Mercurial > hg > mpdl-group
annotate software/eXist/mpdl-modules/src/de/mpg/mpiwg/berlin/mpdl/lt/analyzer/MpdlNormalizer.java @ 6:2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
| author | Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de> |
|---|---|
| date | Tue, 08 Feb 2011 14:54:09 +0100 |
| parents | 408254cf2f1d |
| children | 1ec29fdd0db8 |
| rev | line source |
|---|---|
| 0 | 1 package de.mpg.mpiwg.berlin.mpdl.lt.analyzer; |
| 2 | |
| 3 import java.io.IOException; | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
4 import java.io.StringReader; |
| 0 | 5 import java.util.ArrayList; |
| 6 | |
| 7 import de.mpg.mpiwg.berlin.mpdl.exception.ApplicationException; | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
8 import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexAll; |
| 0 | 9 import de.mpg.mpiwg.berlin.mpdl.lt.doc.regularization.Regularization; |
| 10 import de.mpg.mpiwg.berlin.mpdl.lt.doc.regularization.RegularizationManager; | |
| 11 import de.mpg.mpiwg.berlin.mpdl.lt.general.Language; | |
| 12 | |
| 13 public class MpdlNormalizer { | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
14 public static int MODE_4LEXICA = 1; // normalization for lexica etc. which have sometimes only ascii in it |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
15 public static int MODE_4HUMAN_READERS = 2; // normalization for human readers |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
16 private int normMode = MODE_4LEXICA; // Default |
| 0 | 17 private String[] normFunctionsToUse = {"reg", "norm"}; // default is to use all of these normalization functions |
| 18 private String language; | |
| 19 private int[] offsets; | |
| 20 | |
| 21 public MpdlNormalizer(String[] normFunctionsToUse, String lang) { | |
| 22 this.normFunctionsToUse = normFunctionsToUse; | |
| 23 String language = Language.getInstance().getLanguageId(lang); | |
| 24 this.language = language; | |
| 25 } | |
| 26 | |
| 27 public MpdlNormalizer(String language) { | |
| 28 this.language = language; | |
| 29 } | |
| 30 | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
31 public void setNormMode(int normMode) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
32 this.normMode = normMode; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
33 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
34 |
| 0 | 35 /** |
| 36 * Applies the normalization rules in <code>language</code> to | |
| 37 * <code>s</code>, without offset tracking. | |
| 38 * | |
| 39 * @param s source string | |
| 40 * @return normalized string | |
| 41 */ | |
| 42 public String normalize(String s) throws ApplicationException { | |
| 43 String normStr = s; | |
| 44 if (useRegFunction()) { | |
| 45 // try to regularize the string to the norm form over predefined regularizations | |
| 46 RegularizationManager regManager = RegularizationManager.getInstance(); | |
| 47 ArrayList<Regularization> regs = regManager.findRegsByOrig(language, s); | |
| 48 if (regs != null && regs.size() > 0) { | |
| 49 Regularization reg = regs.get(0); // only one: the first one | |
| 50 String regNormStr = reg.getNorm(); | |
| 51 normStr = regNormStr; | |
| 52 } | |
| 53 } | |
| 54 if (useNormFunction()) { | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
55 // normalize the string by string replacements |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
56 if (normMode == MODE_4LEXICA) |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
57 normStr = normalize4Lexica(normStr, null); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
58 else if (normMode == MODE_4HUMAN_READERS) |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
59 normStr = normalize4HumanReaders(normStr); |
| 0 | 60 } |
| 61 return normStr; | |
| 62 } | |
| 63 | |
| 64 private boolean useRegFunction() { | |
| 65 boolean useReg = false; | |
| 66 for (int i=0; i< normFunctionsToUse.length; i++) { | |
| 67 String function = normFunctionsToUse[i]; | |
| 68 if (function.equals("reg")) | |
| 69 return true; | |
| 70 } | |
| 71 return useReg; | |
| 72 } | |
| 73 | |
| 74 private boolean useNormFunction() { | |
| 75 boolean useNorm = false; | |
| 76 for (int i=0; i< normFunctionsToUse.length; i++) { | |
| 77 String function = normFunctionsToUse[i]; | |
| 78 if (function.equals("norm")) | |
| 79 return true; | |
| 80 } | |
| 81 return useNorm; | |
| 82 } | |
| 83 | |
| 84 /** | |
| 85 * Applies the normalization rules in <code>language</code> to | |
| 86 * <code>s</code>, with offset tracking.<p> | |
| 87 * | |
| 88 * <strong>WARNING:</strong> | |
| 89 * Arboreal will not work properly if a normalization substitution | |
| 90 * replaces a source character with more than two target characters! | |
| 91 * This is simply a BUG, and should be fixed. Fortunately, however, | |
| 92 * one does not often need such a replacement.<p> | |
| 93 * | |
| 94 * @param s source string | |
| 95 * @param offsets character offset table | |
| 96 * @return normalized string | |
| 97 */ | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
98 private String normalize4Lexica(String s, int[] offsets) { |
| 0 | 99 this.offsets = offsets; |
| 100 if (language.equals("la") || language.equals("lat")) { | |
| 101 StringBuffer buf = new StringBuffer(); | |
| 102 int n = 0; | |
| 103 for (int i = 0; i < s.length(); i++) { | |
| 104 char c = s.charAt(i); | |
| 105 String replace = new String(); | |
| 106 switch (c) { | |
| 107 case 'j': replace = "i"; break; | |
| 108 case 'v': replace = "u"; break; | |
| 109 /* | |
| 110 * Linguistic note: /u/ and /v/ are rarely phonemic | |
| 111 * in Latin, as in alui 's/he nourished' vs. | |
| 112 * alvi 'of a belly', volui 's/he wished' or 'it rolled' | |
| 113 * vs. volvi 'to be rolled', (in)seruit 's/he joined | |
| 114 * together' vs. (in)servit 's/he serves'. | |
| 115 */ | |
| 116 case 'q': | |
| 117 if ((i < s.length() - 1) && (s.charAt(i + 1) == ';')) | |
| 118 replace = "qu"; | |
| 119 else | |
| 120 replace = "q"; | |
| 121 break; | |
| 122 case ';': | |
| 123 if ((i > 0) && (s.charAt(i - 1) == 'q')) | |
| 124 replace = "e"; | |
| 125 else if ((i == 0) || ! Character.isLetter(s.charAt(i - 1))) | |
| 126 replace = ";"; | |
| 127 else | |
| 128 replace = ""; | |
| 129 break; | |
| 130 case '\u0300': replace = ""; break; // COMBINING GRAVE ACCENT | |
| 131 case '\u0301': replace = ""; break; // COMBINING ACCUTE ACCENT | |
| 132 case '\u0302': replace = ""; break; // COMBINING CIRCUMFLEX ACCENT | |
| 133 | |
| 134 case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE | |
| 135 case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE | |
| 136 case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX | |
| 137 case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS | |
