root / tmp / org.txm.groovy.core / src / groovy / org / txm / scripts / importer / xmltxm / compiler.groovy @ 1000
History | View | Annotate | Download (12.5 kB)
| 1 |
// Copyright © 2010-2013 ENS de Lyon.
|
|---|---|
| 2 |
// Copyright © 2007-2010 ENS de Lyon, CNRS, INRP, University of
|
| 3 |
// Lyon 2, University of Franche-Comté, University of Nice
|
| 4 |
// Sophia Antipolis, University of Paris 3.
|
| 5 |
//
|
| 6 |
// The TXM platform is free software: you can redistribute it
|
| 7 |
// and/or modify it under the terms of the GNU General Public
|
| 8 |
// License as published by the Free Software Foundation,
|
| 9 |
// either version 2 of the License, or (at your option) any
|
| 10 |
// later version.
|
| 11 |
//
|
| 12 |
// The TXM platform is distributed in the hope that it will be
|
| 13 |
// useful, but WITHOUT ANY WARRANTY; without even the implied
|
| 14 |
// warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
|
| 15 |
// PURPOSE. See the GNU General Public License for more
|
| 16 |
// details.
|
| 17 |
//
|
| 18 |
// You should have received a copy of the GNU General
|
| 19 |
// Public License along with the TXM platform. If not, see
|
| 20 |
// http://www.gnu.org/licenses.
|
| 21 |
//
|
| 22 |
//
|
| 23 |
//
|
| 24 |
// $LastChangedDate: 2016-05-26 17:42:36 +0200 (jeu. 26 mai 2016) $
|
| 25 |
// $LastChangedRevision: 3219 $
|
| 26 |
// $LastChangedBy: mdecorde $
|
| 27 |
//
|
| 28 |
|
| 29 |
|
| 30 |
package org.txm.scripts.importer.xmltxm;
|
| 31 |
|
| 32 |
import java.util.ArrayList;; |
| 33 |
|
| 34 |
import org.txm.*; |
| 35 |
import org.txm.core.engines.*; |
| 36 |
import org.txm.importer.cwb.BuildCwbEncodeArgs; |
| 37 |
import org.txm.importer.cwb.CwbEncode |
| 38 |
import org.txm.importer.cwb.CwbMakeAll |
| 39 |
import org.txm.scripts.importer.*; |
| 40 |
import org.txm.scripts.*; |
| 41 |
import org.txm.importer.scripts.xmltxm.*; |
| 42 |
import org.txm.utils.treetagger.TreeTagger; |
| 43 |
|
| 44 |
import javax.xml.stream.*; |
| 45 |
import java.net.URL; |
| 46 |
import java.io.File; |
| 47 |
import java.util.HashMap; |
| 48 |
import java.util.List; |
| 49 |
|
| 50 |
// TODO: Auto-generated Javadoc
|
| 51 |
/**
|
| 52 |
* The Class compiler.
|
| 53 |
*/
|
| 54 |
class compiler |
| 55 |
{
|
| 56 |
String sortMetadata;
|
| 57 |
/** The debug. */
|
| 58 |
private boolean debug= false; |
| 59 |
|
| 60 |
/** The input data. */
|
| 61 |
private def inputData; |
| 62 |
|
| 63 |
/** The factory. */
|
| 64 |
private def factory; |
| 65 |
|
| 66 |
/** The parser. */
|
| 67 |
private XMLStreamReader parser;
|
| 68 |
|
| 69 |
/** The dir. */
|
| 70 |
private def dir; |
| 71 |
|
| 72 |
/** The output. */
|
| 73 |
private Writer output; |
| 74 |
|
| 75 |
/** The url. */
|
| 76 |
private def url; |
| 77 |
|
| 78 |
/** The text. */
|
| 79 |
String text=""; |
| 80 |
|
| 81 |
/** The base. */
|
| 82 |
String base=""; |
| 83 |
|
| 84 |
/** The project. */
|
| 85 |
String project=""; |
| 86 |
|
| 87 |
/** The lang. */
|
| 88 |
private String lang ="fr"; |
| 89 |
|
| 90 |
/** The s attribs. */
|
| 91 |
private static HashMap<String, List<String>> sAttribs; |
| 92 |
|
| 93 |
/** The anatypes. */
|
| 94 |
private static anatypes = [] |
| 95 |
private static anavalues = [:] |
| 96 |
|
| 97 |
/**
|
| 98 |
* initialize.
|
| 99 |
*
|
| 100 |
*/
|
| 101 |
public compiler(){}
|
| 102 |
|
| 103 |
/**
|
| 104 |
* Instantiates a new compiler.
