/**
 * MailArchiver is an application that provides services for storing and managing e-mail messages through a Web Services SOAP interface.
 * Copyright (C) 2012  Marcio Andre Scholl Levien and Fernando Alberto Reuter Wendt and Jose Ronaldo Nogueira Fonseca Junior
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU Affero General Public License as
 * published by the Free Software Foundation, either version 3 of the
 * License, or (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU Affero General Public License for more details.
 *
 * You should have received a copy of the GNU Affero General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */

/******************************************************************************\
*
*  This product was developed by
*
*        SERVIÇO FEDERAL DE PROCESSAMENTO DE DADOS (SERPRO),
*
*  a government company established under Brazilian law (5.615/70),
*  at Department of Development of Porto Alegre.
*
\******************************************************************************/

package serpro.mailarchiver.domain.metaarchive;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.nio.charset.Charset;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

import javax.jdo.JDOHelper;
import javax.jdo.annotations.NotPersistent;
import javax.jdo.annotations.PersistenceCapable;

import net.htmlparser.jericho.Element;
import net.htmlparser.jericho.HTMLElementName;
import net.htmlparser.jericho.Source;
import net.htmlparser.jericho.Tag;

import org.apache.commons.lang3.StringEscapeUtils;
import org.apache.commons.lang3.mutable.MutableBoolean;

import org.apache.james.mime4j.util.CharsetUtil;

import serpro.mailarchiver.util.BodyVisitor;
import serpro.mailarchiver.util.Charsets;
import serpro.mailarchiver.util.Logger;

@PersistenceCapable
public class TextBody
    extends SingleBody
{
    @NotPersistent
    private static final Logger log = Logger.getLocalLogger();

    @NotPersistent private static final String CR = "\015";
    @NotPersistent private static final String LF = "\012";
    @NotPersistent private static final String CRLF = "\015\012";

    //**** P E R S I S T E N T ****
    private String preview;
    //*****************************

    public final String getPreview() {
        return preview;
    }

    public final void setPreview(String preview) {
        this.preview = preview;
    }

    //--------------------------------------------------------------------------
    @Override
    final String toString(String pad) {
        return String.format(
                "TextBody%n"
              + "%1$sjdoState: %2$s%n"
              + "%1$soid: %3$s%n"
              + "%1$shash: %4$x%n"
              + "%1$soffset: %5$d%n"
              + "%1$slength: %6$d%n"
              + "%1$ssize: %7$d%n"
              + "%1$spreview: %8$s"
              , pad
              , JDOHelper.getObjectState(this)
              , getOid()
              , hashCode()
              , getOffset()
              , getLength()
              , getSize()
              , getPreview());
    }

    //--------------------------------------------------------------------------
    public String getText() throws IOException {
        Entity entity = getEntity();
        if(entity != null) {
            ContentTypeField contentTypeField = entity.getContentTypeField();

            Charset cs = null;
            if(contentTypeField != null) {
                cs = CharsetUtil.lookup(contentTypeField.getCharset());
            }
            if(cs == null) {
                cs = Charsets.Windows_1252;
            }

            InputStream is = getDecoderInputStream();
            InputStreamReader isr = new InputStreamReader(is, cs);
            BufferedReader reader = new BufferedReader(isr);

            StringBuilder sb = new StringBuilder();

            char[] cbuf = new char[0x1000];
            int len;
            while((len = reader.read(cbuf)) > 0) {
                sb.append(cbuf, 0, len);
            }

            reader.close();
            isr.close();
            is.close();

            return sb.toString();
        }
        throw new IllegalStateException();
    }

    public String getAdaptedText() throws IOException {
        String text = getText();
        Entity entity = getEntity();
        if(entity != null) {
            ContentTypeField contentTypeField = entity.getContentTypeField();
            if((contentTypeField == null) || (contentTypeField.isTextPlainMimeType())) {
                text = text.replaceAll(CRLF + "|" + CR + "|" + LF, "<br>" + LF);
            }
        }

        text = StringEscapeUtils.unescapeHtml4(text);
        text = replaceMailToRef(text);
        text = replaceContentIdRef(text);
        text = replaceUrlRef(text);
        return text;
    }

    //--------------------------------------------------------------------------
    @NotPersistent
    private static final String contentIdRegex = "src\\s*=(\\.*(?!cid))(" +
            "'\\s*cid:\\s*([^'\\s<>]*)\\s*'" +
            "|" +
            "\"\\s*cid:\\s*([^\"\\s<>]*)\\s*\"" +
            "|" +
            "cid:\\s*([^\\s<>]+)" +
            ")";
    @NotPersistent
    private static final Pattern contentIdPattern = Pattern.compile(contentIdRegex, Pattern.CASE_INSENSITIVE);

    private String replaceContentIdRef(String text) {
        Matcher m = contentIdPattern.matcher(text);
        final StringBuilder sb = new StringBuilder();
        int lastEnd = 0;
        while(m.find()) {
            sb.append(text.substring(lastEnd, m.start()))
              .append("src=\"");

