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: 2017-01-24 18:11:42 +0100 (mar. 24 janv. 2017) $

// $LastChangedRevision: 3400 $

// $LastChangedBy: mdecorde $

//

package org.txm.importer.scripts.xmltxm

import java.util.ArrayList;

import java.util.HashMap;

import javax.xml.stream.*;

import java.net.URL;

import org.txm.importer.filters.*;

// TODO: Auto-generated Javadoc

/**

 * The Class AnnotationExtraction.

 *

 * @author mdecorde

 *

 * Extract ana tags from a xml-tei-txm file

 * saved into a stand-off file

 */

public class AnnotationExtraction

{

        /** The input data. */

        private def inputData;

        /** The factory. */

        private def factory;

        /** The parser. */

        private XMLStreamReader parser;

        /** The dir. */

        private def dir;

        /** The writer. */

        private Writer writer;

        /** The url. */

        private def url;

        /** The posfile. */

        private def posfile;

        /** The anafile. */

        private def anafile;

        /** The encoding. */

        private def encoding;

        /** The extract types. */

        private List<String> extractTypes;

        /** The resp stmt. */

        private String respStmt;

        /** The solotags. */

        ArrayList<String> solotags;

        /**

         * initialize.

         *

         * @param anafile the path to the xml-tei-txm file

         * @param posfile the output stand-off file

         * @param encoding the encoding of the xml-tei-txm file

         * @param milestones milestones tags (temporary)

         */

        public AnnotationExtraction(File anafile, File posfile,String encoding,milestones) {

                this.url = anafile.toURI().toURL();

                this.posfile = posfile;

                this.encoding = encoding;

                this.solotags = milestones;

                inputData = url.openStream();

                factory = XMLInputFactory.newInstance();

        }

        /**

         * process !!!.

         */

        public void process() {

                assert(extractTypes != null);

                if (this.createOutput(posfile,encoding)) {

                        this.writeHead();

                        this.writeBody( posfile, encoding);

                        this.writeTail();

                }

                writer.close();

        }

        /**

         * Creates the output.

         *

         * @param outfile the outfile

         * @param encoding the encoding

         * @return true, if successful

         */

        private boolean createOutput(File outfile, String encoding){

                try {

                        writer = new OutputStreamWriter(new FileOutputStream(outfile) , "UTF-8");

                        return true;

                } catch (Exception e) {

                        System.out.println(e.getLocalizedMessage());

                        return false;

                }

        }

        /**

         * set the respStmt tag of the stand-off file by reading the respstmt of the xml-tei-txm file.

         */

        private void setResp()

        {

                this.respStmt = "";

                boolean begincapture = false;

                String lastopenlocalname= "";

                inputData = url.openStream();

                parser = factory.createXMLStreamReader(inputData);

                for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next())

                {

                        String prefix = parser.getPrefix();

                        if (prefix == null || prefix == "")

                                prefix = "";

                        else

                                prefix +=":";

                        if (event == XMLStreamConstants.START_ELEMENT) {

                                if (parser.getLocalName().matches("resp")) {

                                        String id = parser.getAttributeValue("xml","id")

                                        if (id == null)

                                                id = parser.getAttributeValue(null,"id")

                                        if (extractTypes.contains(id)) {

                                                this.respStmt = "<respStmt>\n";

                                                begincapture= true;

                                        }

                                }

                                if (begincapture) {

                                        lastopenlocalname = parser.getLocalName();

                                        respStmt += ("\n<"+prefix+parser.getLocalName());

                                        for (int i= 0 ; i < parser.getAttributeCount() ;i++ )

                                                respStmt += (" "+parser.getAttributeLocalName(i)+"=\""+parser.getAttributeValue(i)+"\"" );

                                        if (solotags.contains(lastopenlocalname))

                                                respStmt += ("/>");

                                        else

                                                respStmt += (">");

                                }

                        }

                        else if (event == XMLStreamConstants.END_ELEMENT) {

                                if (parser.getLocalName().matches("respStmt")) {

                                        begincapture= false;

                                        this.respStmt += "\n</respStmt>"

                                        System.out.println(this.respStmt);

                                        parser.close();

                                        inputData.close()

                                        return;

                                } else {

                                        if (!solotags.contains(parser.getLocalName()))

                                                if (lastopenlocalname.equals(parser.getLocalName()))

                                                        respStmt += ("</"+prefix+parser.getLocalName()+">");

                                                else

                                                        respStmt += ("\n</"+prefix+parser.getLocalName()+">");

                                }

                        } else if (event == XMLStreamConstants.CHARACTERS) {

                                String txt = parser.getText().trim();

                                respStmt += txt;

                        }

                }

                parser.close();

                inputData.close()

                return;

        }

        /**

         * write the header of the stand-off ile.

