root / tmp / org.txm.core / src / java / org / txm / importer / scripts / xmltxm / AnnotationInjection.groovy @ 1681
History | View | Annotate | Download (12.7 kB)
| 1 |
|
|---|---|
| 2 |
|
| 3 |
// Copyright © 2010-2013 ENS de Lyon.
|
| 4 |
// Copyright © 2007-2010 ENS de Lyon, CNRS, INRP, University of
|
| 5 |
// Lyon 2, University of Franche-Comté, University of Nice
|
| 6 |
// Sophia Antipolis, University of Paris 3.
|
| 7 |
//
|
| 8 |
// The TXM platform is free software: you can redistribute it
|
| 9 |
// and/or modify it under the terms of the GNU General Public
|
| 10 |
// License as published by the Free Software Foundation,
|
| 11 |
// either version 2 of the License, or (at your option) any
|
| 12 |
// later version.
|
| 13 |
//
|
| 14 |
// The TXM platform is distributed in the hope that it will be
|
| 15 |
// useful, but WITHOUT ANY WARRANTY; without even the implied
|
| 16 |
// warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
|
| 17 |
// PURPOSE. See the GNU General Public License for more
|
| 18 |
// details.
|
| 19 |
//
|
| 20 |
// You should have received a copy of the GNU General
|
| 21 |
// Public License along with the TXM platform. If not, see
|
| 22 |
// http://www.gnu.org/licenses.
|
| 23 |
//
|
| 24 |
//
|
| 25 |
//
|
| 26 |
// $LastChangedDate: 2016-03-29 09:51:35 +0200 (mar. 29 mars 2016) $
|
| 27 |
// $LastChangedRevision: 3185 $
|
| 28 |
// $LastChangedBy: mdecorde $
|
| 29 |
//
|
| 30 |
package org.txm.importer.scripts.xmltxm
|
| 31 |
|
| 32 |
import javax.xml.stream.* |
| 33 |
|
| 34 |
import org.txm.importer.StaxIdentityParser |
| 35 |
import org.txm.importer.filters.* |
| 36 |
import org.txm.importer.PersonalNamespaceContext |
| 37 |
import java.io.File |
| 38 |
import java.io.IOException |
| 39 |
|
| 40 |
import javax.xml.stream.XMLStreamException |
| 41 |
|
| 42 |
/**
|
| 43 |
* The Class AnnotationInjection.
|
| 44 |
*
|
| 45 |
* @author mdecorde
|
| 46 |
*
|
| 47 |
* inject annotation from a stand-off file into a xml-tei-txm
|
| 48 |
* file
|
| 49 |
*/
|
| 50 |
|
| 51 |
public class AnnotationInjection extends StaxIdentityParser { |
| 52 |
|
| 53 |
public static String TXMNS = "http://textometrie.org/1.0" |
| 54 |
|
| 55 |
/** The xml reader factory. */
|
| 56 |
private def factory; |
| 57 |
|
| 58 |
/** The links. */
|
| 59 |
private LinkedHashSet<String> links; |
| 60 |
boolean replace
|
| 61 |
|
| 62 |
/** The linkparsers. key=type*/
|
| 63 |
private LinkedHashMap<String, XMLStreamReader> linkparsers; |
| 64 |
|
| 65 |
/** The anaurl. */
|
| 66 |
private def anaurl; |
| 67 |
|
| 68 |
/** The anainput data. */
|
| 69 |
private def anainputData; |
| 70 |
|
| 71 |
/** The anafactory. */
|
| 72 |
private XMLInputFactory anafactory = XMLInputFactory.newInstance();
|
| 73 |
|
| 74 |
/** The anaparser. */
|
| 75 |
private XMLStreamReader anaparser;
|
| 76 |
private XMLStreamReader headerparser;
|
| 77 |
|
| 78 |
/** The resp stmt id. */
|
| 79 |
String respStmtID = ""; |
| 80 |
|
| 81 |
/** The present taxonomies. */
|
| 82 |
ArrayList<String> presentTaxonomies = new ArrayList(); |
| 83 |
|
| 84 |
/**
|
| 85 |
* Instantiates a new annotation injection.
