shoeboxencoder.java

编辑视频文件

/*
 * File:     ShoeboxEncoder.java
 * Project:  MPI Linguistic Application
 * Date:     02 May 2007
 *
 * Copyright (C) 2001-2007  Max Planck Institute for Psycholinguistics
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 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 General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

/*
 * Created on Oct 15, 2004
 */
package mpi.eudico.server.corpora.clomimpl.shoebox;

import mpi.eudico.server.corpora.clom.AnnotationDocEncoder;
import mpi.eudico.server.corpora.clom.EncoderInfo;
import mpi.eudico.server.corpora.clom.Tier;
import mpi.eudico.server.corpora.clom.Transcription;

import mpi.eudico.server.corpora.clomimpl.abstr.MediaDescriptor;
import mpi.eudico.server.corpora.clomimpl.abstr.TierImpl;
import mpi.eudico.server.corpora.clomimpl.abstr.TranscriptionImpl;
import mpi.eudico.server.corpora.clomimpl.shoebox.interlinear.Interlinearizer;
import mpi.eudico.server.corpora.clomimpl.shoebox.interlinear.TimeCodedTranscriptionImpl;
import mpi.eudico.server.corpora.clomimpl.shoebox.utr22.SimpleConverter;

import java.io.FileOutputStream;
import java.io.OutputStreamWriter;

import java.text.DecimalFormat;

import java.util.ArrayList;
import java.util.Hashtable;
import java.util.Iterator;
import java.util.List;
import java.util.Vector;

import javax.swing.JOptionPane;


/**
 * Encodes information from a Transcription to Shoebox/Toolbox format and stores it.
 *
 * @author hennie
 */
public class ShoeboxEncoder implements AnnotationDocEncoder {
    /** Holds value of property DOCUMENT ME! */
    public static final String defaultDBType = "ElanExport";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanELANLabel = "\\ELANExport";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanBlockStart = "\\block";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanBeginLabel = "\\ELANBegin";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanEndLabel = "\\ELANEnd";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanParticipantLabel = "\\ELANParticipant";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanMediaURLLabel = "\\ELANMediaURL";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanMediaExtractedLabel = "\\ELANMediaExtracted";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanMediaMIMELabel = "\\ELANMediaMIME";

    /** Holds value of property DOCUMENT ME! */
    public static final String elanMediaOriginLabel = "\\ELANMediaOrigin";
    private OutputStreamWriter isoLatinWriter;
    private OutputStreamWriter utf8Writer;
    private SimpleConverter simpleConverter;

    /**
     * Creates a new ShoeboxEncoder instance
     *
     * @param path DOCUMENT ME!
     */
    public ShoeboxEncoder(String path) {
        try {
            FileOutputStream out = new FileOutputStream(path);

            isoLatinWriter = new OutputStreamWriter(out, "ISO-8859-1");
            utf8Writer = new OutputStreamWriter(out, "UTF-8");
        } catch (Exception ex) {
            ex.printStackTrace();
        }
    }

    /*
     * @see mpi.eudico.server.corpora.clom.AnnotationDocEncoder#encodeAndSave(mpi.eudico.server.corpora.clom.Transcription, java.util.Vector, java.lang.String)
     */
    public void encodeAndSave(Transcription theTranscription,
        EncoderInfo encoderInfo, Vector tierOrder, String path) {
        try {
            writeHeader(theTranscription, encoderInfo);
            writeBlocks(theTranscription, tierOrder, encoderInfo);

