root / tmp / org.txm.groovy.core / src / groovy / org / txm / scripts / importer / transcriber / compiler.groovy @ 2361
History | View | Annotate | Download (18.3 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 |
// $LastChangedDate:$
|
| 25 |
// $LastChangedRevision:$
|
| 26 |
// $LastChangedBy:$
|
| 27 |
//
|
| 28 |
package org.txm.scripts.importer.transcriber
|
| 29 |
|
| 30 |
import java.io.File; |
| 31 |
|
| 32 |
import javax.xml.stream.*; |
| 33 |
|
| 34 |
import java.io.OutputStreamWriter; |
| 35 |
import java.util.LinkedHashMap; |
| 36 |
import org.txm.objects.* |
| 37 |
import org.txm.searchengine.cqp.corpus.* |
| 38 |
import org.txm.importer.cwb.CwbEncode |
| 39 |
import org.txm.importer.cwb.CwbMakeAll |
| 40 |
import org.txm.importer.cwb.CwbProcess; |
| 41 |
import org.txm.importer.cwb.PatchCwbRegistry; |
| 42 |
import org.txm.utils.ConsoleProgressBar |
| 43 |
import org.txm.utils.Pair; |
| 44 |
|
| 45 |
/**
|
| 46 |
* The Class compiler.
|
| 47 |
*/
|
| 48 |
class compiler { |
| 49 |
|
| 50 |
boolean ADD_TEXTID_TO_REF = true |
| 51 |
|
| 52 |
/** The input data. */
|
| 53 |
private def inputData; |
| 54 |
|
| 55 |
/** The factory. */
|
| 56 |
private def factory; |
| 57 |
|
| 58 |
/** The parser. */
|
| 59 |
private XMLStreamReader parser;
|
| 60 |
|
| 61 |
/** The output. */
|
| 62 |
OutputStreamWriter output;
|
| 63 |
|
| 64 |
/** The basename. */
|
| 65 |
String corpusname;
|
| 66 |
|
| 67 |
/** The projectname. */
|
| 68 |
String projectname
|
| 69 |
|
| 70 |
/** The outdir. */
|
| 71 |
String outdir;
|
| 72 |
|
| 73 |
/** The debug. */
|
| 74 |
boolean debug = false; |
| 75 |
|
| 76 |
/** The removeinterviewers. */
|
| 77 |
boolean removeinterviewers = false; |
| 78 |
|
| 79 |
/** The trans. */
|
| 80 |
HashMap<String, ArrayList<Pair<String, String>>> trans; |
| 81 |
|
| 82 |
/** The speakers. */
|
| 83 |
HashMap<String, ArrayList<Pair<String, String>>> speakers; |
| 84 |
|
| 85 |
/** The speakersname. */
|
| 86 |
HashMap<String, String> speakersname = new HashMap<String, String>(); |
| 87 |
|
| 88 |
/** The topics. */
|
| 89 |
HashMap<String, ArrayList<Pair<String, String>>> topics; |
| 90 |
|
| 91 |
|
| 92 |
/** The interviewers. */
|
| 93 |
ArrayList<String> interviewers = []; |
| 94 |
static HashSet<String> sectionAttrs; |
| 95 |
|
| 96 |
/** The anatypes. */
|
| 97 |
private static anatypes = [] |
| 98 |
private static anavalues = [:] |
| 99 |
|
| 100 |
/**
|
| 101 |
* Removes the interviewers.
|
| 102 |
*
|
| 103 |
* @param value the value
|
| 104 |
* @return the java.lang. object
|
| 105 |
*/
|
| 106 |
public removeInterviewers(boolean value) { |
| 107 |
this.removeinterviewers = value;
|
| 108 |
} |
| 109 |
|
| 110 |
File cqpFile
|
| 111 |
/**
|
| 112 |
* Run.