|
| 86 |
*
|
| 87 |
* @param url the xml-tei-txm file
|
| 88 |
* @param anaurl the stand-off file
|
| 89 |
*/
|
| 90 |
public AnnotationInjection(URL url, URL anaurl) { |
| 91 |
this(url, anaurl, false) |
| 92 |
} |
| 93 |
|
| 94 |
/**
|
| 95 |
* Instantiates a new annotation injection.
|
| 96 |
*
|
| 97 |
* @param url the xml-tei-txm file
|
| 98 |
* @param anaurl the stand-off file
|
| 99 |
*/
|
| 100 |
public AnnotationInjection(URL url, URL anaurl, boolean replace) { |
| 101 |
super(url); // init reader and writer |
| 102 |
try {
|
| 103 |
this.anaurl = anaurl;
|
| 104 |
this.replace = replace
|
| 105 |
factory = XMLInputFactory.newInstance(); |
| 106 |
this.buildLinkParsers();// build a parser per linkgroup |
| 107 |
} catch (XMLStreamException ex) {
|
| 108 |
System.out.println(ex);
|
| 109 |
} catch (IOException ex) { |
| 110 |
System.out.println("IOException while parsing "); |
| 111 |
} |
| 112 |
} |
| 113 |
|
| 114 |
private void getHeaderInfos(String containertag, boolean captureTheTag) |
| 115 |
{
|
| 116 |
anainputData = new BufferedInputStream(anaurl.openStream()); |
| 117 |
headerparser = anafactory.createXMLStreamReader(anainputData); |
| 118 |
boolean start = false; |
| 119 |
String localname;
|
| 120 |
for (int event = headerparser.next(); event != XMLStreamConstants.END_DOCUMENT; event = headerparser.next()) |
| 121 |
{
|
| 122 |
//String prefix = headerparser.getNamespaceURI();
|
| 123 |
if (event == XMLStreamConstants.START_ELEMENT) {
|
| 124 |
localname = headerparser.getLocalName(); |
| 125 |
if (captureTheTag && localname == containertag) // start copy after the tag |
| 126 |
start = true;
|
| 127 |
if (start) { // copy header |
| 128 |
String prefix = headerparser.getPrefix();
|
| 129 |
if (prefix.length() > 0) |
| 130 |
writer.writeStartElement(Nscontext.getNamespaceURI(prefix), localname) |
| 131 |
else
|
| 132 |
writer.writeStartElement(localname); |
| 133 |
for (int i = 0 ; i < headerparser.getNamespaceCount(); i++) |
| 134 |
writer.writeNamespace(headerparser.getNamespacePrefix(i), headerparser.getNamespaceURI(i)); |
| 135 |
for (int i = 0 ; i < headerparser.getAttributeCount(); i++) |
| 136 |
writer.writeAttribute(headerparser.getAttributeLocalName(i), headerparser.getAttributeValue(i)); |
| 137 |
} |
| 138 |
if (!captureTheTag && localname == containertag) // start copy after the tag |
| 139 |
start = true;
|
| 140 |
} else if (event == XMLStreamConstants.END_ELEMENT) { |
| 141 |
localname = headerparser.getLocalName(); |
| 142 |
if (!captureTheTag && localname == containertag)
|
| 143 |
break;// stop looping |
| 144 |
|
| 145 |
if (start)
|
| 146 |
writer.writeEndElement(); |
| 147 |
|
| 148 |
if (captureTheTag && localname == containertag)
|
| 149 |
break;// stop looping |
| 150 |
} else if (event == XMLStreamConstants.CHARACTERS) { |
| 151 |
if (start)
|
| 152 |
writer.writeCharacters(headerparser.getText()); |
| 153 |
} else if (event == XMLStreamConstants.COMMENT) { |
| 154 |
if (start)
|
| 155 |
writer.writeComment(headerparser.getText()); |
| 156 |
} |
| 157 |
} |
| 158 |
headerparser.close(); |
| 159 |
} |
| 160 |
|
| 161 |
/**
|
| 162 |
* find all refs.
