Laboratoire ICAR » Plateforme TXM

package org.txm.macro.frolex

import java.io.File

import java.io.IOException

import java.nio.charset.Charset

import java.util.LinkedHashMap

import org.apache.commons.io.filefilter.IOFileFilter;

import org.apache.commons.lang.StringUtils;

import org.eclipse.jface.bindings.keys.KeyStroke

import org.txm.utils.CsvReader

import org.txm.utils.io.IOUtils;

import org.kohsuke.args4j.*

import groovy.transform.Field

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

// PARAMETERS

@Field @Option(name="workingDirectory", usage="workingDirectory containing all files needed", widget="Folder", required=true, def="dir")

File workingDirectory = null

if (!ParametersDialog.open(this)) {

        System.out.println("Aborting CLBFMLEX creation.")

        return null

}

File cldmflexsitetsv = new File(workingDirectory, "cldmf-site.tsv")

if (!cldmflexsitetsv.exists()) {

        println "Aborting: file is missing: $cldmflexsitetsv"

        return

}

File cldmflexlgermtsv = new File(workingDirectory, "cldmf-lgerm.tsv")

if (!cldmflexlgermtsv.exists()) {

        println "Aborting: file is missing: $cldmflexlgermtsv"

        return

}

File tsvFile = new File(workingDirectory, "cldmf.tsv")

// START

LinkedHashMap<String, String[]> finalvalues = new LinkedHashMap<String, String[]>()

CsvReader reader_site = new CsvReader(cldmflexsitetsv.getAbsolutePath(), "\t".charAt(0), Charset.forName("UTF-8"))

CsvReader reader_lgerm = new CsvReader(cldmflexlgermtsv.getAbsolutePath(), "\t".charAt(0), Charset.forName("UTF-8"))

reader_site.readHeaders()

reader_lgerm.readHeaders()

reader_site.readRecord()

reader_lgerm.readRecord()

String key_site = reader_site.get("lemma")//+"_"+reader_site.get("category")

String key_lgerm = reader_lgerm.get("lemma")//+"_"+reader_lgerm.get("category")

def fixSiteCategory(String s) { // special fix for dmf_site category troncated

                if (s == "s" || s == "su" || s == "sub" || s == "subs" || s == "subst" || s == "subst. " ) {

                        return "subst."

                } else if (s == "v" || s == "ve" || s == "ver" || s == "verb" ) {

                        return "verbe"

                } else if (s == "a" || s == "adj" ) {

                        return "adj."

                }

                return s

}

int nMerge = 0;

int nSite = 0;

int nLgerm = 0;

// merge lgerm and site lemmas

while (key_site != null && key_lgerm != null) {

        def line = ["","","","","","","","","","","",""]

        if (key_site.equals(key_lgerm)) {

                //println "$key_site VS $key_lgerm"

                nMerge++

                line[0] = reader_site.get("lemma")

                line[1] = reader_lgerm.get("category") // site category is not broken

                if (reader_site.get("AND") != reader_lgerm.get("AND") && reader_lgerm.get("AND").length() > 0  && reader_site.get("AND").length() > 0) {

                        println "Different AND: $key_site\n site ="+reader_site.get("AND")+"\n lgerm="+reader_lgerm.get("AND");

                }

                if (reader_site.get("AND").length() > 0)

                        line[2] = reader_site.get("AND")

                else

                        line[2] = reader_lgerm.get("AND")

                line[3] = reader_lgerm.get("DEAF")

                line[4] = reader_site.get("DECT")

                line[5] = reader_site.get("FEW")

                line[6] = reader_site.get("GDF")

                line[7] = reader_site.get("GDC")

                line[8] = reader_site.get("HUG")

                line[9] = reader_site.get("TLF")

                if (reader_site.get("TL") != reader_lgerm.get("TL") && reader_lgerm.get("TL").length() > 0 && reader_site.get("TL").length() > 0) {

                        println "Different TL: $key_site\n site ="+reader_site.get("TL")+"\n lgerm="+reader_lgerm.get("TL");

                }

                if (reader_site.get("TL").length() > 0)

                        line[10] = reader_site.get("TL")

                else

                        line[10] = reader_lgerm.get("TL")

                //line[10] = reader_site.get("TLF")

                if (reader_site.get("lemma_source") != reader_lgerm.get("lemma_source")) {

                        line[11] = "DMF"