|
| 113 |
*
|
| 114 |
* @param xmlfiles the xmlfiles
|
| 115 |
* @param basename the basename
|
| 116 |
* @param projectname the projectname
|
| 117 |
* @param outdir the outdir
|
| 118 |
* @return true, if successful
|
| 119 |
*/
|
| 120 |
public boolean run(Project project, List<File> xmlfiles, String corpusname, String projectname, File binDir) { |
| 121 |
|
| 122 |
//println "run compiler with $xmlfiles, $basename and $outdir"
|
| 123 |
this.outdir = binDir;
|
| 124 |
this.corpusname = corpusname;
|
| 125 |
this.projectname = projectname;
|
| 126 |
|
| 127 |
anatypes = ["event"] // reset |
| 128 |
anavalues = [:] // reset
|
| 129 |
|
| 130 |
sectionAttrs = new HashSet<String>() // reset section attributs set |
| 131 |
|
| 132 |
CorpusBuild corpus = project.getCorpusBuild(project.getName(), MainCorpus.class); |
| 133 |
if (corpus != null) { |
| 134 |
if (project.getDoUpdate()) {
|
| 135 |
corpus.clean(); // remove old files
|
| 136 |
} else {
|
| 137 |
corpus.delete(); // remove old files and TXMResult children
|
| 138 |
} |
| 139 |
} else {
|
| 140 |
corpus = new MainCorpus(project);
|
| 141 |
corpus.setID(project.getName()); |
| 142 |
corpus.setName(project.getName()); |
| 143 |
} |
| 144 |
corpus.setDescription("Built with the XML-TRS import module");
|
| 145 |
|
| 146 |
cqpFile = new File(binDir,"cqp/"+corpusname+".cqp"); |
| 147 |
cqpFile.delete() |
| 148 |
new File(binDir,"cqp").mkdirs() |
| 149 |
new File(binDir,"data").mkdirs() |
| 150 |
new File(binDir,"registry").mkdirs() |
| 151 |
|
| 152 |
// get all anatypes
|
| 153 |
for (File f : xmlfiles) { |
| 154 |
getAnaTypes(f) |
| 155 |
} |
| 156 |
//println "ANATYPES: "+anatypes
|
| 157 |
if (!createOutput(cqpFile)) return false; |
| 158 |
output.write("<txmcorpus lang=\"fr\">\n")
|
| 159 |
output.close(); |
| 160 |
|
| 161 |
println("Compiling "+xmlfiles.size()+" files") |
| 162 |
ConsoleProgressBar cpb = new ConsoleProgressBar(xmlfiles.size())
|
| 163 |
for (File txmFile :xmlfiles) { |
| 164 |
if (txmFile.exists()) {
|
| 165 |
cpb.tick() |
| 166 |
if (!process(txmFile)) {
|
| 167 |
println("Failed to compile "+txmFile)
|
| 168 |
} |
| 169 |
} |
| 170 |
} |
| 171 |
cpb.done() |
| 172 |
if (!createOutput(cqpFile)) return false; |
| 173 |
output.write("</txmcorpus>\n")
|
| 174 |
output.close(); |
| 175 |
|
| 176 |
//2- Import into CWB
|
| 177 |
File registryFile = new File(binDir, "registry/"+corpusname.toLowerCase()) |
| 178 |
File dataDir = new File(binDir, "data/$corpusname") |
| 179 |
|
| 180 |
new File(binDir, "registry").mkdir(); |
| 181 |
if (!new File(binDir, "registry").exists()) { |
| 182 |
println "Can't create registry directory"
|
| 183 |
return false; |
| 184 |
} |
| 185 |
|
| 186 |
if (!(CwbEncode.isExecutableAvailable() && CwbMakeAll.isExecutableAvailable())) {
|
| 187 |
println ("Error: CWB executables not well set.")
