Laboratoire ICAR » Plateforme TXM

// Copyright © 2010-2013 ENS de Lyon.

// Copyright © 2007-2010 ENS de Lyon, CNRS, INRP, University of

// Lyon 2, University of Franche-Comté, University of Nice

// Sophia Antipolis, University of Paris 3.

//

// The TXM platform 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.

//

// The TXM platform 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 the TXM platform. If not, see

// http://www.gnu.org/licenses.

//

//

//

// $LastChangedDate: 2017-06-26 16:53:47 +0200 (lun. 26 juin 2017) $

// $LastChangedRevision: 3451 $

// $LastChangedBy: mdecorde $

//

package org.txm.scripts.importer

import org.txm.utils.CharsetDetector

import org.txm.utils.i18n.DetectBOM;

import org.w3c.dom.Document;

import org.w3c.dom.Element;

import org.w3c.dom.NodeList;

import org.xml.sax.SAXException;

import java.io.File;

import java.io.FileInputStream;

import java.io.InputStreamReader;

import javax.xml.stream.*;

import java.net.URL;

// TODO: Auto-generated Javadoc

/**

 * Create a xml file per text declared in a Hyperbase file (old format).

 *

 * @author mdecorde

 */

class Hyperbase2Xml {

        /**

         * Run.

         *

         * @param infile the hyperbase file (old format)

         * @param outdir where the xml files are created

         * @param encoding the encoding of the hyperbase file

         * @return true, if successful

         */

        public boolean run(File infile, File outdir, String encoding)

        {

                if (encoding == "??") {

                        encoding = new CharsetDetector(infile).getEncoding();

                        println "Guessing encoding of $infile : $encoding"

                }

                outdir.mkdir();

                String textname = null

                String partname = null;

                int textcount = 1;

                def input = new FileInputStream(infile)

                Reader reader = new InputStreamReader(input , encoding);

                DetectBOM bomdetector = new DetectBOM(infile);

                for (int ibom = 0 ; ibom < bomdetector.getBOMSize() ; ibom++) input.read()

                XMLOutputFactory factory = XMLOutputFactory.newInstance();

                BufferedOutputStream output = null;

                XMLStreamWriter writer = null;

                boolean firsttext=true;

                String line = reader.readLine();

                while(line != null) {

                        line = CleanFile.clean(line);

                        if ((line.startsWith("&&&"))) // text declaration &&& longtitle, title, shorttitle &&&

                        {

                                if (partname != null) // close previous part

                                {

                                        //println("close last part")

                                        writer.writeEndElement();

                                        partname= null;

                                }

                                if (textname != null) // close previous text

                                {

                                        //println("close text "+textname)

                                        writer.writeEndElement();

                                        writer.close();

                                        output.close();

                                }

                                textname = line.substring(3, line.length()-3);

                                if (textname.contains(","))

                                {

                                        String[] textnames = textname.split(",");

                                        if(textnames.length > 1)

                                                textname = textnames[1].trim();

                                }

                                // create a new text

                                output = new BufferedOutputStream(new FileOutputStream(new File(outdir,textname+".xml")))

                                writer = factory.createXMLStreamWriter(output, "UTF-8");

                                writer.writeStartElement("text");

                                writer.writeAttribute("id",textname);

                                println "text : "+new File(outdir,textname+".xml");

                        }

                        else if((line.startsWith("%%%%"))) // an other type of text declaration

                        {

                                if(partname != null)

                                {

                                        //println("close last part")

                                        writer.writeEndElement();

                                        partname= null;

                                }

                                if(textname != null)

                                {

                                        //println("close text "+textname)

                                        writer.writeEndElement();

                                        writer.close();

                                        output.close();

                                }

                                textname = infile.getName().substring(0, infile.getName().length()-4)+(textcount++);

                                output =  new BufferedOutputStream(new FileOutputStream(new File(outdir,textname+".xml")))

                                writer = factory.createXMLStreamWriter(output, "UTF-8");

                                writer.writeStartElement("text");

                                writer.writeAttribute("id",textname);

                                println "text : "+new File(outdir,textname+".xml");

                        }

                        else if((line.startsWith("\$"))) // paragraph declaration

                        {

                                if(partname != null)

                                {

                                        //println("close part "+partname)

                                        writer.writeEndElement();

                                }

                                partname = line.substring(1);

                                writer.writeStartElement("p");

                                writer.writeAttribute("id",partname);

                        }

                        else // brut text > write chars

                        {

                                if(writer != null)

                                        writer.writeCharacters(line+"\n");

                        }

                        line = reader.readLine();

                }

                if(partname != null)

                {

                        //println("close last part")

                        writer.writeEndElement();

                }

                if(writer != null)

                        writer.writeEndElement();

                reader.close();

                if(output != null)

                        output.close();

                return true;

        }

        /**

         * Main.

         *

         * @param args the args

         */

        static main(args)

        {

                File infile = new File(System.getProperty("user.home")+"/xml/presdiscfra/src/PresDiscFranc");

                File outfile = new File(System.getProperty("user.home")+"/xml/presdiscfra/split/");

                new Hyperbase2Xml().run(infile, outfile,"ISO-8859-1");

                println "done"

        }

}