            final String cid =
                m.group(4) != null ? m.group(4) :
                m.group(5) != null ? m.group(5) :
                m.group(6) != null ? m.group(6) :
                null;

            final MutableBoolean binaryBodyFound = new MutableBoolean(false);

            getRootMessage().visitBodies(new BodyVisitor() {
                @Override
                public void visitBinaryBody(BinaryBody binaryBody) {
                    Entity entity = binaryBody.getEntity();
                    UnstructuredField contentIdField = entity.getContentIdField();
                    if(contentIdField != null) {
                        if(contentIdField.getText().equalsIgnoreCase(cid)) {
                            sb.append("\" name=\"embedded_img_").append(binaryBody.getOid()).append("/").append(binaryBody.getFileName()).append("\" ");
                            binaryBodyFound.setValue(true);
                            quit();
                        }
                    }
                }
            });

            if(binaryBodyFound.isFalse()) {
                sb.append("cid:").append(cid).append("\" ");
            }

            lastEnd = m.end();
        }
        sb.append(text.substring(lastEnd));
        return sb.toString();
    }

    //--------------------------------------------------------------------------
    @NotPersistent
    private static final String mailToRegex = "href\\s*=\\s*(" +
            "'\\s*mailto:\\s*([^'\\s<>]*)\\s*'" +
            "|" +
            "\"\\s*mailto:\\s*([^\"\\s<>]*)\\s*\"" +
            "|" +
            "mailto:\\s*([^\\s<>]+)" +
            ")";
    @NotPersistent
    private static final Pattern mailToPattern = Pattern.compile(mailToRegex, Pattern.CASE_INSENSITIVE);

    private String replaceMailToRef(String text) {
        Matcher m = mailToPattern.matcher(text);
        StringBuilder sb = new StringBuilder();
        int lastEnd = 0;
        while(m.find()) {
            sb.append(text.substring(lastEnd, m.start()))
              .append("name=\"_mailto\" href=\"javascript:new_message_to('");

            if(m.group(2) != null) {
                sb.append(m.group(2));
            }
            else if(m.group(3) != null) {
                sb.append(m.group(3));
            }
            else if(m.group(4) != null) {
                sb.append(m.group(4));
            }
            sb.append("');\"");
            lastEnd = m.end();
        }
        sb.append(text.substring(lastEnd));
        return sb.toString();
    }

    //--------------------------------------------------------------------------
    @NotPersistent
    private static final String urlRegex = "\\b(" +
            "(?:(https?://)|www\\d{0,3}[.]|[a-z0-9.\\-]+[.](?=(?:com|org|net|gov|mil|info)(?:[.]br)?)|[a-z0-9.\\-]+[.][a-z]{2,4}/)" +
            "(?:[^\\s()<>]+|\\(([^\\s()<>]+|(\\([^\\s()<>]+\\)))*\\))+" +
            "(?:\\(([^\\s()<>]+|(\\([^\\s()<>]+\\)))*\\)|[^\\s`!()\\[\\]{};:'\".,<>\\?«»“”‘’]))";
    @NotPersistent
    private static final Pattern urlPattern = Pattern.compile(urlRegex, Pattern.CASE_INSENSITIVE);

    private String replaceUrlRef(String text) {

        Source source = new Source(text);
        source.fullSequentialParse();

        Matcher m = urlPattern.matcher(text);
        StringBuilder sb = new StringBuilder();
        int lastEnd = 0;
        find:
        while(m.find()) {

            Element element = source.getEnclosingElement(m.start());
            while(element != null) {
                String elementName = element.getName();

                if(elementName == HTMLElementName.A
                || elementName == HTMLElementName.LINK
                || elementName == HTMLElementName.AREA
                || elementName == HTMLElementName.BASE
                || elementName == HTMLElementName.META
                || elementName == HTMLElementName.SCRIPT
                || elementName == HTMLElementName.IMG
                || elementName == HTMLElementName.VIDEO
                || elementName == HTMLElementName.AUDIO
                || elementName == HTMLElementName.SOURCE
                || elementName == HTMLElementName.BLOCKQUOTE
                || elementName == HTMLElementName.DEL
                || elementName == HTMLElementName.INS
                || elementName == HTMLElementName.Q
                || elementName == HTMLElementName.BUTTON
                || elementName == HTMLElementName.INPUT
                || elementName == HTMLElementName.OBJECT
                || elementName == HTMLElementName.EMBED
                || elementName == HTMLElementName.COMMAND)
                {
                    continue find;
                }

                Tag startTag = element.getStartTag();

                if((startTag.getBegin() < m.start()) && (m.start() < startTag.getEnd())) {
                    continue find;
                }

                element = element.getParentElement();
            }

            sb.append(text.substring(lastEnd, m.start()))
              .append("<a href=\"");

            if(m.group(2) == null) {
                sb.append("http://");
            }

            sb.append(m.group()).append("\">").append(m.group(1)).append("</a>");
            lastEnd = m.end();
        }
        sb.append(text.substring(lastEnd));
        return sb.toString();
    }
}