Laboratoire ICAR » Plateforme TXM

eclipse.preferences.version=1

groovy.compiler.level=25

eclipse.preferences.version=1

org.eclipse.jdt.core.compiler.codegen.inlineJsrBytecode=enabled

org.eclipse.jdt.core.compiler.codegen.targetPlatform=1.7

org.eclipse.jdt.core.compiler.compliance=1.7

org.eclipse.jdt.core.compiler.problem.assertIdentifier=error

org.eclipse.jdt.core.compiler.problem.enumIdentifier=error

org.eclipse.jdt.core.compiler.source=1.7

<?xml version="1.0" encoding="UTF-8"?>

<classpath>

	<classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.7"/>

	<classpathentry kind="con" path="org.eclipse.pde.core.requiredPlugins">

		<accessrules>

			<accessrule kind="accessible" pattern="**"/>

		</accessrules>

	</classpathentry>

	<classpathentry kind="src" path="src"/>

	<classpathentry kind="src" path="groovy"/>

	<classpathentry exported="true" kind="con" path="GROOVY_SUPPORT"/>

	<classpathentry exported="true" kind="con" path="GROOVY_DSL_SUPPORT"/>

	<classpathentry kind="output" path="bin"/>

</classpath>

Manifest-Version: 1.0

Bundle-ManifestVersion: 2

Bundle-Name: CONNLU

Bundle-SymbolicName: org.txm.connlu.core;singleton:=true

Bundle-Version: 1.0.0.qualifier

Automatic-Module-Name: org.txm.connlu.core

Bundle-RequiredExecutionEnvironment: JavaSE-1.7

Require-Bundle: org.txm.core;bundle-version="0.8.0";visibility:=reexport,

 org.txm.groovy.core;bundle-version="1.0.0";visibility:=reexport

<?xml version="1.0" encoding="UTF-8"?>

<projectDescription>

	<name>org.txm.connlu.core</name>

	<comment></comment>

	<projects>

	</projects>

	<buildSpec>

		<buildCommand>

			<name>org.eclipse.jdt.core.javabuilder</name>

			<arguments>

			</arguments>

		</buildCommand>

		<buildCommand>

			<name>org.eclipse.pde.ManifestBuilder</name>

			<arguments>

			</arguments>

		</buildCommand>

		<buildCommand>

			<name>org.eclipse.pde.SchemaBuilder</name>

			<arguments>

			</arguments>

		</buildCommand>

	</buildSpec>

	<natures>

		<nature>org.eclipse.jdt.groovy.core.groovyNature</nature>

		<nature>org.eclipse.pde.PluginNature</nature>

		<nature>org.eclipse.jdt.core.javanature</nature>

	</natures>

</projectDescription>

package org.txm.scripts.importer.connlu

import org.txm.Toolbox

import org.txm.importer.xtz.ImportModule;

import org.txm.metadatas.Metadatas

import org.txm.utils.io.FileCopy

import org.txm.utils.io.IOUtils

import org.txm.importer.xtz.*

import org.txm.scripts.importer.xtz.*

import org.txm.importer.ApplyXsl2;

import javax.xml.stream.*

import org.txm.utils.AsciiUtils

/**

 * Only build the Metadatas object since all XML-TXM files already exists.

 * Metadatas is used to build text order.

 * 

 * 

 * @author mdecorde

 *

 */

class CONNLUImporter extends XTZImporter {

	public CONNLUImporter(ImportModule module) {

		super(module);

	}

	@Override

	public void process() {

		File srcDirectory = new File(outputDirectory.getParentFile(), "src")

		srcDirectory.mkdirs();

		convert(inputDirectory, srcDirectory)

		inputDirectory = srcDirectory // switch source directory

		super.process();

	}

	public static void convert(File inputDirectory, File srcDirectory) {

		File master = null;

		def files = inputDirectory.listFiles()

		for (def f : files) {

			if (f.getName().endsWith(".connlu")) {

				master = f;

				break;

			}

		}

		if (master == null) {

			println "Aborting. No CONNLU file found in $inputDirectory."

			isSuccessFul = false;

			return

		}

		String text_id = null;

		String sent_id = null;

		def content = [:];

		master.eachLine("UTF-8") { line ->

			if (line.startsWith("# text = ")) {

				text_id = line.substring("# text = ".length())

			} else if (line.startsWith("# sent_id = ")) {

				sent_id = line.substring("# sent_id = ".length())

			} else {

				if (text_id != null && sent_id != null) {

					if (!content.containsKey(text_id)) {

						content[text_id] = [:]

					}

					def text = content[text_id]

					if (!text.containsKey(sent_id)) {

						text[sent_id] = []

					}

					text[sent_id] << line.split("\t")

				}

			}

		}

		for (def text_id2 : content.keySet()) {

			File xmlFile = new File(srcDirectory, AsciiUtils.buildAttributeId(text_id)+".xml")

			BufferedOutputStream output = new BufferedOutputStream(new FileOutputStream(xmlFile))

			XMLOutputFactory factory = XMLOutputFactory.newInstance();

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

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

			writer.writeStartElement ("TEI");

			writer.writeDefaultNamespace("http://www.tei-c.org/ns/1.0");

			writer.writeNamespace("txm", "http://textometrie.org/1.0");

			writer.writeStartElement ("teiHeader");

			writer.writeEndElement()

			writer.writeStartElement ("text");

			def text = content[text_id2]

			for (def sent_id2 : text.keySet()) {

				writer.writeStartElement ("s");

				for (def line : text[sent_id]) {

					writer.writeStartElement ("w");

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

						writer.writeAttribute("p"+(i+1), line[i])

