jspwiki-main/src/main/java/org/apache/wiki/parser/CreoleToJSPWikiTranslator.java

/*
    Licensed to the Apache Software Foundation (ASF) under one
    or more contributor license agreements.  See the NOTICE file
    distributed with this work for additional information
    regarding copyright ownership.  The ASF licenses this file
    to you under the Apache License, Version 2.0 (the
    "License"); you may not use this file except in compliance
    with the License.  You may obtain a copy of the License at

       http://www.apache.org/licenses/LICENSE-2.0

    Unless required by applicable law or agreed to in writing,
    software distributed under the License is distributed on an
    "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
    KIND, either express or implied.  See the License for the
    specific language governing permissions and limitations
    under the License.
 */
package org.apache.wiki.parser;

import java.io.UnsupportedEncodingException;
import java.security.MessageDigest;
import java.security.NoSuchAlgorithmException;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.Calendar;
import java.util.HashMap;
import java.util.Map;
import java.util.Properties;
import java.util.Set;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * <p>Translates Creole markp to JSPWiki markup. Simple translator uses regular expressions.
 * See http://www.wikicreole.org for the WikiCreole spec.</p>
 *
 * <p>This translator can be configured through properties defined in
 * jspwiki.properties starting with "creole.*". See the
 * jspwiki.properties file for an explanation of the properties</p>
 *
 * <p><b>WARNING</b>: This is an experimental feature, and known to be
 * broken.  Use at your own risk.</o>
 *
 *
 * @see <a href="http://www.wikicreole.org/">Wiki Creole Spec</a>
 */
public class CreoleToJSPWikiTranslator
{

    // These variables are expanded so that admins
    // can display information about the current installed
    // pagefilter
    //
    // The syntax is the same as a wiki var. Unlike a wiki
    // war though, the CreoleTranslator itself
    //
    // [{$creolepagefilter.version}]
    // [{$creolepagefilter.creoleversion}]
    // [{$creolepagefilter.linebreak}] -> bloglike/wikilike

    /** The version of the filter. */
    public static final String VAR_VERSION = "1.0.3";

    /** The version of Creole that this filter supports. */
    public static final String VAR_CREOLE_VERSION = "1.0";

    /** The linebreak style "bloglike". */
    public static final String VAR_LINEBREAK_BLOGLIKE = "bloglike";

    /** The linebreak style "c2like". */
    public static final String VAR_LINEBREAK_C2LIKE = "c2like";

    private static final String CREOLE_BOLD = "\\*\\*((?s:.)*?)(\\*\\*|(\n\n|\r\r|\r\n\r\n))";

    private static final String JSPWIKI_BOLD = "__$1__$3";

    private static final String CREOLE_ITALIC = "//((?s:.)*?)(//|(\n\n|\r\r|\r\n\r\n))";

    private static final String JSPWIKI_ITALIC = "''$1''$3";

    private static final String CREOLE_SIMPLELINK = "\\[\\[([^\\]]*?)\\]\\]";

    private static final String JSPWIKI_SIMPLELINK = "[$1]";

    private static final String CREOLE_LINK = "\\[\\[([^\\]]*?)\\|([^\\[\\]]*?)\\]\\]";

    private static final String JSPWIKI_LINK = "[$2|$1]";

    private static final String CREOLE_HEADER_0 = "(\n|\r|\r\n|^)=([^=\\r\\n]*)={0,2}";

    private static final String JSPWIKI_HEADER_0 = "$1!!!$2";

    private static final String CREOLE_HEADER_1 = "(\n|\r|\r\n|^)==([^=\\r\\n]*)={0,2}";

    private static final String JSPWIKI_HEADER_1 = "$1!!!$2";

    private static final String CREOLE_HEADER_2 = "(\n|\r|\r\n|^)===([^=\\r\\n]*)={0,3}";

    private static final String JSPWIKI_HEADER_2 = "$1!!$2";

    private static final String CREOLE_HEADER_3 = "(\n|\r|\r\n|^)====([^=\\r\\n]*)={0,4}";

    private static final String JSPWIKI_HEADER_3 = "$1!$2";

    private static final String CREOLE_HEADER_4 = "(\n|\r|\r\n|^)=====([^=\\r\\n]*)={0,5}";

    private static final String JSPWIKI_HEADER_4 = "$1__$2__";

    private static final String CREOLE_SIMPLEIMAGE = "\\{\\{([^\\}]*?)\\}\\}";