|
| 105 |
*
|
| 106 |
* @param url the url
|
| 107 |
* @param text the text
|
| 108 |
* @param base the base
|
| 109 |
* @param project the project
|
| 110 |
*/
|
| 111 |
public compiler(URL url,String text,String base, String project) |
| 112 |
{
|
| 113 |
this.text = text
|
| 114 |
this.base = base;
|
| 115 |
this.project = project;
|
| 116 |
try {
|
| 117 |
this.url = url;
|
| 118 |
inputData = url.openStream(); |
| 119 |
|
| 120 |
factory = XMLInputFactory.newInstance(); |
| 121 |
parser = factory.createXMLStreamReader(inputData); |
| 122 |
} catch (Exception ex) { |
| 123 |
System.out.println("Error while creating indexes: $ex"); |
| 124 |
ex.printStackTrace(); |
| 125 |
} |
| 126 |
} |
| 127 |
|
| 128 |
public void setSortMetadata(String sortMetadata) |
| 129 |
{
|
| 130 |
this.sortMetadata = sortMetadata;
|
| 131 |
} |
| 132 |
|
| 133 |
/**
|
| 134 |
* set the language of the corpus.
|
| 135 |
*
|
| 136 |
* @param lang the lang
|
| 137 |
* @return the java.lang. object
|
| 138 |
*/
|
| 139 |
public setLang(String lang) |
| 140 |
{
|
| 141 |
this.lang = lang;
|
| 142 |
} |
| 143 |
|
| 144 |
/**
|
| 145 |
* Creates the output.
|
| 146 |
*
|
| 147 |
* @param dirPathName the dir path name
|
| 148 |
* @param fileName the file name
|
| 149 |
* @return true, if successful
|
| 150 |
*/
|
| 151 |
private boolean createOutput(File f) { |
| 152 |
try {
|
| 153 |
//File f = new File(dirPathName, fileName)
|
| 154 |
output = new OutputStreamWriter(new BufferedOutputStream(new FileOutputStream(f,f.exists())) , "UTF-8"); |
| 155 |
return true; |
| 156 |
} catch (Exception e) { |
| 157 |
println "Error while create CQP otput file: "+e
|
| 158 |
e.printStackTrace(); |
| 159 |
|
| 160 |
return false; |
| 161 |
} |
| 162 |
} |
| 163 |
|
| 164 |
/**
|
| 165 |
* Go to text.
|
| 166 |
*/
|
| 167 |
private boolean GoToText() |
| 168 |
{
|
| 169 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) { |
| 170 |
if (event == XMLStreamConstants.END_ELEMENT)
|
| 171 |
if (parser.getLocalName().equals("teiHeader") || parser.getLocalName().equals("teiheader")) { |
| 172 |
return true; |
| 173 |
} |
| 174 |
} |
| 175 |
return false; |
| 176 |
} |
| 177 |
|
| 178 |
/**
|
| 179 |
* Transfom file cqp.
|
| 180 |
*
|
| 181 |
* @param dirPathName the dir path name
|
| 182 |
* @param fileName the file name
|
| 183 |
* @return true, if successful
|
| 184 |
*/
|
| 185 |
public boolean transfomFileCqp(File cqpFile) |
| 186 |
{
|
| 187 |
createOutput(cqpFile); |
| 188 |
String headvalue="" |
| 189 |
String vAna = ""; |
| 190 |
String vForm = ""; |
| 191 |
String wordid= ""; |
| 192 |
String vHead = ""; |
| 193 |
String anatype = null; |
| 194 |
String anavalue = null; |
| 195 |
int p_id = 0; |
| 196 |
int s_id = 0; |
| 197 |
|
| 198 |
boolean captureword = false; |
| 199 |
boolean flagForm = false; |
| 200 |
boolean flagAna = false; |
| 201 |
boolean inW = false; |
| 202 |
int wcounter = 1; |
| 203 |
if (!GoToText()) {
|
| 204 |
println "Error: no teiHeader tag found in text '"+this.text+"' (please check file format or content)" |
| 205 |
return false; |
| 206 |
} |
| 207 |
int wc = 0; // TEMP FOR TEST ONLY |
| 208 |
try {
|
| 209 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) |
| 210 |
{
|
| 211 |
switch (event) {
|
| 212 |
case XMLStreamConstants.START_ELEMENT:
|
| 213 |
//println "start: "+parser.getLocalName()
|
| 214 |
switch (parser.getLocalName()) {
|
| 215 |
case "w": |
| 216 |
|
| 217 |