|
| 188 |
return false; |
| 189 |
} |
| 190 |
CwbEncode cwbEn = new CwbEncode();
|
| 191 |
CwbMakeAll cwbMa = new CwbMakeAll();
|
| 192 |
cwbEn.setDebug(debug); |
| 193 |
cwbMa.setDebug(debug); |
| 194 |
|
| 195 |
String uAttr = "u:0+spkid+spk+scope+accent+s+time+check+dialect+type"; |
| 196 |
String textAttr ="text:0+base+project" |
| 197 |
if (trans != null) { |
| 198 |
for (String key : trans.keySet()) { |
| 199 |
for (Pair p : trans.get(key)) {
|
| 200 |
if (ignoreTranscriberMetadata) {
|
| 201 |
String meta =p.getFirst();
|
| 202 |
if ( meta != "scribe" && meta != "audio_filename" && |
| 203 |
meta != "version" && meta != "version_date") |
| 204 |
textAttr+="+"+meta
|
| 205 |
} else {
|
| 206 |
textAttr+="+"+p.getFirst()
|
| 207 |
} |
| 208 |
} |
| 209 |
break;
|
| 210 |
} |
| 211 |
} |
| 212 |
|
| 213 |
String sectionAttr = "div:0" // "div:0+id+topic+endtime+starttime+type" |
| 214 |
for (String attr : sectionAttrs) { |
| 215 |
sectionAttr += "+"+attr
|
| 216 |
} |
| 217 |
|
| 218 |
List<String> pargs = ["spk", "ref", "id", "entitytype", "entityid"] |
| 219 |
for (String ana : anatypes) pargs.add(ana) |
| 220 |
|
| 221 |
String[] pAttributes = pargs |
| 222 |
|
| 223 |
String[] sAttributes = ["txmcorpus:0+lang", uAttr , textAttr, "event:0+id+desc+type+extent", sectionAttr, "sp:0+id+speaker+endtime+starttime+overlap+time"]; |
| 224 |
|
| 225 |
println "pAttributes: $pAttributes"
|
| 226 |
println "sAttributes: $sAttributes"
|
| 227 |
//return;
|
| 228 |
try {
|
| 229 |
cwbEn.run(dataDir.getAbsolutePath(), |
| 230 |
cqpFile.getAbsolutePath(), |
| 231 |
registryFile.getAbsolutePath(), pAttributes, sAttributes); |
| 232 |
if (!registryFile.exists()) {
|
| 233 |
println "Error: The registry file was not created: $regPath. See https://groupes.renater.fr/wiki/txm-users/public/faq"
|
| 234 |
return false; |
| 235 |
} |
| 236 |
cwbMa.run(corpusname, registryFile.getParent()); |
| 237 |
|
| 238 |
} catch (Exception ex) {System.out.println(ex); return false;} |
| 239 |
|
| 240 |
return true; |
| 241 |
} |
| 242 |
|
| 243 |
/**
|
| 244 |
* Creates the output.
|
| 245 |
*
|
| 246 |
* @param dirPathName the dir path name
|
| 247 |
* @param fileName the file name
|
| 248 |
* @return true, if successful
|
| 249 |
*/
|
| 250 |
private boolean createOutput(File f){ |
| 251 |
try {
|
| 252 |
//File f = new File(dirPathName, fileName)
|
| 253 |
output = new OutputStreamWriter(new BufferedOutputStream(new FileOutputStream(f,f.exists())) , "UTF-8"); |
| 254 |
return true; |
| 255 |
} catch (Exception e) { |
| 256 |
System.out.println(e.getLocalizedMessage());
|
| 257 |
return false; |
| 258 |
} |
| 259 |
} |
| 260 |
|
| 261 |
/** The text_id. */
|
| 262 |
String text_id
|
| 263 |
|
| 264 |
/** The u opened. */
|
| 265 |
boolean uOpened = false; |
| 266 |
|
| 267 |
/** The idturn. */
|
| 268 |
int idturn = 1; |
| 269 |
|
| 270 |
/** The idsection. */
|
| 271 |
int idsection = 1; |
| 272 |
|
| 273 |
/** The idu. */
|
| 274 |
int idu = 1; |
| 275 |
|
| 276 |
/** The idevent. */
|
| 277 |
int idevent = 1; |
| 278 |
|
| 279 |
/** The events. */
|
| 280 |
List<String> events = []; |
| 281 |
static int vEntityId = 0; |
| 282 |
static int vEntityIdCount = 1; |
| 283 |
|
| 284 |
/**
|
| 285 |
* Process.