         */

        private void writeHead() {

                this.setResp();

                writer.write("<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");

                writer.write("<!DOCTYPE TEI SYSTEM \"tei_bfm_v1.dtd\">\n");

                writer.write("<TEI xmlns:txm=\"http://textometrie.ens-lyon.fr/1.0\">\n");

                writer.write("<teiHeader xml:lang=\"eng\">\n");

                writer.write("<fileDesc>\n");

                writer.write("<titleStmt>\n");

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

                        writer.write("<title>"+extractTypes[i]+"</title>\n");

                writer.write(this.respStmt+"\n");

                writer.write("</titleStmt>\n");

                writer.write("<publicationStmt>\n");

                writer.write("<distributor>BFM project - http://bfm.ens-lsh.fr</distributor>\n");

                writer.write("<availability>\n");

                writer.write("<p>(c) 2010 Projet BFM - CNRS/ENS-LSH.\n");

                writer.write("<hi>Conditions d'utilisation</hi> : \n");

                writer.write("Sous licence <ref target=\"http://creativecommons.org/licenses/by-sa/2.0/fr/\">Creative Commons</ref>.\n");

                writer.write("</p>\n");

                writer.write("</availability>\n");

                writer.write("</publicationStmt>\n");

                writer.write("<sourceDesc>\n");

                writer.write("<p>born digital : TXM project - http://textometrie.org</p>\n");

                writer.write("</sourceDesc>\n");

                writer.write("</fileDesc>\n");

                writer.write("</teiHeader>\n");

                writer.write("<text xml:lang=\"fr\" type=\"standoff\">\n");

                writer.write("<body>\n");

                writer.write("<div>\n");

        }

        /**

         * write the group of ana tags with the attribute type = type.

         *

         * @param type the type

         */

        private void writeGroup(String type)

        {

                System.out.println("write group : "+type);

                inputData = url.openStream();

                factory = XMLInputFactory.newInstance();

                parser = factory.createXMLStreamReader(inputData);

                boolean flagAna = false;

                boolean flagForm = false;

                boolean flagW = false;

                String wordid="";

                String vAna="";

                String currentType;

                writer.write("<linkGrp type=\""+type+"\">\n")

                for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next())

                {

                        switch (event) {

                                case XMLStreamConstants.START_ELEMENT:

                                        switch (parser.getLocalName()) {

                                                case "w":

                                                wordid = parser.getAttributeValue(null,"id");

                                                flagW = true;

                                                break;

                                                case "ana":

                                                if (flagW) {

                                                        currentType = (parser.getAttributeValue(null,"type"));

                                                        if (currentType.matches(type)) {

                                                                flagAna = true;

                                                                vAna ="";

                                                        }

                                                }

                                                break;

                                        }

                                        break;

                                case XMLStreamConstants.END_ELEMENT:

                                        switch (parser.getLocalName()) {

                                                case "w":

                                                flagW = false;

                                                break;

                                                case "ana":

                                                if (flagW) {

                                                        writer.write("<link targets=\"#"+wordid+" #"+vAna+"\"/>\n")

                                                        flagAna = false;

                                                }

                                                break;

                                        }

                                        break;

                                case XMLStreamConstants.CHARACTERS:

                                        if (flagAna)

                                                vAna += parser.getText().trim();//catch interp text

                                        break;

                        }

                }

                writer.write("</linkGrp>\n");

                parser.close();

                inputData.close();

        }

        /**

         * write a group per ana type to extract from the anafile.

         *

         * @param TTrez the t trez

         * @param encoding the encoding

         * @return the java.lang. object

         */

        private writeBody(File TTrez, String encoding)

        {

                for (String type : extractTypes) {

                        writeGroup(type);

                }

        }

        /**

         * write the tail of the stand-off file = close body, text and TEI tags.

         */

        private void writeTail()

        {

                writer.write("</div>\n");

                writer.write("</body>\n");

                writer.write("</text>\n");

                writer.write("</TEI>\n");

        }

        /**

         * define the types to extract.

         *

         * @param types the new types to extract

         */

        public void setTypesToExtract(List<String> types)

        {

                extractTypes = types;

        }

        /**

         * The main method.

         *

         * @param args the arguments

         */

        public static void main(String[] args) {

                String rootDir = "~/xml/rgaqcj/"

                new File(rootDir+"/extract/").mkdir()

                File anafile = new File(rootDir+"/injection/","roland-ana.xml")

                File posfile =  new File(rootDir+"/extract/","roland-pos.xml")

                String encoding = "UTF-8"

                //extrait les txm:ana dont la ref est dans cette liste

                List<String> types = ["#t1","#POS"]

                ArrayList<String> milestones = new ArrayList<String>()

                //the tags you want to keep as milestones

                milestones.add("tagUsage")

                milestones.add("pb")

                milestones.add("lb")

                milestones.add("catRef")

                AnnotationExtraction builder = new AnnotationExtraction(anafile,posfile,encoding,milestones)

                builder.setTypesToExtract(types)

                builder.process()

                return

        }

}