Laboratoire ICAR » Plateforme TXM

package org.txm.macro.txt

import org.kohsuke.args4j.*

import groovy.transform.Field

import java.nio.charset.Charset

import org.txm.rcp.swt.widget.parameters.*

import org.txm.utils.*

import javax.xml.stream.*

import java.net.URL

@Field @Option(name="inputDirectory", usage="CQP directory", widget="Folder", required=false, def="")

		File inputDirectory

@Field @Option(name="encoding", usage="File encoding", widget="String", required=false, def="UTF-8")

		String encoding

@Field @Option(name="separator", usage="Column separator (TAB by default)", widget="String", required=false, def="\\t")

		String separator

@Field @Option(name="columns", usage="Column names separated by comma ('word,pos,lemma' by default)", widget="String", required=true, def="word,pos,lemma")

		String columns

def rootTag = "text"

if (!ParametersDialog.open(this)) return;

encoding = encoding.trim()

if (separator == "\\t") {

	separator = "\t"

} else {

	separator = separator[0]

}

columnNames = columns.trim().split(",")

outputDirectory = new File(inputDirectory, "out")

outputDirectory.mkdir()

println "processing: "+inputDirectory

for (File inputfile : inputDirectory.listFiles()) {

	if (inputfile.isDirectory() || inputfile.isHidden() || !inputfile.getName().endsWith(".cqp")) {

		continue // ignore

	}

	println " file: "+inputfile

	String name = inputfile.getName()

	int idx = name.lastIndexOf(".")

	if (idx > 0) {

		name = name.substring(0, idx)

	}

	File outputfile = new File(outputDirectory, name+".xml")

	// output = new FileOutputStream(outputfile)

	output = new PrintWriter(new BufferedWriter(new FileWriter(outputfile, true)));

	outputStr = new StringWriter()

	factory = XMLOutputFactory.newInstance()

	writer = factory.createXMLStreamWriter(outputStr)

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

	writer.writeCharacters("\n") // manage XML reserved characters

	writer.writeStartElement(rootTag)

	writer.writeAttribute("id", name)

	writer.writeCharacters("\n")

	inputfile.withReader(encoding) { reader ->

		def line

		while ((line = reader.readLine()) != null) {

			if (line.length() > 0) {

				if (line[0] == "<") {

					writer.flush()

					outputStr.write(line)

					outputStr.flush()

				} else {

					writer.writeStartElement("w")

					def word = ""

					println "line = "+line

					println "columnNames = "+columnNames

					line.split(separator).eachWithIndex { columnValue, index ->

						if (columnNames[index] == "word") {

							word = columnValue

						} else {

							writer.writeAttribute(columnNames[index], columnValue)

						}

					}

					writer.writeCharacters(word)

					writer.writeEndElement() // w

				}

				writer.writeCharacters("\n")

			}

		}

	}

	writer.writeEndElement() // rootTag

	writer.close()

	output.println(outputStr.toString())

	output.close()

}