Laboratoire ICAR » Plateforme TXM

command.name.0  = in CoNLL-U format (.conllu)...

command.name.0  = au format CoNLL-U (.conllu)...

            name="%command.name.0">

      <command

            categoryId="TreeSearch4TXM.commands.category"

            defaultHandler="org.txm.conllu.rcp.commands.CoNLLUCorpusPreferences"

            id="org.txm.conllu.rcp.commands.CoNLLUCorpusPreferences"

            name="CoNLLU Corpus preferences">

      </command>

      <menuContribution

            locationURI="menu:menu.file.export.properties">

         <command

               commandId="org.txm.conllu.rcp.commands.ExportCorpusAsFullCoNLLU"

               icon="icons/functions/UD.png"

               style="push">

            <visibleWhen

                  checkEnabled="false">

               <reference

                     definitionId="OneMainCorpusSelected">

               </reference>

            </visibleWhen>

         </command>

      </menuContribution>

      <menuContribution

            locationURI="menu:menu.edit?after=org.txm.rcp.separator1">

         <command

               commandId="org.txm.conllu.rcp.commands.CoNLLUCorpusPreferences"

               style="push">

            <visibleWhen

                  checkEnabled="false">

               <reference

                     definitionId="OneMainCorpusSelected">

               </reference>

            </visibleWhen>

         </command>

      </menuContribution>

import jline.internal.Log;

		Log.info("Writing CoNLLu files in "+outdir.getAbsolutePath());

		int nTokens = 0;

			nTokens += words.length;

		Log.info("Done: "+texts.length+" texts and "+nTokens+" words.");

// Copyright © 2010-2020 ENS de Lyon., University of Franche-Comté

// 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.conllu.rcp.commands;

import java.util.HashMap;

import org.eclipse.core.commands.AbstractHandler;

import org.eclipse.core.commands.ExecutionEvent;

import org.eclipse.core.commands.ExecutionException;

import org.eclipse.jface.dialogs.InputDialog;

import org.eclipse.jface.dialogs.MessageDialog;

import org.eclipse.ui.handlers.HandlerUtil;

import org.kohsuke.args4j.Option;

import org.osgi.service.prefs.Preferences;

import org.txm.conllu.core.preferences.UDPreferences;

import org.txm.rcp.swt.widget.parameters.ParametersDialog;

import org.txm.rcp.views.corpora.CorporaView;

import org.txm.searchengine.cqp.corpus.MainCorpus;

import org.txm.utils.logger.Log;

/**

 * Export the conllu properties and CQP words into a conllu corpus of several files (one per text)

 * 

 * @author mdecorde.

 */

public class CoNLLUCorpusPreferences extends AbstractHandler {

	public static final String ID = CoNLLUCorpusPreferences.class.getName();

	/*

	 * (non-Javadoc)

	 * @see org.eclipse.core.commands.AbstractHandler#execute(org.eclipse.core.commands.ExecutionEvent)

	 */

	@Override

	public Object execute(final ExecutionEvent event) throws ExecutionException {

		Object s = CorporaView.getFirstSelectedObject();

		if (!(s instanceof MainCorpus)) {

			Log.warning("Selection is not a corpus. Aborting.");

			return null;

		}

		MainCorpus c = (MainCorpus)s;

		Preferences customNode = c.getProject().getPreferencesScope().getNode(UDPreferences.getInstance().getPreferencesNodeQualifier());

		String udPrefix = customNode.get(UDPreferences.UDPREFIX, UDPreferences.getInstance().getString(UDPreferences.UDPREFIX));

		InputDialog dialog = new InputDialog(HandlerUtil.getActiveShell(event), "CoNLL-U Corpus preferences", "Prefix of the CQP properties encoding the UD fields", udPrefix, null);

		if (dialog.open() == InputDialog.OK) {

			customNode.put(UDPreferences.UDPREFIX, dialog.getValue());

			Log.info("Set "+c.getName()+" prefix to "+dialog.getValue());

			return s;

		}

		return null;

	}

}

		return exportCorpus(mainCorpus, outputDirectory, sentenceStructure, lemmaProperty, posProperty, encoding);

	}

	public static boolean exportCorpus(MainCorpus mainCorpus, File outputDirectory, String sentenceStructure, String lemmaProperty, String posProperty, String encoding) {

			return new ToCoNLL2009().process(outputDirectory, mainCorpus, mainCorpus.getStructuralUnit(sentenceStructure), mainCorpus.getProperty("word"), mainCorpus.getProperty(lemmaProperty), mainCorpus.getProperty(posProperty), encoding);

		return false;