    private static final String JSPWIKI_SIMPLEIMAGE = "[{Image src='$1'}]";

    private static final String CREOLE_IMAGE = "\\{\\{([^\\}]*?)\\|([^\\}]*?)\\}\\}";

    private static final String JSPWIKI_IMAGE = "[{Image src='$1' caption='$2'}]";

    private static final String CREOLE_IMAGE_LINK = "\\[\\[(.*?)\\|\\{\\{(.*?)\\}\\}\\]\\]";

    private static final String JSPWIKI_IMAGE_LINK = "[{Image src='$2' link='$1'}]";

    private static final String CREOLE_IMAGE_LINK_DESC = "\\[\\[(.*?)\\|\\{\\{(.*?)\\|(.*?)\\}\\}\\]\\]";

    private static final String JSPWIKI_IMAGE_LINK_DESC = "[{Image src='$2' link='$1' caption='$3'}]";

    private static final String PREFORMATTED_PROTECTED = "\\Q{{{\\E.*?\\Q}}}\\E";

    //private static final String CREOLE_LINEBREAKS = "([^\\s\\\\])(\r\n|\r|\n)+(?=[^\\s\\*#])";

    //private static final String JSPWIKI_LINEBREAKS = "$1\\\\\\\\$2";

    private static final String CREOLE_TABLE = "(\n|\r|\r\n|^)(\\|[^\n\r]*)\\|(\\t| )*(\n|\r|\r\n|$)";

    private static final String CREOLE_PLUGIN = "\\<\\<((?s:.)*?)\\>\\>";

    private static final String JSPWIKI_PLUGIN = "[{$1}]";

    private static final String WWW_URL = "(\\[\\[)\\s*(www\\..*?)(\\]\\])";

    private static final String HTTP_URL = "$1http://$2$3";

    private static final String CREOLE_IMAGE_X = "\\{\\{(.*?)((\\|)(.*?)){0,1}((\\|)(.*?)){0,1}\\}\\}";

    private static final String JSPWIKI_IMAGE_X = "[{\u2016 src='$1' caption='$4' \u2015}]";

    private static final String CREOLE_LINK_IMAG_X = "\\[\\[([^|]*)\\|\\{\\{([^|]*)((\\|)([^|]*)){0,1}((\\|)([^}]*)){0,1}\\}\\}\\]\\]";

    private static final String JSPWIKI_LINK_IMAGE_X = "[{\u2016 src='$2' link='$1' caption='$5' \u2015}]";

    private static final String JSPWIKI_TABLE = "$1$2$4";

    /* TODO Is it possible to use just protect :// ? */
    private static final String URL_PROTECTED = "http://|ftp://|https://";

    private static final String TABLE_HEADER_PROTECTED = "((\n|\r|\r\n|^)(\\|.*?)(\n|\r|\r\n|$))";

    private static final String SIGNATURE = "--~~~";

    private static final String SIGNATURE_AND_DATE = "--~~~~";

    private static final String DEFAULT_DATEFORMAT = "yyyy-MM-dd";

    private static final String ESCAPE_PROTECTED = "~(\\*\\*|~|//|-|#|\\{\\{|}}|\\\\|~\\[~~[|]]|----|=|\\|)";

    private static Map<String, String> c_protectionMap = new HashMap<String, String>();

    private        ArrayList<String> m_hashList = new ArrayList<String>();

    /**
     *  I have no idea what this method does.  Could someone please tell me?
     *
     * @param wikiProps A property set
     * @param content The content to translate?
     * @param username The username in the signature?
     * @return Probably some translated content.
     */
    public String translateSignature(Properties wikiProps, final String content, String username)
    {

        String dateFormat = wikiProps.getProperty("creole.dateFormat");

        if (dateFormat == null)
        {
            dateFormat = DEFAULT_DATEFORMAT;
        }

        SimpleDateFormat df = null;
        try
        {
            df = new SimpleDateFormat(dateFormat);
        }
        catch (Exception e)
        {
            e.printStackTrace();
            df = new SimpleDateFormat(DEFAULT_DATEFORMAT);
        }

        String result = content;
        result = protectMarkup(result, PREFORMATTED_PROTECTED, "", "");
        result = protectMarkup(result, URL_PROTECTED, "", "");

        Calendar cal = Calendar.getInstance();
        result = translateElement(result, SIGNATURE_AND_DATE, "-- [[" + username + "]], " + df.format(cal.getTime()));
        result = translateElement(result, SIGNATURE, "-- [[" + username + "]]");
        result = unprotectMarkup(result, false);
        return result;
    }