					}

					writer.writeCharacters(line[1])

					writer.writeEndElement() // w

					writer.writeCharacters(" ")

				}

				writer.writeEndElement() // s

				writer.writeCharacters("\n")

			}

			writer.writeEndElement() // text

			writer.writeEndElement() // TEI

		}

	}

}

// 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: 2015-06-03 15:04:53 +0200 (mer., 03 juin 2015) $

// $LastChangedRevision: 2984 $

// $LastChangedBy: mdecorde $

//

package org.txm.scripts.importer.connlu;

import javax.xml.stream.*

import org.txm.*

import org.txm.metadatas.*

import org.txm.objects.*

import org.txm.importer.scripts.xmltxm.*

import org.txm.utils.*

import org.txm.utils.i18n.*

import org.txm.importer.xtz.*

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

def MONITOR;

Project project;

try {project=projectBinding;MONITOR=monitor} catch (Exception)

{	}

if (project == null) { println "no project set. Aborting"; return; }

CONNLUImport i = new CONNLUImport(project);

/*

 * To customize the XTZ import, replace the importer, compiler, annotater or pager objects before calling process()

		i.importer = new XTZImporter(i)

		i.compiler = new XTZCompiler(i)

		i.annotater = new TTAnnotater(i);

		i.pager = new XTZPager(i)

 */

i.process();

readyToLoad = i.successFul && project.save()

package org.txm.scripts.importer.connlu;

import org.apache.log4j.BasicConfigurator;

import org.txm.importer.xtz.*

import org.txm.objects.Project

import org.txm.scripts.importer.xtz.*

import ims.tiger.index.writer.*

import ims.tiger.system.*

class CONNLUImport extends XTZImport {

	public CONNLUImport(Project params) {

		super(params);

	}

	@Override

	public void init(Project p) {

		super.init(p);

		importer = new CONNLUImporter(this); // select TIGER XML files then do XTZImporter step

		compiler = new XTZCompiler(this)

		annotater = null; // no annotater step to do

		pager = new XTZPager(this)

	}

	/**

	 * Do a XTZ Import then build the TIGERSearch indexes in the binary corpus "tiger" directory

	 */

	@Override

	public void start() throws InterruptedException {

		super.start(); // call the usual XTZ import

		if (successFul) {

			println "BUILD CONNL-UD indexes ??"

		}

	}

}

<html>

<body>

<p>TIGERSearch import module. This is a prototype that can only manage SRCMF TIGERSearch sources</p>

</body>

</html>

package org.txm.connlu.core;

import java.io.File;

import org.txm.PostTXMHOMEInstallationStep;

import org.txm.Toolbox;

import org.txm.objects.Workspace;

import org.txm.utils.BundleUtils;

public class InstallGroovyCONNLUFiles extends PostTXMHOMEInstallationStep {

	final static String createfolders[] = {"scripts/groovy/lib", "scripts/groovy/user", "scripts/groovy/system"}; //$NON-NLS-1$ //$NON-NLS-2$ //$NON-NLS-3$ //$NON-NLS-4$

	@Override

	public boolean do_install(Workspace workspace) {

		File txmhomedir = new File(Toolbox.getTxmHomePath());

		for (String folder : createfolders) {

			new File(txmhomedir, folder).mkdirs();

		}

		File scriptsDirectory = new File(txmhomedir, "scripts/groovy");

		File userDirectory = new File(scriptsDirectory, "user");

		File systemDirectory = new File(scriptsDirectory, "system");

		String bundle_id = "org.txm.connlu.core";

		File scriptsPackageDirectory = new File(userDirectory, "org/txm/scripts/importer");

		File scriptsPackageDirectory2 = new File(systemDirectory, "org/txm/scripts/importer");

		scriptsPackageDirectory.mkdirs();

		scriptsPackageDirectory2.mkdirs();

		BundleUtils.copyFiles(bundle_id, "groovy", "org/txm/scripts/importer", "", scriptsPackageDirectory);

		BundleUtils.copyFiles(bundle_id, "groovy", "org/txm/scripts/importer", "", scriptsPackageDirectory2);

		return scriptsDirectory.exists();

	}

	@Override

	public String getName() {

		return "CONNLU (org.txm.connlu.core)";

	}

	@Override

	public boolean needsReinstall(Workspace workspace) {

		File txmhomedir = new File(Toolbox.getTxmHomePath());

		for (String folder : createfolders) {

			if (!new File(txmhomedir, folder).exists()) {

				return true;

			}

		}

		return false;

	}

}

source.. = src/

output.. = bin/

bin.includes = META-INF/,\

               .,\

               groovy/,\

               plugin.xml

qualifier=svn

<?xml version="1.0" encoding="UTF-8"?>

<?eclipse version="3.4"?>

<plugin>

   <extension

         point="org.eclipse.ui.menus">

      <menuContribution

            locationURI="menu:menu.file.import?before=menu.file.import.separator.software">

         <command

               commandId="org.txm.rcp.handlers.scripts.ExecuteImportScript"

               label="Connl-UD"

               style="push">

            <parameter

                  name="org.txm.rcp.commands.commandParameter3"

                  value="connlu/connluLoader.groovy">

            </parameter>

         </command>

      </menuContribution>

   </extension>

   <extension

         point="org.txm.PostTXMHOMEInstallationStep">

      <PostTXMHOMEInstallationStep

            class="org.txm.connlu.core.InstallGroovyCONNLUFiles"

            description="copy scripts for import module"

            name="CONNLUFiles">

      </PostTXMHOMEInstallationStep>

   </extension>

</plugin>