                }

                line[11] = reader_site.get("lemma_source")

                finalvalues[key_site] = line

                if (reader_site.readRecord()) key_site = reader_site.get("lemma")//+"_"+reader_site.get("category")

                else key_site = null;

                if (reader_lgerm.readRecord()) key_lgerm = reader_lgerm.get("lemma")//+"_"+reader_lgerm.get("category")

                else key_lgerm = null

        } else if (key_site.compareTo(key_lgerm) >= 0) {

                nLgerm++

                line[0] = reader_lgerm.get("lemma")

                line[1] = reader_lgerm.get("category")

                line[2] = reader_lgerm.get("AND")

                line[3] = reader_lgerm.get("DEAF")

                line[4] = reader_lgerm.get("DECT")

                line[5] = reader_lgerm.get("FEW")

                line[6] = reader_lgerm.get("GDF")

                line[7] = reader_lgerm.get("GDC")

                line[8] = reader_lgerm.get("HUG")

                line[9] = reader_lgerm.get("TLF")

                line[10] = reader_lgerm.get("TL")

                line[11] = reader_lgerm.get("lemma_source")

                finalvalues[key_lgerm] = line

                if (reader_lgerm.readRecord()) key_lgerm = reader_lgerm.get("lemma")//+"_"+reader_lgerm.get("category")

                else key_lgerm = null

        } else {

                nSite++

                /*// ignore entries only in DMF site

                line[0] = reader_site.get("lemma")

                line[1] = reader_site.get("category")

                line[1] = fixSiteCategory(line[1])

                line[2] = reader_site.get("AND")

                line[3] = reader_site.get("DEAF")

                line[4] = reader_site.get("DECT")

                line[5] = reader_site.get("FEW")

                line[6] = reader_site.get("GDF")

                line[7] = reader_site.get("GDC")

                line[8] = reader_site.get("HUG")

                line[9] = reader_site.get("TLF")

                line[10] = reader_site.get("TL")

                line[11] = reader_site.get("lemma_source")

                finalvalues[key_site] = line

                */

                if (reader_site.readRecord()) key_site = reader_site.get("lemma")//+"_"+reader_site.get("category")

                else key_site = null;

        }

}

// write last Lgerm lemmas

while (key_lgerm != null) {

        nLgerm++

        def line = ["","","","","","","","","","","",""]

        line[0] = reader_lgerm.get("lemma")

        line[1] = reader_lgerm.get("category")

        line[2] = reader_lgerm.get("AND")

        line[3] = reader_lgerm.get("DEAF")

        line[4] = reader_lgerm.get("DECT")

        line[5] = reader_lgerm.get("FEW")

        line[6] = reader_lgerm.get("GDF")

        line[7] = reader_lgerm.get("GDC")

        line[8] = reader_lgerm.get("HUG")

        line[9] = reader_lgerm.get("TLF")

        line[10] = reader_lgerm.get("TL")

        line[11] = reader_lgerm.get("lemma_source")

        finalvalues[key_lgerm] = line

        if (reader_lgerm.readRecord()) key_lgerm = reader_lgerm.get("lemma")//+"_"+reader_lgerm.get("category")

        else key_lgerm = null

}

// write last site lemmas

while (key_site != null) {

        nSite++

        def line = ["","","","","","","","","","","",""]

        line[0] = reader_site.get("lemma")

        line[1] = reader_site.get("category")

        line[1] = fixSiteCategory(line[1])

        line[2] = reader_site.get("AND")

        line[3] = reader_site.get("DEAF")

        line[4] = reader_site.get("DECT")

        line[5] = reader_site.get("FEW")

        line[6] = reader_site.get("GDF")

        line[7] = reader_site.get("GDC")

        line[8] = reader_site.get("HUG")

        line[9] = reader_site.get("TLF")

        line[10] = reader_site.get("TL")

        line[11] = reader_site.get("lemma_source")

        finalvalues[key_site] = line

        if (reader_site.readRecord()) key_site = reader_site.get("lemma")//+"_"+reader_site.get("category")

        else key_site = null;

}

def writer = IOUtils.getWriter(tsvFile);

writer.println("lemma        category        AND        DEAF        DECT        FEW        GDF        GDC        HUG        TLF        TL        lemma_source");

for (String key : finalvalues.keySet()) {

        writer.println(StringUtils.join(finalvalues.get(key), "\t"));

}

writer.close();

println "nMerge=$nMerge"

println "nLgerm=$nLgerm"

println "nSite=$nSite"

// END