    /**
     *  Translates Creole markup to JSPWiki markup
     *
     *  @param wikiProps A set of Wiki Properties
     *  @param content Creole markup
     *  @return Wiki markup
     */
    public String translate(Properties wikiProps, final String content)
    {
        boolean blogLineBreaks = false;
        /*
        // BROKEN, breaks on different platforms.
        String tmp = wikiProps.getProperty("creole.blogLineBreaks");
        if (tmp != null)
        {
            if (tmp.trim().equals("true"))
                blogLineBreaks = true;
        }
        */
        String imagePlugin = wikiProps.getProperty("creole.imagePlugin.name");

        String result = content;
        //
        // Breaks on OSX.  It is never a good idea to tamper with the linebreaks.  JSPWiki always
        // stores linebreaks as \r\n, regardless of the platform.
        //result = result.replace("\r\n", "\n");
        //result = result.replace("\r", "\n");

        /* Now protect the rest */
        result = protectMarkup(result);
        result = translateLists(result, "*", "-", "Nothing");
        result = translateElement(result, CREOLE_BOLD, JSPWIKI_BOLD);
        result = translateElement(result, CREOLE_ITALIC, JSPWIKI_ITALIC);
        result = translateElement(result, WWW_URL, HTTP_URL);

        if (imagePlugin != null && !imagePlugin.equals(""))
        {
            result = this.replaceImageArea(wikiProps, result, CREOLE_LINK_IMAG_X, JSPWIKI_LINK_IMAGE_X, 6, imagePlugin);
            result = this.replaceImageArea(wikiProps, result, CREOLE_IMAGE_X, JSPWIKI_IMAGE_X, 5, imagePlugin);
        }
        result = translateElement(result, CREOLE_IMAGE_LINK_DESC, JSPWIKI_IMAGE_LINK_DESC);
        result = translateElement(result, CREOLE_IMAGE_LINK, JSPWIKI_IMAGE_LINK);
        result = translateElement(result, CREOLE_LINK, JSPWIKI_LINK);
        result = translateElement(result, CREOLE_SIMPLELINK, JSPWIKI_SIMPLELINK);
        result = translateElement(result, CREOLE_HEADER_4, JSPWIKI_HEADER_4);
        result = translateElement(result, CREOLE_HEADER_3, JSPWIKI_HEADER_3);
        result = translateElement(result, CREOLE_HEADER_2, JSPWIKI_HEADER_2);
        result = translateElement(result, CREOLE_HEADER_1, JSPWIKI_HEADER_1);
        result = translateElement(result, CREOLE_HEADER_0, JSPWIKI_HEADER_0);
        result = translateElement(result, CREOLE_IMAGE, JSPWIKI_IMAGE);
        result = translateLists(result, "-", "*", "#");
        result = translateElement(result, CREOLE_SIMPLEIMAGE, JSPWIKI_SIMPLEIMAGE);
        result = translateElement(result, CREOLE_TABLE, JSPWIKI_TABLE);
        result = replaceArea(result, TABLE_HEADER_PROTECTED, "\\|=([^\\|]*)=|\\|=([^\\|]*)", "||$1$2");

        /*
        if (blogLineBreaks)
        {
            result = translateElement(result, CREOLE_LINEBREAKS, JSPWIKI_LINEBREAKS);
        }
        */
        result = unprotectMarkup(result, true);

        result = translateVariables(result, blogLineBreaks);
        //result = result.replace("\n", System.getProperty("line.separator"));
        return result;
    }

    /** Translates lists. */
    private static String translateLists(String content, String sourceSymbol, String targetSymbol, String sourceSymbol2)
    {
        String[] lines = content.split("\n");
        StringBuilder result = new StringBuilder();
        int counter = 0;
        int inList = -1;
        for (int i = 0; i < lines.length; i++)
        {
            String line = lines[i];
            String actSourceSymbol = "";
            while ((line.startsWith(sourceSymbol) || line.startsWith(sourceSymbol2))
                   && (actSourceSymbol.equals("") || line.substring(0, 1).equals(actSourceSymbol)))
            {
                actSourceSymbol = line.substring(0, 1);
                line = line.substring(1, line.length());
                counter++;
            }
            if ((inList == -1 && counter != 1) || (inList != -1 && inList + 1 < counter))
            {
                for (int c = 0; c < counter; c++)
                {
                    result.append(actSourceSymbol);
                }
                inList = -1;
            }
            else
            {
                for (int c = 0; c < counter; c++)
                {
                    if (actSourceSymbol.equals(sourceSymbol2))
                    {
                        result.append(sourceSymbol2);
                    }
                    else
                    {
                        result.append(targetSymbol);
                    }
                }
                inList = counter;
            }
            result.append(line);
            if (i < lines.length - 1)
            {
                result.append("\n");
            }
            counter = 0;
        }

