/*
 * HighlightFilter.java
 *
 * Brazil project web application toolkit,
 * export version: 2.3 
 * Copyright (c) 2007 Sun Microsystems, Inc.
 *
 * Sun Public License Notice
 *
 * The contents of this file are subject to the Sun Public License Version 
 * 1.0 (the "License"). You may not use this file except in compliance with 
 * the License. A copy of the License is included as the file "license.terms",
 * and also available at http://www.sun.com/
 * 
 * The Original Code is from:
 *    Brazil project web application toolkit release 2.3.
 * The Initial Developer of the Original Code is: suhler.
 * Portions created by suhler are Copyright (C) Sun Microsystems, Inc.
 * All Rights Reserved.
 * 
 * Contributor(s): suhler.
 *
 * Version:  1.2
 * Created by suhler on 07/06/21
 * Last modified by suhler on 07/06/22 14:16:34
 *
 * Version Histories:
 *
 * 1.2 07/06/22-14:16:34 (suhler)
 *   - don't filter when there is no RE
 *   - handle encoding properly
 *
 * 1.2 07/06/21-16:21:32 (Codemgr)
 *   SunPro Code Manager data about conflicts, renames, etc...
 *   Name history : 1 0 sunlabs/HighlightFilter.java
 *
 * 1.1 07/06/21-16:21:31 (suhler)
 *   date and time created 07/06/21 16:21:31 by suhler
 *
 */

package sunlabs.brazil.filter;

import java.io.ByteArrayOutputStream;
import java.io.OutputStream;
import java.io.UnsupportedEncodingException;
import java.util.StringTokenizer;
import java.util.Hashtable;
import java.util.Enumeration;
import sunlabs.brazil.server.Request;
import sunlabs.brazil.server.Server;
import sunlabs.brazil.util.Format;
import sunlabs.brazil.util.LexML;
import sunlabs.brazil.util.http.HttpInputStream;
import sunlabs.brazil.util.http.HttpUtil;
import sunlabs.brazil.util.http.MimeHeaders;
import sunlabs.brazil.util.regexp.Regexp;
import sunlabs.brazil.util.http.HttpRequest;
import sunlabs.brazil.handler.HtmlRewriter;

/**
 * Rewrite text between markup (and not in links) based on re sub's.
 * <dl class=props>
 * <dt>sub<dd>The substitution pattern
 * <dt>noCase<dd>true for case insensitive matches.
 * <dt>trailer<dd>Markup to add at end of document (before &lt;/body&gt;)
 * </dl>
 * The values for "re", "sub", and "trailer" are recomputed at each
 * request, and ${...} substitutions are evaluated.
 * <p>
 * Sample usage:
 * <pre>
 * re=test 
 * sub=<font color=red>&</font>
 * </pre>
 * will cause every occurance of the word "test" in the html content
 * to be colored red.
 */

public class HighlightFilter implements Filter {
    String prefix;	// our properties prefix
    String sub;		// the substitution string
    Regexp re;		// the regular expression to match
    String reString;	// the re string
    String trailer;     // text to add at end of all rewritten content
    boolean noCase;

    public boolean
    init(Server server, String prefix) {
	this.prefix = prefix;
	reString = null;
        noCase = (server.props.getProperty(prefix + "noCase") != null);
        sub = server.props.getProperty(prefix + "sub", "");
        trailer = server.props.getProperty(prefix + "trailer");
	return true;
    }

   /*
    * be lazy and don't handle content encodings
    */

    public boolean respond(Request request) {
	String encoding = request.headers.get("accept-encoding");
	if (encoding != null) {
            request.headers.remove("accept-encoding");
	}

        String s = Format.subst(request.props,request.props.getProperty(
		prefix + "re"));
	if (s != null && !s.equals(reString)) {
	    reString = s;
	    try {
		re = new Regexp(reString, noCase);
	    } catch (IllegalArgumentException e) {
		request.log(Server.LOG_WARNING, prefix, "Invalid regexp");
		re=null;
	    }
	}
	return false;
    }

    public boolean
    shouldFilter(Request request, MimeHeaders headers) {
	String type = headers.get("content-type");
	String encoding = headers.get("content-encoding");
	request.log(Server.LOG_DIAGNOSTIC, request.url + ": " + type);
	boolean ok = type!=null && type.startsWith("text/html") && encoding==null;
	return ok;
    }

    public byte[]
    filter(Request request, MimeHeaders headers, byte[] content) {
	HtmlRewriter hr = null;
        String encoding = HttpRequest.getEncoding(headers);
	try {
	    hr = new HtmlRewriter(new String(content, encoding));
	} catch (UnsupportedEncodingException e) {
	    request.log(Server.LOG_WARNING, "bad encoding: " + encoding);
	    encoding = null;
	}
        if (hr == null) {
	    hr = new HtmlRewriter(new String(content));
        }
	request.log(Server.LOG_DIAGNOSTIC, "filtering: " + request.url);
	boolean changed=false;
        boolean skip = false;

	while (hr.nextToken()) {
	    int type = hr.getType();
	    if (type == LexML.TAG) {
	        String tag = hr.getTag(); // lower case tag
		skip = (tag.equals("script") || tag.equals("style") ||
			tag.equals("title"));
		if (trailer!=null && tag.equals("/body")) {
		    request.log(Server.LOG_DIAGNOSTIC, "adding trailer");
		    String append = Format.subst(request.props, trailer);
		    hr.append(append);
		    hr.appendToken();
		    changed = true;
	        }
	    } else if (type==LexML.STRING && !skip && reString!=null &&
			!reString.equals("")) {
		changed |= process(hr);
            }
	}
	if (changed) {
            try {
	        return hr.toString().getBytes(encoding);
            } catch (UnsupportedEncodingException e) {
		request.log(Server.LOG_WARNING, "bad encoding: " + encoding);
		return hr.toString().getBytes();
	    }
        } else {
	    return content;
        }
    }

    /**
     * Process literal text for rewriting
     * hr: current rewrite context
     */

    boolean process(HtmlRewriter hr) {
	String src = hr.getToken();
        String result = re.subAll(src, sub);
	if (result.equals(src)) {
	    return false;
	} else {
	    // System.out.println("Map: " + src + " => " + result);
	    hr.append(result);
	    return true;
	}
    }
}