inW = true;
|
| 218 |
anavalues = [:] |
| 219 |
wordid = parser.getAttributeValue(null, "id") |
| 220 |
if (wordid == null) |
| 221 |
wordid = "w_"+text+"_"+(wcounter++) |
| 222 |
|
| 223 |
vAna ="";
|
| 224 |
break;
|
| 225 |
|
| 226 |
case "form": |
| 227 |
String type2 = parser.getAttributeValue(null, "type"); |
| 228 |
if(type2 == null || type2.equals("default")) { |
| 229 |
flagForm = true;
|
| 230 |
vForm = "";
|
| 231 |
} else {
|
| 232 |
flagAna = true;
|
| 233 |
vAna += "\t";
|
| 234 |
if(!anatypes.contains(type2))
|
| 235 |
anatypes << type2; |
| 236 |
} |
| 237 |
|
| 238 |
break;
|
| 239 |
|
| 240 |
case "ana": |
| 241 |
flagAna = true;
|
| 242 |
anavalue = "";
|
| 243 |
anatype = parser.getAttributeValue(null, "type"); |
| 244 |
if (anatype != null) { |
| 245 |
if(anatype.startsWith("#")) |
| 246 |
anatype = anatype.substring(1)
|
| 247 |
break;
|
| 248 |
} |
| 249 |
break;
|
| 250 |
|
| 251 |
default:
|
| 252 |
if (!inW) {
|
| 253 |
output.write("<"+parser.getLocalName().toLowerCase());
|
| 254 |
if (!sAttribs.containsKey(parser.getLocalName()))
|
| 255 |
sAttribs.put(parser.getLocalName().toLowerCase(), []);
|
| 256 |
|
| 257 |
for (int i = 0 ; i < parser.getAttributeCount() ; i++) { |
| 258 |
String attrname = parser.getAttributeLocalName(i).toLowerCase();
|
| 259 |
String attrvalue = parser.getAttributeValue(i);
|
| 260 |
if (!(parser.getLocalName() == "text" && attrname == "id")) |
| 261 |
output.write(" "+attrname+"=\""+attrvalue.replace("\"", "'")+"\""); |
| 262 |
|
| 263 |
if (!sAttribs.get(parser.getLocalName().toLowerCase()).contains(attrname))
|
| 264 |
sAttribs.get(parser.getLocalName().toLowerCase()).add(attrname) |
| 265 |
} |
| 266 |
|
| 267 |
if (parser.getLocalName() == "text") { |
| 268 |
output.write(" id=\""+text+"\" base=\""+base+"\" project=\""+project+"\""); |
| 269 |
} |
| 270 |
output.write(">\n");
|
| 271 |
} |
| 272 |
} |
| 273 |
break;
|
| 274 |
|
| 275 |
case XMLStreamConstants.END_ELEMENT:
|
| 276 |
switch (parser.getLocalName()) {
|
| 277 |
case "TEI": |
| 278 |
break;
|
| 279 |
case "w": |
| 280 |
for (String t : anatypes) { |
| 281 |
def v = anavalues.get(t);
|
| 282 |
if (v != null) vAna +="\t"+v; |
| 283 |
else vAna +="\t"; |
| 284 |
} |
| 285 |
|
| 286 |
output.write( vForm.replaceAll("&", "&").replaceAll("<", "<") +"\t"+wordid+vAna+"\n"); |
| 287 |
vAna = "";
|
| 288 |
vForm = "";
|
| 289 |
inW = false;
|
| 290 |
break;
|
| 291 |
|
| 292 |
case "form": |
| 293 |
flagForm = false;
|
| 294 |
flagAna = false;
|
| 295 |
break;
|
| 296 |
|
| 297 |
case "ana": |
| 298 |
anavalues.put(anatype, anavalue) |
| 299 |
flagAna = false;
|
| 300 |
break;
|
| 301 |
|
| 302 |
default:
|
| 303 |
if(!inW)
|
| 304 |
output.write("</"+parser.getLocalName().toLowerCase()+">\n"); |
| 305 |
|
| 306 |
} |
| 307 |
break;
|
| 308 |
|
| 309 |
case XMLStreamConstants.CHARACTERS:
|
| 310 |
if(inW)
|
| 311 |
{
|
| 312 |
if(flagForm) {
|
| 313 |
vForm += parser.getText().trim(); |
| 314 |
} |
| 315 |
else if (flagAna) { |
| 316 |
anavalue += parser.getText().trim(); |
| 317 |
} |
| 318 |
} |
| 319 |
break;
|
| 320 |
} |
| 321 |
} |
| 322 |
|
| 323 |
output.close(); |
| 324 |
parser.close(); |
| 325 |
} |
| 326 |
catch (Exception ex) { |
| 327 |
System.out.println("Error while writing CQP file $ex"); |
| 328 |
ex.printStackTrace(); |
| 329 |
return false; |
| 330 |
} |
| 331 |
return true; |
| 332 |
} |
| 333 |
|
| 334 |
|
| 335 |
|
| 336 |
/**
|
| 337 |
* Run.