| 138 case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E | |
| 139 case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA | |
| 140 case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE | |
| 141 case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE | |
| 142 case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX | |
| 143 case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS | |
| 144 case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; | |
| 145 case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE | |
| 146 case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX | |
| 147 case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS | |
| 148 case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE | |
| 149 case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE | |
| 150 case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX | |
| 151 case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS | |
| 152 case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE | |
| 153 case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE | |
| 154 case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX | |
| 155 case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS | |
| 156 case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE | |
| 157 case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE | |
| 158 case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX | |
| 159 case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS | |
| 160 case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E | |
| 161 case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA | |
| 162 case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE | |
| 163 case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE | |
| 164 case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX | |
| 165 case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS | |
| 166 case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE | |
| 167 case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE | |
| 168 case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX | |
| 169 case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS | |
| 170 case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE | |
| 171 case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE | |
| 172 case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX | |
| 173 case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS | |
| 174 case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE | |
| 175 case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE | |
| 176 case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX | |
| 177 case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS | |
| 178 case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON | |
| 179 case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON | |
| 180 case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE | |
| 181 case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE | |
| 182 case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON | |
| 183 case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON | |
| 184 case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE | |
| 185 case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE | |
| 186 case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK | |
| 187 case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK | |
| 188 case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON | |
| 189 case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON | |
| 190 case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE | |
| 191 case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE | |
| 192 case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON | |
| 193 case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON | |
| 194 case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE | |
| 195 case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE | |
| 196 case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E | |
| 197 case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E | |
| 198 case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON | |
| 199 case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON | |
| 200 case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE | |
| 201 case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE | |
| 202 case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S | |
| 203 case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S | |
| 204 case '\u00ad': break; // soft hyphen | |
| 205 // new in MPDL project by J. Willenborg | |
| 206 case '\u1e14': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 207 case '\u1e15': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 208 case '\u1e16': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 209 case '\u1e17': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 210 case '\u1e18': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 211 case '\u1e19': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 212 case '\u1e1a': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 213 case '\u1e1b': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 214 case '\u1e1c': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 215 case '\u1e1d': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 216 case '\u1eb8': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 217 case '\u1eb9': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 218 case '\u1eba': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 219 case '\u1ebb': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 220 case '\u1ebc': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 221 case '\u1ebd': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 222 case '\u1ebe': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 223 case '\u1ebf': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 224 case '\u1ec0': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 225 case '\u1ec1': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 226 case '\u1ec2': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 227 case '\u1ec3': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 228 case '\u1ec4': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 229 case '\u1ec5': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 230 case '\u1ec6': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 231 case '\u1ec7': replace = "e"; break; // LATIN ... LETTER E WITH ... | |
| 232 // by Malcolm | |
| 233 case '\u2329': break; // BRA | |
| 234 case '\u232a': break; // KET | |
| 235 default: replace += c; break; | |
| 236 } | |
| 237 buf.append(replace); | |