|
| 286 |
*
|
| 287 |
* @param xmlfile the xmlfile
|
| 288 |
* @return true, if successful
|
| 289 |
*/
|
| 290 |
private boolean process(File xmlfile) { |
| 291 |
text_id = xmlfile.getName(); |
| 292 |
text_id = text_id.substring(0, text_id.length() -4); |
| 293 |
|
| 294 |
idturn = 1;
|
| 295 |
idsection = 1;
|
| 296 |
idu = 1;
|
| 297 |
|
| 298 |
boolean flagAna;
|
| 299 |
boolean flagForm;
|
| 300 |
boolean flagWord;
|
| 301 |
String vWord=""; |
| 302 |
String vForm=""; |
| 303 |
String vAna=""; |
| 304 |
String vEvents = "N/A"; |
| 305 |
String vEntityType = "N/A" |
| 306 |
String wordid= ""; |
| 307 |
//def wordattributes = [:];
|
| 308 |
String anatype = ""; |
| 309 |
String anavalue = ""; |
| 310 |
|
| 311 |
String formatedTime;
|
| 312 |
|
| 313 |
LinkedHashMap<String, String> anahash = new LinkedHashMap<String, String>(); |
| 314 |
String currentType;
|
| 315 |
|
| 316 |
URL url = xmlfile.toURI().toURL();
|
| 317 |
inputData = url.openStream(); |
| 318 |
factory = XMLInputFactory.newInstance(); |
| 319 |
parser = factory.createXMLStreamReader(inputData); |
| 320 |
String filename = xmlfile.getName()
|
| 321 |
String textid = filename.substring(0, filename.length() - 4); |
| 322 |
|
| 323 |
createOutput(cqpFile); |
| 324 |
String localname;
|
| 325 |
|
| 326 |
//get all metadatas declared before Episode tag
|
| 327 |
speakers = new HashMap<String, ArrayList<Pair<String, String>>>(); |
| 328 |
trans = new HashMap<String, ArrayList<Pair<String, String>>>(); |
| 329 |
topics = new HashMap<String, ArrayList<Pair<String, String>>>(); |
| 330 |
//println "parse infos"
|
| 331 |
parseInfos(); |
| 332 |
|
| 333 |
// println "Trans: $trans"
|
| 334 |
// println "Topics: $topics"
|
| 335 |
// println "Speakers: $speakers"
|
| 336 |
// def transproperties = ""
|
| 337 |
// for (String key : trans.keySet()) {
|
| 338 |
// for (Pair p : trans.get(key))
|
| 339 |
// transproperties+="\t"+p.getSecond();
|
| 340 |
// break;
|
| 341 |
// }
|
| 342 |
// println "Trans properties: "+transproperties
|
| 343 |
List<String> localspeakers; |
| 344 |
|
| 345 |
|
| 346 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) { |
| 347 |
//print "event: "+event +" "
|
| 348 |
switch (event) {
|
| 349 |
case XMLStreamConstants.START_ELEMENT:
|
| 350 |
localname = parser.getLocalName(); |
| 351 |
//println localname
|
| 352 |
switch(localname) {
|
| 353 |
case "div": |
| 354 |
output.write("<div");
|
| 355 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 356 |
String name = parser.getAttributeLocalName(i).replace("_","").toLowerCase() |
| 357 |
output.write(" "+name+"=\""+parser.getAttributeValue(i)+"\""); |
| 358 |
sectionAttrs << name |
| 359 |
} |
| 360 |
output.write ">\n"
|
| 361 |
break;
|
| 362 |
case "sp": |
| 363 |
output.write("<sp");
|
| 364 |
writeAttributes(); |
| 365 |
output.write ">\n"
|
| 366 |
break;
|
| 367 |
case "u": |
| 368 |
output.write("<u");
|
| 369 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 370 |
String name = parser.getAttributeLocalName(i).replace("_","").toLowerCase() |
| 371 |
output.write(" "+name+"=\""+parser.getAttributeValue(i)+"\""); |
| 372 |
if (name == "time") formatedTime = parser.getAttributeValue(i) |
| 373 |
else if (name == "spk") u_name = parser.getAttributeValue(i) |
| 374 |
} |
| 375 |
output.write ">\n"
|
| 376 |
break;
|
| 377 |
case "event": |