|
| 338 |
*
|
| 339 |
* @param rootDirFile the root dir file
|
| 340 |
* @param basename the basename
|
| 341 |
* @return true, if successful
|
| 342 |
*/
|
| 343 |
public boolean run(File binDir, File txmDir, String basename, String corpusname, List<File> files) |
| 344 |
{
|
| 345 |
anatypes = new ArrayList<String>();// init only 1 time |
| 346 |
anavalues = [:] |
| 347 |
sAttribs = new HashMap<String, List<String>>();// init only 1 time |
| 348 |
String rootDir = binDir.getAbsolutePath();
|
| 349 |
|
| 350 |
if (!(CwbEncode.isExecutableAvailable() && CwbMakeAll.isExecutableAvailable())) {
|
| 351 |
println ("Error: CWB executables not well set.")
|
| 352 |
return false; |
| 353 |
} |
| 354 |
if(!new File(rootDir).exists()) { |
| 355 |
println ("binary directory does not exists: "+rootDir)
|
| 356 |
return false; |
| 357 |
} |
| 358 |
|
| 359 |
File cqpFile = new File(binDir,"cqp/"+corpusname.toLowerCase()+".cqp"); |
| 360 |
new File(rootDir,"/cqp/").deleteDir(); |
| 361 |
new File(rootDir,"/cqp/").mkdir(); |
| 362 |
new File(rootDir,"data/"+corpusname).deleteDir(); |
| 363 |
new File(rootDir,"data/"+corpusname).mkdir(); |
| 364 |
new File(rootDir,"registry/").mkdir(); |
| 365 |
|
| 366 |
String textid = ""; |
| 367 |
int counttext = 0; |
| 368 |
//List<File> files = txmDir.listFiles();
|
| 369 |
|
| 370 |
// get all anatypes
|
| 371 |
for (File f : files) { |
| 372 |
getAnaTypes(f) |
| 373 |
} |
| 374 |
|
| 375 |
//0 set Lang
|
| 376 |
if (createOutput(cqpFile)) {
|
| 377 |
output.write("<txmcorpus lang=\""+lang+"\">\n"); |
| 378 |
output.close(); |
| 379 |
} |
| 380 |
//1- Transform into CQP file
|
| 381 |
def builder = null; |
| 382 |
for (File f : files) { |
| 383 |
counttext++; |
| 384 |
if (!f.exists()) {
|
| 385 |
println("file "+f+ " does not exists") |
| 386 |
} else {
|
| 387 |
//println("process file "+f)
|
| 388 |
String txtname = f.getName().substring(0,f.getName().length()-4); |
| 389 |
builder = new compiler(f.toURI().toURL(), txtname, corpusname.toLowerCase(), "default"); |
| 390 |
builder.setLang(lang); |
| 391 |
if(!builder.transfomFileCqp(cqpFile))
|
| 392 |
return false; |
| 393 |
} |
| 394 |
} |
| 395 |
|
| 396 |
//end corpus
|
| 397 |
if (createOutput(cqpFile)) {
|
| 398 |
output.write("</txmcorpus>\n");
|
| 399 |
output.close(); |
| 400 |
} |
| 401 |
|
| 402 |
//2- Import into CWB
|
| 403 |
|
| 404 |
CwbEncode cwbEn = new CwbEncode();
|
| 405 |
cwbEn.setDebug(debug); |
| 406 |
CwbMakeAll cwbMa = new CwbMakeAll();
|
| 407 |
cwbMa.setDebug(debug); |
| 408 |
List<String> pargs = ["id"]; |
| 409 |
for(String ana : anatypes) |
| 410 |
pargs.add(ana); |
| 411 |
|
| 412 |
List<String> sargs = []; |
| 413 |
//println "Found Sattributes "+this.sAttribs;
|
| 414 |
if(sAttribs.containsKey("text")) { |
| 415 |
if(!sAttribs.get("text").contains("id")) |
| 416 |
sAttribs.get("text").add("id"); |
| 417 |
if(!sAttribs.get("text").contains("base")) |
| 418 |
sAttribs.get("text").add("base"); |
| 419 |
if(!sAttribs.get("text").contains("project")) |
| 420 |
sAttribs.get("text").add("project"); |