|
| 163 |
*
|
| 164 |
* @return the list of link parser
|
| 165 |
*/
|
| 166 |
private LinkedHashSet<String> findGrpLink() |
| 167 |
{
|
| 168 |
LinkedHashSet<String> links = new LinkedHashSet<String>(); |
| 169 |
anainputData = anaurl.openStream(); |
| 170 |
anaparser = anafactory.createXMLStreamReader(anainputData); |
| 171 |
|
| 172 |
for (int event = anaparser.next(); event != XMLStreamConstants.END_DOCUMENT; event = anaparser.next()) { |
| 173 |
if (event == XMLStreamConstants.START_ELEMENT) {
|
| 174 |
if (anaparser.getLocalName().equals("linkGrp")) { |
| 175 |
String targetsvalue = anaparser.getAttributeValue(0) |
| 176 |
|
| 177 |
if (links.contains(targetsvalue)) {
|
| 178 |
System.err.println("Warning: Multiple group declaration : "+targetsvalue+" has already been added, the first one will be used") |
| 179 |
} else {
|
| 180 |
links.add(targetsvalue); // add the taxonomy type
|
| 181 |
} |
| 182 |
} else if (anaparser.getLocalName().equals("respStmt")) { |
| 183 |
respStmtID = anaparser.getAttributeValue(0); // one attribute (id) only |
| 184 |
} |
| 185 |
} |
| 186 |
} |
| 187 |
anaparser.close(); |
| 188 |
return links;
|
| 189 |
} |
| 190 |
|
| 191 |
/**
|
| 192 |
* Builds the link parsers.
|
| 193 |
* I need to know what groups exists to build a parser per taxonomy and go to the first link element
|
| 194 |
*/
|
| 195 |
private void buildLinkParsers() { |
| 196 |
|
| 197 |
// link group of the standoff file
|
| 198 |
links = findGrpLink(); |
| 199 |
linkparsers = new LinkedHashMap<String, XMLStreamReader>(); |
| 200 |
|
| 201 |
// build one parser per link group
|
| 202 |
for (String link : links) { // build a parser per group |
| 203 |
anainputData = new BufferedInputStream(anaurl.openStream()); |
| 204 |
linkparsers.put(link, anafactory.createXMLStreamReader(anainputData)); |
| 205 |
} |
| 206 |
|
| 207 |
//for each parser
|
| 208 |
for (String link : links) { |
| 209 |
anaparser = linkparsers.get(link); |
| 210 |
for (int event = anaparser.next(); event != XMLStreamConstants.END_DOCUMENT; event = anaparser.next()) { |
| 211 |
if (event == XMLStreamConstants.START_ELEMENT) {
|
| 212 |
if (anaparser.getLocalName().equals("linkGrp")) { // position the parser to the right group |
| 213 |
String targetsvalue = anaparser.getAttributeValue(0) |
| 214 |
if (targetsvalue.equals(link)) {
|
| 215 |
break; // next element is a link start tag |
| 216 |
} |
| 217 |
} |
| 218 |
} |
| 219 |
} |
| 220 |
} |
| 221 |
} |
| 222 |
|
| 223 |
public boolean process(File outfile) throws XMLStreamException, IOException |
| 224 |
{
|
| 225 |
boolean ret = super.process(outfile); |
| 226 |
releaseLinkParsers(); |
| 227 |
} |
| 228 |
|
| 229 |
/**
|
| 230 |
* Release the link parsers.
|
| 231 |
*/
|
| 232 |
private void releaseLinkParsers() { |
| 233 |
if (linkparsers == null) return; |
| 234 |
|
| 235 |
for (String l : linkparsers.keySet()) { |
| 236 |
XMLStreamReader p = linkparsers.get(l); |
| 237 |
if (p != null) { |
| 238 |
try {
|
| 239 |
p.close(); |
| 240 |
} catch(Exception e) { |
| 241 |
println "** Can not close $l link parser $p: $e"
|
| 242 |
} |
| 243 |
} |
| 244 |
} |
| 245 |
} |
| 246 |
|
| 247 |
/**
|
| 248 |
* get the next tei:link value of a tei:LinkGrp.