| 378 |
output.write("<event");
|
| 379 |
writeAttributes(); |
| 380 |
output.write ">\n"
|
| 381 |
|
| 382 |
if (parser.getAttributeValue(null, "type") == "entities") { |
| 383 |
if (parser.getAttributeValue(null, "extent") == "begin") { |
| 384 |
vEntityType = parser.getAttributeValue(null, "desc"); |
| 385 |
vEntityId = vEntityIdCount++; |
| 386 |
} else {
|
| 387 |
vEntityType = "N/A";
|
| 388 |
vEntityId = 0;
|
| 389 |
} |
| 390 |
} else if (parser.getAttributeValue(null, "type") == "pronounce") { |
| 391 |
if (parser.getAttributeValue(null, "extent") == "begin") |
| 392 |
events.add(parser.getAttributeValue(null, "desc")) |
| 393 |
else if (parser.getAttributeValue(null, "extent") == "end") |
| 394 |
events.remove(parser.getAttributeValue(null, "desc")) |
| 395 |
vEvents = "";
|
| 396 |
for (String s : events) |
| 397 |
vEvents += s+"#";
|
| 398 |
if (vEvents.length() > 0) |
| 399 |
vEvents = vEvents.substring(0, vEvents.length()-1); |
| 400 |
else
|
| 401 |
vEvents = ""
|
| 402 |
} |
| 403 |
break;
|
| 404 |
case "w": |
| 405 |
for (int i = 0 ; i < parser.getAttributeCount(); i++) { |
| 406 |
if (parser.getAttributeLocalName(i).equals("id")) { |
| 407 |
wordid = parser.getAttributeValue(i); |
| 408 |
break;
|
| 409 |
} |
| 410 |
} |
| 411 |
anavalues = [:]; |
| 412 |
break;
|
| 413 |
case "form": |
| 414 |
flagForm = true;
|
| 415 |
vForm = "";
|
| 416 |
vAna ="";
|
| 417 |
break;
|
| 418 |
case "ana": |
| 419 |
flagAna = true;
|
| 420 |
anavalue = "";
|
| 421 |
for (int i = 0 ; i < parser.getAttributeCount(); i++) |
| 422 |
if (parser.getAttributeLocalName(i).equals("type")) { |
| 423 |
anatype = parser.getAttributeValue(i).substring(1);//remove the # |
| 424 |
break;
|
| 425 |
} |
| 426 |
break;
|
| 427 |
} |
| 428 |
break;
|
| 429 |
case XMLStreamConstants.END_ELEMENT:
|
| 430 |
localname = parser.getLocalName(); |
| 431 |
|
| 432 |
switch (localname) {
|
| 433 |
case "text": |
| 434 |
output.write("</text>\n")
|
| 435 |
break;
|
| 436 |
case "Topics": |
| 437 |
break;
|
| 438 |
case "Topic": |
| 439 |
break;
|
| 440 |
case "Speakers": |
| 441 |
break;
|
| 442 |
case "Speaker": |
| 443 |
break;
|
| 444 |
case "Episode": |
| 445 |
break;
|
| 446 |
case "div": |
| 447 |
output.write("</div>\n")
|
| 448 |
break;
|
| 449 |
case "sp": |
| 450 |
output.write("</sp>\n")
|
| 451 |
break;
|
| 452 |
case "u": |
| 453 |
output.write("</u>\n")
|
| 454 |
break;
|
| 455 |
case "event": |
| 456 |
output.write("</event>\n")
|
| 457 |
break;
|
| 458 |
case "form": |
| 459 |
flagForm = false;
|
| 460 |
break;
|
| 461 |
case "ana": |
| 462 |
anavalues.put(anatype, anavalue) |
| 463 |
flagAna = false;
|
| 464 |
break;
|
| 465 |
case "w": |
| 466 |
// concat spk id and ref
|
| 467 |
String isEnq = (interviewers.contains(u_name))?"*":""; |
| 468 |
String ref = (u_name+", "+formatedTime+""+isEnq) |
| 469 |
if (ADD_TEXTID_TO_REF) ref = textid+", "+ref |
| 470 |
vForm +="\t"+u_name+"\t"+ref |
| 471 |
|
| 472 |
// concat entity and entity ID
|
| 473 |
vAna+= "\t"+vEntityType+"\t"+vEntityId; |
| 474 |
|
| 475 |
//concat ana values
|
| 476 |
for (String type : anatypes) { |
| 477 |
def v = anavalues.get(type);
|
| 478 |
if (v == null) v = ""; |
| 479 |
|
| 480 |
if ("event" == type) { |
| 481 |
if (v.length() > 0) |
| 482 |
vAna+="\t#"+v;
|
| 483 |
else
|
| 484 |
vAna+="\t";
|
| 485 |
|
| 486 |
//concat <Event> values