        // Fixes testExtensions5
        if( content.endsWith( "\n" ) && result.charAt( result.length()-1 ) != '\n' )
        {
            result.append( '\n' );
        }

        return result.toString();
    }

    private String translateVariables(String result, boolean blogLineBreaks)
    {
        result = result.replace("[{$creolepagefilter.version}]", VAR_VERSION);
        result = result.replace("[{$creolepagefilter.creoleversion}]", VAR_CREOLE_VERSION);
        String linebreaks = blogLineBreaks ? VAR_LINEBREAK_BLOGLIKE : VAR_LINEBREAK_C2LIKE;
        result = result.replace("[{$creolepagefilter.linebreak}]", linebreaks);
        return result;
    }

    /**
     * Undoes the protection. This is done by replacing the md5 hashes by the
     * original markup.
     *
     * @see #protectMarkup(String)
     */
    private String unprotectMarkup(String content,boolean replacePlugins)
    {
        Object[] it = this.m_hashList.toArray();

        for (int i = it.length - 1; i >= 0; i--)
        {
            String hash = (String) it[i];
            String protectedMarkup = c_protectionMap.get(hash);
            content = content.replace(hash, protectedMarkup);
            if ((protectedMarkup.length() < 3 || (protectedMarkup.length() > 2 &&
                !protectedMarkup.substring(0, 3).equals("{{{")))&&replacePlugins)
                content = translateElement(content, CREOLE_PLUGIN, JSPWIKI_PLUGIN);

        }
        return content;
    }

    /**
     * Protects markup that should not be processed. For now this includes:
     * <ul>
     * <li>Preformatted sections, they should be ignored</li>
     * </li>
     * <li>Protocol strings like <code>http://</code>, they cause problems
     * because of the <code>//</code> which is interpreted as italic</li>
     * </ul>
     * This protection is a simple method to keep the regular expressions for
     * the other markup simple. Internally the protection is done by replacing
     * the protected markup with the the md5 hash of the markup.
     *
     * @param content
     * @return The content with protection
     */
    private String protectMarkup(String content)
    {
        c_protectionMap.clear();
        m_hashList = new ArrayList<String>();
        content = protectMarkup(content, PREFORMATTED_PROTECTED, "", "");
        content = protectMarkup(content, URL_PROTECTED, "", "");
        content = protectMarkup(content, ESCAPE_PROTECTED, "", "");
        content = protectMarkup(content, CREOLE_PLUGIN, "", "");

        // content = protectMarkup(content, LINE_PROTECTED);
        // content = protectMarkup(content, SIGNATURE_PROTECTED);
        return content;
    }

    private ArrayList< String[] > readPlaceholderProperties(Properties wikiProps)
    {
        Set< Object > keySet = wikiProps.keySet();
        Object[] keys = keySet.toArray();
        ArrayList<String[]> result = new ArrayList<String[]>();

        for( int i = 0; i < keys.length; i++ )
        {
            String key = keys[i] + "";
            String value = wikiProps.getProperty( keys[i] + "" );
            if( key.indexOf( "creole.imagePlugin.para.%" ) > -1 )
            {
                String[] pair = new String[2];
                pair[0] = key.replaceAll( "creole.imagePlugin.para.%", "" );
                pair[1] = value;
                result.add( pair );
            }
        }
        return result;
    }

    private String replaceImageArea(Properties wikiProps, String content, String markupRegex, String replaceContent, int groupPos,
                                    String imagePlugin)
    {
        Matcher matcher = Pattern.compile(markupRegex, Pattern.MULTILINE | Pattern.DOTALL).matcher(content);
        String contentCopy = content;

        ArrayList< String[] > plProperties = readPlaceholderProperties(wikiProps);

        while (matcher.find())
        {
            String protectedMarkup = matcher.group(0);
            String paramsField = matcher.group(groupPos);
            String paramsString = "";

            if (paramsField != null)
            {
                String[] params = paramsField.split(",");

                for (int i = 0; i < params.length; i++)
                {
                    String param = params[i].replaceAll("\\||\\s", "").toUpperCase();

