Laboratoire ICAR » Plateforme TXM

// 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.

								flagAna = true;

								anaType = parser.getAttributeValue(null, "type")

								if (anaType.length() > 0) anaType = anaType.substring(1);

								anahash.put(anaType, "");

								if (firstWord) {

									anaTypes << anaType;

							case "w":

								firstWord = false;

								output.write( vForm.replaceAll("&", "&amp;").replaceAll("<", "&lt;")

								for(String type : anaTypes) {

									output.write("\t"+anahash.get(type));

								}

			}

			output.close();

			if (parser != null) parser.close();

		}

		catch (Exception ex) {

			System.out.println("Exception while parsing " + inputData);

	{

		this.langGroups = groups;

	}

	public static HashMap<Integer, ArrayList<String>> corpusIDS;

	public void setCorpusIDS(HashMap<Integer, ArrayList<String>> corpusIDS)

	{

		this.corpusIDS = corpusIDS;

	}

	/**

	 * Run.

	 *

		sattrsListener = null; // reset SAttribute Listener for each new import

		String rootDir = binDir.getAbsolutePath();

		seg_id=1;

		if (!(CwbEncode.isExecutableAvailable() && CwbMakeAll.isExecutableAvailable())) {

			println ("Error: CWB executables not well set.")

			return false;

		}

		new File(binDir,"cqp").mkdirs()

		new File(binDir,"data").mkdirs()

		new File(binDir,"registry").mkdirs()

			}

		}

		println "Using corpus ID: $corpusIDS"

		String cqpName;

		for (int group : langGroups.keySet()) {

			String lang = langs.get(langGroups.get(group)[0]);

		for (int group : langGroups.keySet()) {

			//String lang = langs.get(langGroups.get(group)[0]);

			cqpName = basename+"_"+corpusIDS.get(group);

			} else {

			}

			corpus.setDescription("Built with the TMX import module");

			File cqpFile = new File(binDir,"cqp/"+cqpName+".cqp");

			def filenames = langGroups.get(group);

			filenames.sort()

			//println("Process group no $group of files "+filenames)

			}

		}

		println ""

		for (int group : langGroups.keySet()) {

			//String lang = langs.get(langGroups.get(group)[0]);

			createOutput(rootDir+"/cqp", basename+"_"+corpusIDS.get(group)+".cqp");

		def cqpFiles = [:]

		for (File cqpfile : new File(rootDir, "cqp").listFiles()) {

			String corpusname = cqpfile.getName();

			corpusname = corpusname.substring(0, corpusname.length()-4);

			def corpusAnaTypes = anaTypesPerCqp[corpusname];

			CwbEncode cwbEn = new CwbEncode();

			cwbEn.setDebug(debug);

			CwbMakeAll cwbMa = new CwbMakeAll();

					if (!corpusName.equals(targetName)) {

						try {

							PatchCwbRegistry.patchAlignment(new File(registryDirectory, corpusName.getName()), targetName.getName());

							File cqpFile1 = new File(cqpFiles[corpusName.getName()]);

							File cqpFile2 = new File(cqpFiles[targetName.getName()]);

							File alignOutFile = new File(outDir, "align.out");

							BuildAlignOut bao = new BuildAlignOut(cqpFile1, cqpFile2);

							if (!bao.process(alignOutFile, "seg", "id")) {

								println "Error while creating alignement file of $corpusName. Aborting."

								return false;

							}

							CwbAlign tt = new CwbAlign();

							// ./cwb-align-encode -D -r ~/TXM/corpora/tmxtest/registry/ -v out.align

							tt.setD();