|
| 249 |
*
|
| 250 |
* @param link the link
|
| 251 |
* @return the next ana
|
| 252 |
*/
|
| 253 |
private String getNextAnaValue(String link, String wordId) { |
| 254 |
anaparser = linkparsers.get(link); |
| 255 |
def m;
|
| 256 |
for (int event = anaparser.next(); event != XMLStreamConstants.END_DOCUMENT; event = anaparser.next()) { |
| 257 |
if (event == XMLStreamConstants.START_ELEMENT) {
|
| 258 |
if (anaparser.getLocalName().equals("link")) { |
| 259 |
String targetsvalue = anaparser.getAttributeValue(0) |
| 260 |
if ((m = targetsvalue =~ /#(.*) #(.*)/)) { // balise externe |
| 261 |
def g1 = m[0][1]; |
| 262 |
def g2 = m[0][2]; |
| 263 |
|
| 264 |
String anavalue = g2;
|
| 265 |
anavalue = anavalue.replace("<", "<") |
| 266 |
return anavalue;
|
| 267 |
} else {
|
| 268 |
System.err.println("Error: getNextAna(): link target is not well formed: = "+anaparser.getAttributeValue(0)); |
| 269 |
} |
| 270 |
} |
| 271 |
} |
| 272 |
} |
| 273 |
return ""; |
| 274 |
} |
| 275 |
|
| 276 |
/**
|
| 277 |
* build the ana tags of a word.
|
| 278 |
*
|
| 279 |
* @param wordId the word id
|
| 280 |
* @return the ana tag
|
| 281 |
*/
|
| 282 |
private void writeAnaTags(String wordId) |
| 283 |
{
|
| 284 |
String anabalises ="\n"; |
| 285 |
for (String link : links) { |
| 286 |
writer.writeStartElement(TXMNS, "ana");
|
| 287 |
writer.writeAttribute("resp", "#"+respStmtID); |
| 288 |
writer.writeAttribute("type", "#"+link); |
| 289 |
if (replace) {
|
| 290 |
if (linkparsers.containsKey(link)) { // order is important |
| 291 |
writer.writeCharacters(getNextAnaValue(link, wordId)); |
| 292 |
} else {
|
| 293 |
writer.writeCharacters(anaValues.get(link)); |
| 294 |
} |
| 295 |
} else { // ok no problem |
| 296 |
writer.writeCharacters(getNextAnaValue(link, wordId)); |
| 297 |
} |
| 298 |
|
| 299 |
writer.writeEndElement(); // txm:ana
|
| 300 |
} |
| 301 |
} |
| 302 |
|
| 303 |
String wordId;
|
| 304 |
HashMap<String, String> anaValues = new HashMap<String, String>(); |
| 305 |
boolean flagSourceDesc = false, flagW = false, flagAna = false; |
| 306 |
String type = null, resp = null, anaValue= ""; |
| 307 |
protected void processStartElement() { |
| 308 |
|
| 309 |
if (localname.equals("taxonomy")) { |
| 310 |
String taxo = parser.getAttributeValue(0) // taxonomy type |
| 311 |
presentTaxonomies.add(taxo); |
| 312 |
} else if (flagW && replace && localname.equals("ana")) { |
| 313 |
flagAna = true
|
| 314 |
anaValue= ""
|
| 315 |
type = null
|
| 316 |
resp = null
|
| 317 |
for (int i= 0 ; i < parser.getAttributeCount() ; i++ ) { |
| 318 |
if (parser.getAttributeLocalName(i) == "resp") { |
| 319 |
resp = parser.getAttributeValue(i); |
| 320 |
} else if (parser.getAttributeLocalName(i) == "type") { |
| 321 |
type = parser.getAttributeValue(i); |
| 322 |
} |
| 323 |
} |
| 324 |
if (type != null) type = type.substring(1); |
| 325 |
if (resp != null) resp = resp.substring(1); |
| 326 |
return; // don't write the "ana" start element |
| 327 |
} else if (localname.equals("w")) { |
| 328 |
for (int i= 0 ; i < parser.getAttributeCount() ; i++ ) { |
| 329 |
if (parser.getAttributeLocalName(i) == "id") { |
| 330 |
wordId = parser.getAttributeValue(i); |
| 331 |
break
|
| 332 |
} |
| 333 |
} |
| 334 |
flagW = true
|
| 335 |
anaValues.clear() |
| 336 |
} |
| 337 |
|
| 338 |
super.processStartElement();
|
| 339 |
} |
| 340 |
|
| 341 |
protected void processCharacters() { |
| 342 |
if (flagAna) anaValue += parser.getText();