|
| 487 |
if (vEvents != null && vEvents.length() > 0 && vEvents != "N/A") |
| 488 |
vAna += "#"+vEvents;
|
| 489 |
} else {
|
| 490 |
vAna+="\t"+v;
|
| 491 |
} |
| 492 |
} |
| 493 |
|
| 494 |
|
| 495 |
vForm = vForm.replaceAll("\n", "").replaceAll("&", "&").replaceAll("<", "<"); |
| 496 |
|
| 497 |
if (removeinterviewers) {
|
| 498 |
if (!interviewers.contains(u_name))
|
| 499 |
output.write(vForm+"\t"+wordid+vAna+"\n"); |
| 500 |
} else {
|
| 501 |
output.write(vForm+"\t"+wordid+vAna+"\n"); |
| 502 |
} |
| 503 |
|
| 504 |
vAna = "";
|
| 505 |
vForm = "";
|
| 506 |
break;
|
| 507 |
} |
| 508 |
break
|
| 509 |
case XMLStreamConstants.CHARACTERS:
|
| 510 |
if (flagForm)
|
| 511 |
vForm += parser.getText().trim(); |
| 512 |
if (flagAna) {
|
| 513 |
anavalue += parser.getText().trim(); |
| 514 |
} |
| 515 |
break;
|
| 516 |
} |
| 517 |
} |
| 518 |
|
| 519 |
parser.close(); |
| 520 |
inputData.close(); |
| 521 |
output.close(); |
| 522 |
return true; |
| 523 |
} |
| 524 |
|
| 525 |
/** The u_name. */
|
| 526 |
String u_name;
|
| 527 |
|
| 528 |
/**
|
| 529 |
* Write start tag.
|
| 530 |
*/
|
| 531 |
private void writeStartTag() { |
| 532 |
output.write("<"+parser.getLocalName());
|
| 533 |
writeAttributes(); |
| 534 |
output.write ">\n"
|
| 535 |
} |
| 536 |
|
| 537 |
/**
|
| 538 |
* Write attributes.
|
| 539 |
*/
|
| 540 |
private void writeAttributes() { |
| 541 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 542 |
output.write(" "+parser.getAttributeLocalName(i).replace("_","").toLowerCase()+"=\""+parser.getAttributeValue(i)+"\""); |
| 543 |
} |
| 544 |
} |
| 545 |
|
| 546 |
private void getAnaTypes(File xmlFile) { |
| 547 |
inputData = xmlFile.toURI().toURL().openStream(); |
| 548 |
factory = XMLInputFactory.newInstance(); |
| 549 |
parser = factory.createXMLStreamReader(inputData); |
| 550 |
String ana = "ana" |
| 551 |
HashSet<String> types = new HashSet<String>(); |
| 552 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) { |
| 553 |
if (event == XMLStreamConstants.START_ELEMENT) { // start elem |
| 554 |
if (ana.equals(parser.getLocalName())) { // ana elem |
| 555 |
for (int i = 0 ; i < parser.getAttributeCount(); i++) { // find @type |
| 556 |
if ("type".equals(parser.getAttributeLocalName(i))) { // @type |
| 557 |
types.add(parser.getAttributeValue(i).substring(1)); //remove the # |
| 558 |
break;
|
| 559 |
} |
| 560 |
} |
| 561 |
} |
| 562 |
} |
| 563 |
} |
| 564 |
if (parser != null) parser.close(); |
| 565 |
if (inputData != null) inputData.close(); |
| 566 |
|
| 567 |
for (String type : types) |
| 568 |
if (!anatypes.contains(type))
|
| 569 |
anatypes << type |
| 570 |
} |
| 571 |
|
| 572 |
/**
|
| 573 |
* Write start tag.
|
| 574 |
*
|
| 575 |
* @param id the id
|
| 576 |
*/
|
| 577 |
private void writeStartTag(int id) { |
| 578 |
output.write("<"+parser.getLocalName().toLowerCase());
|
| 579 |
output.write(" id=\""+id+"\""); |
| 580 |
writeAttributes(); |
| 581 |
output.write ">\n"
|
| 582 |
} |
| 583 |
|
| 584 |
/**
|
| 585 |
* Write end tag.
|
| 586 |
*/
|
| 587 |
private void writeEndTag() { |
| 588 |
output.write("</"+parser.getLocalName().toLowerCase()+">\n"); |
| 589 |
} |
| 590 |
|
| 591 |
/** The ignore transcriber metadata. */
|
| 592 |
boolean ignoreTranscriberMetadata = false; |
| 593 |
|
| 594 |
/**
|
| 595 |
* Sets the ignore transcriber metadata.
