Package org.languagetool.dev.conversion.cg

Examples of org.languagetool.dev.conversion.cg.CgCompositeTag$AnyTag

428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
  private ArrayList<CgSet> groupCompositeTags(CgSet target) {
    HashMap<String,ArrayList<CgCompositeTag>> bf = new HashMap<String,ArrayList<CgCompositeTag>>();
    HashMap<String,ArrayList<CgCompositeTag>> sf = new HashMap<String,ArrayList<CgCompositeTag>>();
    HashMap<String,CgCompositeTag> dict = new HashMap<String,CgCompositeTag>();  // dictionary of sorts
    for (CgCompositeTag ctag : target.tags) {
      CgCompositeTag postags = new CgCompositeTag();
      CgCompositeTag baseforms = new CgCompositeTag();
      CgCompositeTag surfaceforms = new CgCompositeTag();
      for (CgTag tag : ctag.tags) {
        if (isBaseForm(tag.tag)) {
          baseforms.addTag(tag);
        } else if (isSurfaceForm(tag.tag)) {
          surfaceforms.addTag(tag);
        } else if (isPostag(tag.tag)) {
          postags.addTag(tag);
        }
      }
      if (!postags.isEmpty()) {
        if (!baseforms.isEmpty()) {
          bf = (HashMap)smartPut(bf, postags.toString(), baseforms);
        }
        else if (!surfaceforms.isEmpty()) {  // assumes there won't be both sf and bf in the same composite tag
          sf = (HashMap)smartPut(sf, postags.toString(), surfaceforms);
        }
      }
      dict.put(postags.toString(), postags);
    }
    ArrayList<CgSet> ret = new ArrayList<CgSet>();
    for (String postagSet : bf.keySet()) {
      CgSet newSet = new CgSet(target);
      newSet.tags = new HashSet<CgCompositeTag>();
      ArrayList<CgCompositeTag> bfs = bf.get(postagSet);
      for (CgCompositeTag singleBf : bfs) {
        CgCompositeTag newTotalTag = new CgCompositeTag();
        for (CgTag tag : dict.get(postagSet).tags) {
          newTotalTag.addTag(tag);
        }
        for (CgTag tag : singleBf.tags) {
          newTotalTag.addTag(tag);
        }
        newSet.addCompositeTag(newTotalTag);
      }
      ret.add(newSet);
    }
View Full Code Here
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
    }
    HashSet<String> pos = new HashSet<String>();
    HashSet<String> base = new HashSet<String>();
    HashSet<String> surf = new HashSet<String>();
    for (CgCompositeTag ctag : token.compositeTags) {
      CgCompositeTag postagCompile = new CgCompositeTag();
      for (CgTag tag : ctag.tags) {
        if (isPostag(tag.tag)) {
          postagCompile.addTag(tag);
        } else if (isSurfaceForm(tag.tag)) {
          surf.add(tag.tag);
        } else if (isBaseForm(tag.tag)) {
          base.add(tag.tag);
        }
      }
      pos.add(postagCompile.toString());
    }
    if (pos.size() > 1 && (surf.size() > 1 || base.size() > 1)) {
      return false;
    }
    if (surf.size() > 1 && base.size() > 1) {
View Full Code Here
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
    // also, if there're composite tags, they should have the same postags component, and not both base and surface forms
    if (compositeTags.length != 0) {
      ArrayList<String> baseformsList = new ArrayList<String>();
      ArrayList<String> surfaceformsList = new ArrayList<String>();
      for (CgCompositeTag ctag : compositeTags) {
        CgCompositeTag postagCompiled = new CgCompositeTag();
        for (CgTag tag : ctag.tags) {
          if (isPostag(tag.tag)) {
            postagCompiled.addTag(tag);
          } else if (isBaseForm(tag.tag)) {
            baseformsList.add(tag.tag);
          } else if (isSurfaceForm(tag.tag)) {
            surfaceformsList.add(tag.tag);
          }
View Full Code Here
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
          for (CgTag tag : expandedSet1.single_tags) {
            for (CgTag tag2 : expandedSet2.single_tags) {
              if (tag.tag.equals(tag2.tag)) {
                newSet.addTag(tag);
              } else {
                CgCompositeTag ctag = new CgCompositeTag();
                ctag.addTag(tag);
                ctag.addTag(tag2);
                newSet.addCompositeTag(ctag);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet1.tags) {
            for (CgTag tag : expandedSet2.single_tags) {
              if (ctag.tags.contains(tag)) {
                newSet.addCompositeTag(ctag);
              } else {
                CgCompositeTag ctagnew = new CgCompositeTag();
                for (CgTag tag2 : ctag.tags) {
                  ctagnew.addTag(tag2);
                }
                ctagnew.addTag(tag);
                newSet.addCompositeTag(ctagnew);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet2.tags) {
            for (CgTag tag : expandedSet1.single_tags) {
              if (ctag.tags.contains(tag)) {
                newSet.addCompositeTag(ctag);
              } else {
                CgCompositeTag ctagnew = new CgCompositeTag();
                for (CgTag tag2 : ctag.tags) {
                  ctagnew.addTag(tag2);
                }
                ctagnew.addTag(tag);
                newSet.addCompositeTag(ctagnew);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet1.tags) {
            for (CgCompositeTag ctag2 : expandedSet2.tags) {
              CgCompositeTag ctagnew = new CgCompositeTag();
              for (CgTag tag : ctag.tags) {
                if (!ctagnew.tags.contains(tag)) {
                  ctagnew.addTag(tag);
                }
              }
              for (CgTag tag : ctag2.tags) {
                if (!ctagnew.tags.contains(tag)) {
                  ctagnew.addTag(tag);
                }
              }
              newSet.addCompositeTag(ctagnew);
            }
          }