                    // Replace placeholder params
                    for (int j = 0; j < plProperties.size(); j++)
                    {
                        String[] pair = plProperties.get(j);
                        String key = pair[0];
                        String value = pair[1];
                        String code = param.replaceAll("(?i)([0-9]+)" + key, value + "<check>" + "$1" + "</check>");
                        code = code.replaceAll("(.*?)%(.*?)<check>(.*?)</check>", "$1$3$2");
                        if (!code.equals(param)) {
                            paramsString += code;
                        }
                    }

                    // Check if it is a number
                    try
                    {
                        Integer.parseInt(param);
                        paramsString += " width='" + param + "px'";
                    }
                    catch (Exception e)
                    {

                        if (wikiProps.getProperty("creole.imagePlugin.para." + param) != null)
                            paramsString += " "
                                            + wikiProps.getProperty("creole.imagePlugin.para." + param)
                                                .replaceAll("^(\"|')(.*)(\"|')$", "$2");
                    }
                }
            }
            String temp = protectedMarkup;

            protectedMarkup = translateElement(protectedMarkup, markupRegex, replaceContent);
            protectedMarkup = protectedMarkup.replaceAll("\u2015", paramsString);
            protectedMarkup = protectedMarkup.replaceAll("\u2016", imagePlugin);
            protectedMarkup = protectedMarkup.replaceAll("caption=''", "");
            protectedMarkup = protectedMarkup.replaceAll("\\s+", " ");

            int pos = contentCopy.indexOf(temp);
            contentCopy = contentCopy.substring(0, pos) + protectedMarkup
                          + contentCopy.substring(pos + temp.length(), contentCopy.length());
        }
        return contentCopy;
    }

    private String replaceArea(String content, String markupRegex, String replaceSource, String replaceTarget)
    {
        Matcher matcher = Pattern.compile(markupRegex, Pattern.MULTILINE | Pattern.DOTALL).matcher(content);
        String contentCopy = content;

        while (matcher.find())
        {
            String protectedMarkup = matcher.group(0);
            String temp = protectedMarkup;
            protectedMarkup = protectedMarkup.replaceAll(replaceSource, replaceTarget);
            int pos = contentCopy.indexOf(temp);
            contentCopy = contentCopy.substring(0, pos) + protectedMarkup
                          + contentCopy.substring(pos + temp.length(), contentCopy.length());
        }
        return contentCopy;
    }

    /**
     * Protects a specific markup
     *
     * @see #protectMarkup(String)
     */
    private String protectMarkup(String content, String markupRegex, String replaceSource, String replaceTarget)
    {
        Matcher matcher = Pattern.compile(markupRegex, Pattern.MULTILINE | Pattern.DOTALL).matcher(content);
        StringBuffer result = new StringBuffer();
        while (matcher.find())
        {
            String protectedMarkup = matcher.group();
            protectedMarkup = protectedMarkup.replaceAll(replaceSource, replaceTarget);
            try
            {
                MessageDigest digest = MessageDigest.getInstance("MD5");
                digest.reset();
                digest.update(protectedMarkup.getBytes("UTF-8"));
                String hash = bytesToHash(digest.digest());
                matcher.appendReplacement(result, hash);
                c_protectionMap.put(hash, protectedMarkup);
                m_hashList.add(hash);
            }
            catch (NoSuchAlgorithmException e)
            {
                // FIXME: Should log properly
                e.printStackTrace();
            }
            catch (UnsupportedEncodingException e)
            {
                // FIXME: Auto-generated catch block
                e.printStackTrace();
            }
        }
        matcher.appendTail(result);
        return result.toString();
    }

    private String bytesToHash(byte[] b)
    {
        String hash = "";
        for (int i = 0; i < b.length; i++)
        {
            hash += Integer.toString((b[i] & 0xff) + 0x100, 16).substring(1);
        }
        return hash;
    }

    private String translateElement(String content, String fromMarkup, String toMarkup)
    {
        Matcher matcher = Pattern.compile(fromMarkup, Pattern.MULTILINE).matcher(content);
        StringBuffer result = new StringBuffer();

        while (matcher.find())
        {
            matcher.appendReplacement(result, toMarkup);
        }
        matcher.appendTail(result);
        return result.toString();
    }
}