|
| 343 |
else super.processCharacters(); // FORM CONTENT LOST !!!!!!!!!!!!! |
| 344 |
} |
| 345 |
|
| 346 |
boolean applicationWritten = false; |
| 347 |
boolean taxonomiesWritten = false; |
| 348 |
protected void processEndElement() { |
| 349 |
switch (parser.getLocalName()) {
|
| 350 |
case "w": |
| 351 |
writeAnaTags(wordId); |
| 352 |
flagW = false
|
| 353 |
break;
|
| 354 |
case "ana": |
| 355 |
if (flagAna && replace && type != null && resp != null && anaValue != null) { |
| 356 |
anaValues.put(type, anaValue) |
| 357 |
links.add(type) |
| 358 |
flagAna = false
|
| 359 |
return; // don't write the "ana" end element |
| 360 |
} |
| 361 |
flagAna = false
|
| 362 |
break;
|
| 363 |
|
| 364 |
case "appInfo": |
| 365 |
applicationWritten = true;
|
| 366 |
getHeaderInfos("appInfo", false); |
| 367 |
break;
|
| 368 |
|
| 369 |
case "classDecl": |
| 370 |
taxonomiesWritten = true;
|
| 371 |
getHeaderInfos("classDecl", false); |
| 372 |
break;
|
| 373 |
|
| 374 |
case "encodingDesc": |
| 375 |
if (!applicationWritten) {
|
| 376 |
writer.writeStartElement("appInfo");
|
| 377 |
getHeaderInfos("appInfo", false); |
| 378 |
writer.writeEndElement(); // appInfo
|
| 379 |
} |
| 380 |
if (!taxonomiesWritten) {
|
| 381 |
writer.writeStartElement("classDecl");
|
| 382 |
getHeaderInfos("classDecl", false); |
| 383 |
writer.writeEndElement(); // classDecl
|
| 384 |
} |
| 385 |
break;
|
| 386 |
|
| 387 |
case "titleStmt": |
| 388 |
if (flagSourceDesc) {
|
| 389 |
//output.write(this.respStmt+"\n")
|
| 390 |
getHeaderInfos("respStmt", true); |
| 391 |
flagSourceDesc = false;
|
| 392 |
break;
|
| 393 |
} |
| 394 |
break;
|
| 395 |
} |
| 396 |
super.processEndElement();
|
| 397 |
} |
| 398 |
|
| 399 |
/** The declarenamespace. */
|
| 400 |
boolean declarenamespace = false; |
| 401 |
|
| 402 |
/**
|
| 403 |
* Declare namespace.
|
| 404 |
*
|
| 405 |
* @return the java.lang. object
|
| 406 |
*/
|
| 407 |
private declareNamespace() {
|
| 408 |
if (!declarenamespace) {
|
| 409 |
writer.writeDefaultNamespace("http://www.tei-c.org/ns/1.0");
|
| 410 |
writer.writeNamespace("txm", TXMNS);
|
| 411 |
declarenamespace = true;
|
| 412 |
} |
| 413 |
} |
| 414 |
|
| 415 |
/**
|
| 416 |
* The main method.
|
| 417 |
*
|
| 418 |
* @param args the arguments
|
| 419 |
*/
|
| 420 |
public static void main(String[] args) { |
| 421 |
|
| 422 |
String rootDir = "~/xml/rgaqcj/"; |
| 423 |
new File(rootDir + "/injection/").mkdir(); |
| 424 |
|
| 425 |
def milestones = ["tagUsage", "pb", "lb","catRef"]// the tags who |
| 426 |
|
| 427 |
File srcfile = new File(rootDir, "/anainline/", "roland.xml"); |
| 428 |
File pos1file = new File(rootDir, "/pos/", "rolandTT1-w-ana.xml"); |
| 429 |
|
| 430 |
File src2file = new File(rootDir, "/injection/", "roland.xml"); |
| 431 |
File pos2file = new File(rootDir, "/pos/", "rolandTT2-w-ana.xml"); |
| 432 |
|
| 433 |
println("process file : " + srcfile + " with : " + pos1file); |
| 434 |
def builder = new AnnotationInjection(srcfile.toURI().toURL(), |
| 435 |
pos1file.toURI().toURL(), milestones); |
| 436 |
builder.transfomFile(new File(rootDir + "/injection/", "roland.xml")); |
| 437 |
|
| 438 |
println("process file : " + src2file + " with : " + pos1file); |
| 439 |
builder = new AnnotationInjection(src2file.toURI().toURL(), pos2file.toURI().toURL(),
|
| 440 |
milestones); |
| 441 |
builder.transfomFile(rootDir + "/injection/", "roland-FINAL.xml"); |
| 442 |
|
| 443 |
return;
|
| 444 |
} |
| 445 |
} |