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: 2015-12-17 12:11:39 +0100 (jeu. 17 déc. 2015) $

// $LastChangedRevision: 3087 $

// $LastChangedBy: mdecorde $

//

package org.txm.scripts.importer

import java.io.FileInputStream

import java.io.InputStreamReader

import java.nio.charset.Charset

import org.txm.utils.*

import javax.xml.stream.*

import java.net.URL

/**

 * The Class CsvToXml: allow to convert CSV files to XML files

 *

 * @author mdecorde

 */

class CsvToXml {

        /** The titles. */

        File xmlfile

        List<String> titles = []

        private CsvReader getCSVReader(File csvfile, String fieldSeparator, String encoding) {

                CsvReader csvreader = new CsvReader(csvfile.getAbsolutePath(), fieldSeparator.charAt(0), Charset.forName(encoding))

                //get titles

                csvreader.readHeaders()

                titles = []

                for (String title : csvreader.getHeaders()) {

                        titles << title

                }

                if (titles.size() == 0) {

                        println "the CSV file as no column"

                        return null

                }

                if (titles.contains("text") == 0) {

                        println "the CSV file as no 'text' column"

                        return null

                }

                if (titles.contains("id") == 0) {

                        println "the CSV file as no 'id' column"

                        return null

                }

                return csvreader

        }

        /**

         * Convert a CSV file to ONE XML file using the first line to declare metadata

         *

         * @param csvfile the csvfile

         * @param outfile the outfile

         * @param fieldSeparator the field separator

         * @param textSeparator the text separator

         * @param encoding the encoding

         * @return true, if successful

         */

        public boolean toOneXMLFile(File csvfile, File xmlfile, String fieldSeparator, String textSeparator, String encoding) {

                CsvReader csvreader = getCSVReader(csvfile, fieldSeparator, encoding)

                if (csvreader == null) return false

                println "Metadata properties declared: "+titles

                XMLOutputFactory factory = XMLOutputFactory.newInstance()

                FileOutputStream output = new FileOutputStream(xmlfile)

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

                writer.writeStartDocument("UTF-8","1.0")

                writer.writeStartElement("text")

                while (csvreader.readRecord()) {

                        writer.writeStartElement("div")

                        for (int i = 0 ; i < titles.size() ; i++) {

                                if (titles[i] == "text") continue

                                writer.writeAttribute(AsciiUtils.buildId(titles.get(i)), csvreader.get(i))

                        }

                        writer.writeCharacters(csvreader.get("text")) // get text content

                        writer.writeEndElement() // div

                }

                writer.writeEndElement() // text

                writer.close()

                output.close()

                csvreader.close()

                return true;

        }

        /**

        * Convert a CSV file to SEVERAL XML file using the first line to declare metadata

        *

        * @param csvfile the csvfile

        * @param outfile the outfile

        * @param fieldSeparator the field separator

        * @param textSeparator the text separator

        * @param encoding the encoding

        * @return true, if successful

        */

        public boolean toMultipleXMLFiles(File csvfile, File outDir, String fieldSeparator, String textSeparator, String encoding) {

                CsvReader csvreader = getCSVReader(csvfile, fieldSeparator, encoding)

                if (csvreader == null) return false

                if (!outDir.exists()) outDir.mkdirs()

                if (!outDir.exists()) {

                        println "Out directory does not exist and could not create it"

                        return false;

                }

                while (csvreader.readRecord()) {

                        File xmlfile = new File(outDir, csvreader.get("id")+".xml")

                        XMLOutputFactory factory = XMLOutputFactory.newInstance()

                        FileOutputStream output = new FileOutputStream(xmlfile)

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

                        writer.writeStartDocument("UTF-8","1.0")

                        writer.writeStartElement("text")

                        for (int i = 0 ; i < titles.size() ; i++) {

                                if (titles[i] == "text") continue

                                writer.writeAttribute(AsciiUtils.buildId(titles.get(i)), csvreader.get(i))

                        }

                        writer.writeCharacters(csvreader.get("text")) // get text content

                        writer.writeEndElement() //text

                        writer.close()

                        output.close()

                }

                return true

        }

        /**

         * The main method.

         *

         * @param args the arguments

         */

        static public void main(String[] args)

        {

                String home = System.getProperty("user.home")

                File csvfile = new File(home, "xml/csv/test.csv")

                File outfile = new File(home, "xml/csv/test.xml")

                File outdir = new File(home, "xml/csv/out")

                String fieldSeparator = "\t"

                String textSeparator = "" //or "'"

                CsvToXml builder = new CsvToXml()

                builder.toMultipleXMLFiles(csvfile, outdir, fieldSeparator, textSeparator, "UTF-8")

                builder.toOneXMLFile(csvfile, outfile, fieldSeparator, textSeparator, "UTF-8")

        }

}