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.

//

//

//

// $LastChangedDate:$

// $LastChangedRevision:$

// $LastChangedBy:$

//

package org.txm.scripts.filters.ReunitBrokenWords;

//Pre-processing extra-word tags (2)

import org.txm.importer.scripts.filters.*;

import org.txm.tokenizer.*;

// TODO: Auto-generated Javadoc

/**

 * The Class ReunitBrokenWords.

 */

class ReunitBrokenWords extends Filter

{

        /** The tag_all. */

        def tag_all = TokenizerClasses.tag_all;

        /** The div_tags. */

        def div_tags = TokenizerClasses.div_tags;

        /** The q_tags. */

        def q_tags = TokenizerClasses.q_tags;

        /** The extraword_tags. */

        def extraword_tags = TokenizerClasses.extraword_tags;

        /** The corr_tags_no_seg. */

        def corr_tags_no_seg = TokenizerClasses.corr_tags_no_seg;

        /** The corr_tags. */

        def corr_tags = TokenizerClasses.corr_tags;

        /** The word_tags. */

        def word_tags = TokenizerClasses.word_tags;

        /** The intraword_tags. */

        def intraword_tags = TokenizerClasses.intraword_tags;

        /** The tag. */

        def tag;

        /** The tag_end. */

        def tag_end;

        /** The tag_name. */

        def tag_name;

        /** The word_part_tags. */

        List<String> word_part_tags = [];

        /** The unknowned. */

        Set<String> unknowned = new HashSet<String>();

        /** The level. */

        private String level = "extraword";

        /* (non-Javadoc)

         * @see org.txm.importer.filters.Filter#SetUsedParam(java.lang.Object)

         */

        void SetUsedParam(Object args)

        {

                try

                {

                        div_tags = args.get("div_tags");

                        q_tags = args.get("q_tags");

                        extraword_tags = args.get("extraword_tags");

                        corr_tags_no_seg = args.get("corr_tags_no_seg");

                        corr_tags = args.get("corr_tags");

                        word_tags = args.get("word_tags");

                        intraword_tags = args.get("intraword_tags");

                }

                catch(Exception e)

                {

                        System.err.println(e);

                        System.err.println("PreProcess2 needs 1 Map with args  :\n div_tags, q_tags, extraword_tags, " +

                                        "corr_tags_no_seg, corr_tags, word_tags, intraword_tags")

                }

        }

        /* (non-Javadoc)

         * @see org.txm.importer.filters.Filter#before()

         */

        boolean before(){        }

        /* (non-Javadoc)

         * @see org.txm.importer.filters.Filter#after()

         */

        void after()

        {

                if(unknowned.size() > 0)

                        println "Unknown tags: "+unknowned;

        }

        /* (non-Javadoc)

         * @see org.txm.importer.filters.Filter#filter()

         */

        void filter()

        {

                def m;

                def segment;

                boolean firstTest = true;

                while(!(line ==~ /\A\s*\Z/))// tant que la ligne contient des charracters autres que des blancs

                {

                        if ((m = line =~ /^([^<]*)$/))          // si la ligne NE commence PAS par une balise

                        {

                                if(firstTest)

                                {

                                        //output.print(lineSeparator);

                                        firstTest=false;

                                }

                                output.print(line);//on ecrit la ligne

                                line = " "

                        }

                        else if ((m = line =~ /\A\s*(<\/?($extraword_tags)( [^>]*|\/)?>)(.*)\Z/)) // balise externe

                        {

                                tag = m[0][1]; // le nom du tag

                                line = m[0][4];

                                //println "balise ext : "+tag

                                level = "extraword"; //di qu'on est ds un tag extern

                                output.print("\n$tag \n");

                        }

                        else if ((m = line =~ /\A\s*(<($corr_tags)( [^>]*|\/)?>)(.*)\Z/)) //balise corr ouvrante

                        {

                                tag = m[0][1];

                                String tag_end = m[0][3]; // ( [^>]*|\/)?>)

                                line = m[0][4];

                                //println "balise corr : "+tag+" end : "+tag_end

                                if(tag_end != null && tag_end.find("word_part")) //on trouve un word_part

                                {

                                        //println "set level : intra :"+tag

                                        word_part_tags.add(m[0][2]);

                                        level = "intraword";

                                        output.print(tag);

                                }

                                else {

                                        level = "extraword";

                                        output.print("\n$tag\n");

                                }

                        }

                        else if ((m = line =~ /^\s*(<($word_tags)( [^>]*|\/)?>)(.*)$/))

                        {

                                tag = m[0][1];

                                line = m[0][4];

                                //println "balise word : "+tag

                                //output.print("\n$tag");

                                output.write(line)

                                level = "intraword";

                        }

                        else if ((m = line =~ /^\s*(<($intraword_tags)( [^>]*|\/)?>)(.*)$/))

                        {

                                tag = m[0][1];

                                line = m[0][4];

                                //println "balise intra : "+tag

                                output.print(tag);

                                level = "intraword";

                        }

                        else if ((m = line =~ /^\s*(<\/($corr_tags|$word_tags|$intraword_tags)>)\s*(.*)$/))

                        {

                                tag = m[0][1];

                                tag_name = m[0][2];

                                line = m[0][3];

                                //println "balise tous : "+tag+" level : "+level

                                if (level ==~ /extraword/)

                                {//println "write extra"

                                        output.print("\n");

                                }//else{println "dont write intra"}

                                output.print(tag);

                                if (tag_name ==~ /^($corr_tags|$word_tags)$/)

                                {

                                        //println("> corrwordtags "+tag_name+" level "+level+" Word_part_tag "+word_part_tags)

                                        if (level ==~ /extraword/)

                                        {

                                                output.print("\n");

                                        }

                                        //else{println "dont write intra corr"}

                                        if(word_part_tags.size() > 0)

                                        {

                                                //String word_part_tag = word_part_tags[-1];

                                                //if(tag_name == word_part_tag)

                                                //{

                                                        word_part_tags.remove(word_part_tags.size()-1)

                                                        if(word_part_tags.size() == 0 )

                                                        {

                                                                level = "extraword";

                                                                //output.print("\n");

                                                        }

                                                //}

                                        }

                                        else

                                                level = "extraword"

                                }

                        }

                        else if ((m = line =~ /^\s*($tag_all)(.*)$/))

                        {

                                //println "balise autres : "+tag

                                tag = m[0][1];

                                line = m[0][2];

                                output.print("\n$tag\n");

                                if(!unknowned.contains(tag))

                                {

                                        unknowned.add(tag)

                                        //println "Warning: unknown tag $tag in line : "+line;

                                }

                        }

                        else{

                                println "Error in "+ line;

                                line = " ";

                        }

                        firstTest=false;

                }

                // End

        }

}