| 421 |
} else {
|
| 422 |
sargs.add("text:0+id+base+project")
|
| 423 |
} |
| 424 |
|
| 425 |
if (sAttribs.containsKey("txmcorpus")) { |
| 426 |
if(!sAttribs.get("txmcorpus").contains("lang")) |
| 427 |
sAttribs.get("txmcorpus").add("lang"); |
| 428 |
} else {
|
| 429 |
sargs.add("txmcorpus:0+lang")
|
| 430 |
} |
| 431 |
|
| 432 |
for (String tag : this.sAttribs.keySet()) { |
| 433 |
String sAttr = tag;
|
| 434 |
if(sAttribs.get(tag).size() > 0) |
| 435 |
sAttr += ":";
|
| 436 |
for(String attr : sAttribs.get(tag)) |
| 437 |
sAttr +="+"+attr;
|
| 438 |
sargs.add(sAttr) |
| 439 |
} |
| 440 |
|
| 441 |
|
| 442 |
|
| 443 |
String[] sAttributes = sargs; |
| 444 |
String[] pAttributes = pargs; |
| 445 |
println "sAttributes : "+sAttributes;
|
| 446 |
println "pAttributes : "+pAttributes;
|
| 447 |
try {
|
| 448 |
String regPath = rootDir + "/registry/"+corpusname.toLowerCase() |
| 449 |
cwbEn.run( |
| 450 |
rootDir + "/data/$corpusname",
|
| 451 |
cqpFile.getAbsolutePath(), |
| 452 |
regPath, pAttributes, sAttributes); |
| 453 |
if (!new File(regPath).exists()) { |
| 454 |
println "Error: The registry file was not created: $regPath. See https://groupes.renater.fr/wiki/txm-users/public/faq"
|
| 455 |
return false; |
| 456 |
} |
| 457 |
cwbMa.run(corpusname, rootDir + "/registry");
|
| 458 |
|
| 459 |
} catch (Exception ex) { |
| 460 |
System.out.println("Error while creating indexes with CQP tools: $ex"); |
| 461 |
ex.printStackTrace(); |
| 462 |
return false; |
| 463 |
} |
| 464 |
|
| 465 |
return true; |
| 466 |
} |
| 467 |
|
| 468 |
/**
|
| 469 |
* Sets the debug.
|
| 470 |
*/
|
| 471 |
public void setDebug() |
| 472 |
{
|
| 473 |
this.debug = true; |
| 474 |
} |
| 475 |
|
| 476 |
/**
|
| 477 |
* The main method.
|
| 478 |
*
|
| 479 |
* @param args the arguments
|
| 480 |
*/
|
| 481 |
public static void main(String[] args) |
| 482 |
{
|
| 483 |
File dir = new File("~/xml/geo"); |
| 484 |
def c = new compiler(); |
| 485 |
c.setDebug(); |
| 486 |
c.setCwbPath("~/TXM/cwb/bin");
|
| 487 |
c.run(dir,"geo");
|
| 488 |
} |
| 489 |
|
| 490 |
private void getAnaTypes(File xmlFile) { |
| 491 |
inputData = xmlFile.toURI().toURL().openStream(); |
| 492 |
factory = XMLInputFactory.newInstance(); |
| 493 |
parser = factory.createXMLStreamReader(inputData); |
| 494 |
String ana = "ana" |
| 495 |
HashSet<String> types = new HashSet<String>(); |
| 496 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) { |
| 497 |
if (event == XMLStreamConstants.START_ELEMENT) { // start elem |
| 498 |
if (ana.equals(parser.getLocalName())) { // ana elem |
| 499 |
for (int i = 0 ; i < parser.getAttributeCount(); i++) { // find @type |
| 500 |
if ("type".equals(parser.getAttributeLocalName(i))) { // @type |
| 501 |
types.add(parser.getAttributeValue(i).substring(1)); //remove the # |
| 502 |
break;
|
| 503 |
} |
| 504 |
} |
| 505 |
} |
| 506 |
} |
| 507 |
} |
| 508 |
parser.close() |
| 509 |
|
| 510 |
for (String type : types) |
| 511 |
if (!anatypes.contains(type))
|
| 512 |
anatypes << type |
| 513 |
} |
| 514 |
} |