362')); // latin small letter o with grave
		entities.put(new String("oacute"), new Character('\363')); // latin small letter o with acute
		entities.put(new String("ocirc"), new Character('\364')); // latin small letter o with circumflex
		entities.put(new String("otilde"), new Character('\365')); // latin small letter o with tilde
		entities.put(new String("ouml"), new Character('\366')); // latin small letter o with diaeresis
		entities.put(new String("divide"), new Character('\367')); // division sign
		entities.put(new String("oslash"), new Character('\370')); // latin small letter o with stroke = latin small letter o slash
		entities.put(new String("ugrave"), new Character('\371')); // latin small letter u with grave
		entities.put(new String("uacute"), new Character('\372')); // latin small letter u with acute
		entities.put(new String("ucirc"), new Character('\373')); // latin small letter u with circumflex
		entities.put(new String("uuml"), new Character('\374')); // latin small letter u with diaeresis
		entities.put(new String("yacute"), new Character('\375')); // latin small letter y with acute
		entities.put(new String("thorn"), new Character('\376')); // latin small letter thorn
		entities.put(new String("yuml"), new Character('\377')); // latin small letter y with diaeresis

		// Special characters
		entities.put(new String("quot"), new Character('\42')); // quotation mark = APL quote
		entities.put(new String("amp"), new Character('\46')); // ampersand
		entities.put(new String("lt"), new Character('\74')); // less-than sign
		entities.put(new String("gt"), new Character('\76')); // greater-than sign
		// Latin Extended-A
		entities.put(new String("OElig"), new Character('\u0152')); // latin capital ligature OE
		entities.put(new String("oelig"), new Character('\u0153')); // latin small ligature oe,  ligature is a misnomer, this is a separate character in some languages
		entities.put(new String("Scaron"), new Character('\u0160')); // latin capital letter S with caron
		entities.put(new String("scaron"), new Character('\u0161')); // latin small letter s with caron
		entities.put(new String("Yuml"), new Character('\u0178')); // latin capital letter Y with diaeresis
		// Spacing Modifier Letters
		entities.put(new String("circ"), new Character('\u02c6')); // modifier letter circumflex accent
		entities.put(new String("tilde"), new Character('\u02dc')); // small tilde
		// General punctuation
		entities.put(new String("ensp"), new Character('\u2002')); // en space
		entities.put(new String("emsp"), new Character('\u2003')); // em space
		entities.put(new String("thinsp"), new Character('\u2009')); // thin space
		entities.put(new String("zwnj"), new Character('\u200c')); // zero width non-joiner
		entities.put(new String("zwj"), new Character('\u200d')); // zero width joiner
		entities.put(new String("lrm"), new Character('\u200e')); // left-to-right mark
		entities.put(new String("rlm"), new Character('\u200f')); // right-to-left mark
		entities.put(new String("ndash"), new Character('\u2013')); // en dash
		entities.put(new String("mdash"), new Character('\u2014')); // em dash
		entities.put(new String("lsquo"), new Character('\u2018')); // left single quotation mark
		entities.put(new String("rsquo"), new Character('\u2019')); // right single quotation mark
		entities.put(new String("sbquo"), new Character('\u201a')); // single low-9 quotation mark
		entities.put(new String("ldquo"), new Character('\u201c')); // left double quotation mark
		entities.put(new String("rdquo"), new Character('\u201d')); // right double quotation mark
		entities.put(new String("bdquo"), new Character('\u201e')); // double low-9 quotation mark
		entities.put(new String("dagger"), new Character('\u2020')); // dagger
		entities.put(new String("Dagger"), new Character('\u2021')); // double dagger
		entities.put(new String("permil"), new Character('\u2030')); // per mille sign
		entities.put(new String("lsaquo"), new Character('\u2039')); // single left-pointing angle quotation mark, not yet standardized
		entities.put(new String("rsaquo"), new Character('\u203a')); // single right-pointing angle quotation mark, not yet standardized
		entities.put(new String("euro"), new Character('\u20ac')); // euro sign
	}
}