|
| 596 |
*
|
| 597 |
* @param state the new ignore transcriber metadata
|
| 598 |
*/
|
| 599 |
public void setIgnoreTranscriberMetadata(boolean state) { |
| 600 |
this.ignoreTranscriberMetadata = state;
|
| 601 |
} |
| 602 |
|
| 603 |
/**
|
| 604 |
* Parses the infos.
|
| 605 |
*/
|
| 606 |
private void parseInfos() { //until tag Episode |
| 607 |
String localname;
|
| 608 |
//assert(parser != null);
|
| 609 |
for (int event = parser.next(); event != XMLStreamConstants.END_DOCUMENT; event = parser.next()) { |
| 610 |
if (event == XMLStreamConstants.START_ELEMENT) {
|
| 611 |
localname = parser.getLocalName(); |
| 612 |
switch (localname) {
|
| 613 |
case "text": |
| 614 |
output.write("<text project=\""+projectname+"\" base=\""+corpusname+"\"") |
| 615 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 616 |
if (ignoreTranscriberMetadata) {
|
| 617 |
if (parser.getAttributeLocalName(i) != "scribe" && |
| 618 |
parser.getAttributeLocalName(i) != "audio_filename" &&
|
| 619 |
parser.getAttributeLocalName(i) != "version" &&
|
| 620 |
parser.getAttributeLocalName(i) != "version_date")
|
| 621 |
output.write(" "+parser.getAttributeLocalName(i).replace("_","").toLowerCase()+"=\""+parser.getAttributeValue(i)+"\""); |
| 622 |
} else {
|
| 623 |
output.write(" "+parser.getAttributeLocalName(i).replace("_","").toLowerCase()+"=\""+parser.getAttributeValue(i)+"\""); |
| 624 |
} |
| 625 |
} |
| 626 |
|
| 627 |
output.write ">\n"
|
| 628 |
|
| 629 |
ArrayList list = new ArrayList<Pair<String, String>>() |
| 630 |
trans.put("trans", list);
|
| 631 |
|
| 632 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 633 |
list.add(new Pair(parser.getAttributeLocalName(i).replace("_","").toLowerCase(), parser.getAttributeValue(i))); |
| 634 |
if(parser.getAttributeLocalName(i).startsWith("enq")) |
| 635 |
interviewers.add(parser.getAttributeValue(i)); |
| 636 |
} |
| 637 |
return
|
| 638 |
case "Topic": |
| 639 |
String id = parser.getAttributeValue(null, "id"); |
| 640 |
if (id != null) { |
| 641 |
ArrayList list = new ArrayList<Pair<String, String>>() |
| 642 |
topics.put(id, list); |
| 643 |
} else {
|
| 644 |
println "found tag $localname with no id"
|
| 645 |
} |
| 646 |
break;
|
| 647 |
case "Speaker": |
| 648 |
//case "Trans":
|
| 649 |
String id = parser.getAttributeValue(null, "id"); |
| 650 |
String name = parser.getAttributeValue(null, "name"); |
| 651 |
if (id != null && name != null) { |
| 652 |
speakersname.put(id, name); |
| 653 |
} else {
|
| 654 |
println "found tag $localname with no id ($id)or name ($name)"
|
| 655 |
return;
|
| 656 |
} |
| 657 |
if (id != null) { |
| 658 |
ArrayList list = new ArrayList<Pair<String, String>>() |
| 659 |
speakers.put(id, list); |
| 660 |
|
| 661 |
for (int i = 0 ; i < parser.getAttributeCount() ; i ++) { |
| 662 |
list.add(new Pair(parser.getAttributeLocalName(i), parser.getAttributeValue(i)));
|
| 663 |
} |
| 664 |
} else {
|
| 665 |
println "found tag $localname with no id"
|
| 666 |
return;
|
| 667 |
} |
| 668 |
break;
|
| 669 |
} |
| 670 |
} |
| 671 |
} |
| 672 |
} |
| 673 |
|
| 674 |
/**
|
| 675 |
* Sets the debug.
|
| 676 |
*/
|
| 677 |
public void setDebug() { |
| 678 |
debug = true;
|
| 679 |
} |
| 680 |
|
| 681 |
} |