View Full Code Here
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
  private List<CgSet> groupCompositeTags(CgSet target) {
    HashMap<String,ArrayList<CgCompositeTag>> bf = new HashMap<>();
    Map<String, ArrayList<CgCompositeTag>> sf = new HashMap<>();
    Map<String, CgCompositeTag> dict = new HashMap<>();  // dictionary of sorts
    for (CgCompositeTag ctag : target.tags) {
      CgCompositeTag postags = new CgCompositeTag();
      CgCompositeTag baseforms = new CgCompositeTag();
      CgCompositeTag surfaceforms = new CgCompositeTag();
      for (CgTag tag : ctag.tags) {
        if (isBaseForm(tag.tag)) {
          baseforms.addTag(tag);
        } else if (isSurfaceForm(tag.tag)) {
          surfaceforms.addTag(tag);
        } else if (isPostag(tag.tag)) {
          postags.addTag(tag);
        }
      }
      if (!postags.isEmpty()) {
        if (!baseforms.isEmpty()) {
          bf = (HashMap)smartPut(bf, postags.toString(), baseforms);
        }
        else if (!surfaceforms.isEmpty()) {  // assumes there won't be both sf and bf in the same composite tag
          sf = (HashMap)smartPut(sf, postags.toString(), surfaceforms);
        }
      }
      dict.put(postags.toString(), postags);
    }
    List<CgSet> ret = new ArrayList<>();
    for (String postagSet : bf.keySet()) {
      CgSet newSet = new CgSet(target);
      newSet.tags = new HashSet<>();
      Iterable<CgCompositeTag> bfs = bf.get(postagSet);
      for (CgCompositeTag singleBf : bfs) {
        CgCompositeTag newTotalTag = new CgCompositeTag();
        for (CgTag tag : dict.get(postagSet).tags) {
          newTotalTag.addTag(tag);
        }
        for (CgTag tag : singleBf.tags) {
          newTotalTag.addTag(tag);
        }
        newSet.addCompositeTag(newTotalTag);
      }
      ret.add(newSet);
    }
View Full Code Here
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
    }
    Collection<String> pos = new HashSet<>();
    Collection<String> base = new HashSet<>();
    Collection<String> surf = new HashSet<>();
    for (CgCompositeTag ctag : token.compositeTags) {
      CgCompositeTag postagCompile = new CgCompositeTag();
      for (CgTag tag : ctag.tags) {
        if (isPostag(tag.tag)) {
          postagCompile.addTag(tag);
        } else if (isSurfaceForm(tag.tag)) {
          surf.add(tag.tag);
        } else if (isBaseForm(tag.tag)) {
          base.add(tag.tag);
        }
      }
      pos.add(postagCompile.toString());
    }
    if (pos.size() > 1 && (surf.size() > 1 || base.size() > 1)) {
      return false;
    }
    if (surf.size() > 1 && base.size() > 1) {
View Full Code Here
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
    // also, if there're composite tags, they should have the same postags component, and not both base and surface forms
    if (compositeTags.length != 0) {
      ArrayList<String> baseformsList = new ArrayList<>();
      ArrayList<String> surfaceformsList = new ArrayList<>();
      for (CgCompositeTag ctag : compositeTags) {
        CgCompositeTag postagCompiled = new CgCompositeTag();
        for (CgTag tag : ctag.tags) {
          if (isPostag(tag.tag)) {
            postagCompiled.addTag(tag);
          } else if (isBaseForm(tag.tag)) {
            baseformsList.add(tag.tag);
          } else if (isSurfaceForm(tag.tag)) {
            surfaceformsList.add(tag.tag);
          }
View Full Code Here
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
          for (CgTag tag : expandedSet1.single_tags) {
            for (CgTag tag2 : expandedSet2.single_tags) {
              if (tag.tag.equals(tag2.tag)) {
                newSet.addTag(tag);
              } else {
                CgCompositeTag ctag = new CgCompositeTag();
                ctag.addTag(tag);
                ctag.addTag(tag2);
                newSet.addCompositeTag(ctag);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet1.tags) {
            for (CgTag tag : expandedSet2.single_tags) {
              if (ctag.tags.contains(tag)) {
                newSet.addCompositeTag(ctag);
              } else {
                CgCompositeTag ctagnew = new CgCompositeTag();
                for (CgTag tag2 : ctag.tags) {
                  ctagnew.addTag(tag2);
                }
                ctagnew.addTag(tag);
                newSet.addCompositeTag(ctagnew);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet2.tags) {
            for (CgTag tag : expandedSet1.single_tags) {
              if (ctag.tags.contains(tag)) {
                newSet.addCompositeTag(ctag);
              } else {
                CgCompositeTag ctagnew = new CgCompositeTag();
                for (CgTag tag2 : ctag.tags) {
                  ctagnew.addTag(tag2);
                }
                ctagnew.addTag(tag);
                newSet.addCompositeTag(ctagnew);
              }
            }
          }
          for (CgCompositeTag ctag : expandedSet1.tags) {
            for (CgCompositeTag ctag2 : expandedSet2.tags) {
              CgCompositeTag ctagnew = new CgCompositeTag();
              for (CgTag tag : ctag.tags) {
                if (!ctagnew.tags.contains(tag)) {
                  ctagnew.addTag(tag);
                }
              }
              for (CgTag tag : ctag2.tags) {
                if (!ctagnew.tags.contains(tag)) {
                  ctagnew.addTag(tag);
                }
              }
              newSet.addCompositeTag(ctagnew);
            }
          }
View Full Code Here
TOP

Related Classes of org.languagetool.dev.conversion.cg.CgCompositeTag$AnyTag

  • org.languagetool.dev.conversion.CgRuleConverter
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.