            // media descriptors are written at the end of the Toolbox file. When written at begin/in header
            // Toolbox throws them away without any notification.
            writeMediaDescriptors(theTranscription);
            closeFile();
        } catch (Exception e) {
            String txt = "Sorry: unable to export this file to Shoebox." +
                e.getMessage() + ")";
            JOptionPane.showMessageDialog(null, txt, txt,
                JOptionPane.ERROR_MESSAGE);
            e.printStackTrace();
        }
    }

    private void writeHeader(Transcription theTranscription,
        EncoderInfo encoderInfo) {
        String dbType = ((ToolboxEncoderInfo) encoderInfo).getDatabaseType();

        if ((dbType == null) ||
                dbType.equals("") /*||
            ((ToolboxEncoderInfo) encoderInfo).getMarkerSource() == ToolboxEncoderInfo.TIERNAMES*/) {
            dbType = defaultDBType;
        }

        write("\\_sh v3.0  400  " + dbType + "\n");
        write("\\_DateStampHasFourDigitYear\n");

        if (((ToolboxEncoderInfo) encoderInfo).getMarkerSource() == ToolboxEncoderInfo.TIERNAMES) {
            write(("\n" + elanELANLabel + "\n"));
        }
    }

    private void writeMediaDescriptors(Transcription theTranscription) {
        // media descriptors
        Vector mediaDescriptors = null;

        try {
            mediaDescriptors = theTranscription.getMediaDescriptors();
        } catch (Exception rex) {
            rex.printStackTrace();
        }

        for (int i = 0; i < mediaDescriptors.size(); i++) {
            write("\n");

            MediaDescriptor md = (MediaDescriptor) mediaDescriptors.elementAt(i);

            if ((md.mediaURL != null) && !md.mediaURL.equals("")) {
                write((elanMediaURLLabel + " " + md.mediaURL + "\n"));
            }

            if ((md.mimeType != null) && !md.mimeType.equals("")) {
                write((elanMediaMIMELabel + " " + md.mimeType + "\n"));
            }

            if (md.timeOrigin != 0) {
                write((elanMediaOriginLabel + " " + md.timeOrigin + "\n"));
            }

            if ((md.extractedFrom != null) && !md.extractedFrom.equals("")) {
                write((elanMediaExtractedLabel + " " + md.extractedFrom + "\n"));
            }
        }
    }

    /**
     * Jul 2005: added a test on the toplevel tiers:<br>
     *  - if there is only one tier use it as the RecordMarker
     *  - if all top level tiernames start with "xxxx@" use "xxxx" as the RecordMarker
     *
     * in both cases don't add the ELAN "block" marker as RecordMarker
     *
     * @param theTranscription the transcription to export
     * @param tierOrder the order of the tiers
     * @param encoderInfo info for the encoder
     */
    private void writeBlocks(Transcription theTranscription, Vector tierOrder,
        EncoderInfo encoderInfo) {
        boolean lineForRootAnnot = false;
        boolean justOneRoot = false;
        int blockCounter = 1;

        Interlinearizer interlinearizer = new Interlinearizer(new TimeCodedTranscriptionImpl(
                    (TranscriptionImpl) theTranscription));

        setShoeboxArguments(theTranscription, interlinearizer, tierOrder,
            encoderInfo);

        String[] outputLines = interlinearizer.renderAsText();

        // find set of root tier names, for each root tier store participant
        String participantString = "";
        Hashtable rootTierNames = new Hashtable();

        try {
            Vector topTiers = ((TranscriptionImpl) theTranscription).getTopTiers();

            Iterator tierIter = topTiers.iterator();

            while (tierIter.hasNext()) {
                TierImpl t = (TierImpl) tierIter.next();
                rootTierNames.put(t.getName(), t.getParticipant());
            }

            if (topTiers.size() == 1) {
                justOneRoot = true;
            } else {
                // loop over toptiers; if all have a '@' in their name and the prefix
                // is always the same, use this prefix as the record marker
                justOneRoot = true;

                String name;
                String prefix = null;
                int atIndex = -1;
                Iterator it = rootTierNames.keySet().iterator();

                while (it.hasNext()) {
                    name = (String) it.next();
                    atIndex = name.indexOf('@');

                    if (atIndex < 1) {
                        justOneRoot = false;

                        break;
                    } else {
                        String curPref = name.substring(0, atIndex);

                        //System.out.println("Pref: " + curPref);
                        if (prefix == null) {