| 238 // update offsets if replacement is a different length | |
| 239 if (offsets != null) { | |
| 240 int r = replace.length(); | |
| 241 if (r == 0) | |
| 242 this.offsets = arrayKill(this.offsets, i - n); | |
| 243 else if (r == 2) | |
| 244 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 245 n += 1 - r; | |
| 246 } | |
| 247 } | |
| 248 return buf.toString(); | |
| 249 } else if (language.equals("it")) { | |
| 250 // new Mpdl code: added by J. Willenborg: some of Malcolms code did not work without errors so it has to be taken away, also all latin stuff is imported | |
| 251 StringBuffer buf = new StringBuffer(); | |
| 252 int n = 0; | |
| 253 for (int i = 0; i < s.length(); i++) { | |
| 254 char c = s.charAt(i); | |
| 255 String replace = new String(); | |
| 256 switch (c) { | |
| 257 case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE | |
| 258 case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE | |
| 259 case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX | |
| 260 case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS | |
| 261 case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E | |
| 262 case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA | |
| 263 case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE | |
| 264 case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE | |
| 265 case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX | |
| 266 case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS | |
| 267 case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; | |
| 268 case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE | |
| 269 case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX | |
| 270 case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS | |
| 271 case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE | |
| 272 case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE | |
| 273 case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX | |
| 274 case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS | |
| 275 case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE | |
| 276 case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE | |
| 277 case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX | |
| 278 case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS | |
| 279 case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE | |
| 280 case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE | |
| 281 case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX | |
| 282 case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS | |
| 283 case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E | |
| 284 case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA | |
| 285 case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE | |
| 286 case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE | |
| 287 case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX | |
| 288 case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS | |
| 289 case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE | |
| 290 case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE | |
| 291 case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX | |
| 292 case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS | |
| 293 case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE | |
| 294 case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE | |
| 295 case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX | |
| 296 case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS | |
| 297 case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE | |
| 298 case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE | |
| 299 case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX | |
| 300 case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS | |
| 301 case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON | |
| 302 case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON | |
| 303 case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE | |
| 304 case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE | |
| 305 case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON | |
| 306 case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON | |
| 307 case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE | |
| 308 case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE | |
| 309 case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK | |
| 310 case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK | |
| 311 case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON | |
| 312 case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON | |
| 313 case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE | |
| 314 case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE | |
| 315 case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON | |
| 316 case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON | |
| 317 case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE | |
| 318 case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE | |
| 319 case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E | |
| 320 case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E | |
| 321 case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON | |
| 322 case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON | |
| 323 case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE | |
| 324 case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE | |
| 325 case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S | |
| 326 case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S | |
| 327 // new in MPDL project by J. Willenborg | |
| 328 case '\u1e8d': replace = "e"; break; // LATIN SMALL LETTER E WITH TILDE | |
| 329 default: replace += c; break; | |
| 330 } | |
| 331 buf.append(replace); | |
| 332 // update offsets if replacement is a different length | |
| 333 if (offsets != null) { | |
| 334 int r = replace.length(); | |
| 335 if (r == 0) this.offsets = arrayKill(this.offsets, i - n); | |
| 336 else if (r == 2) | |
| 337 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 338 n += 1 - r; | |
| 339 } | |
| 340 } | |
| 341 return buf.toString(); | |
| 342 // new Mpdl code: added by J. Willenborg: most of the latin replacements also in english | |
| 343 } else if (language.equals("en")) { | |
| 344 StringBuffer buf = new StringBuffer(); | |
| 345 int n = 0; | |
| 346 for (int i = 0; i < s.length(); i++) { | |
| 347 char c = s.charAt(i); | |
| 348 String replace = new String(); | |
| 349 switch (c) { | |
| 350 case '\u0300': replace = ""; break; // COMBINING GRAVE ACCENT | |
| 351 case '\u0301': replace = ""; break; // COMBINING ACCUTE ACCENT | |
| 352 case '\u0302': replace = ""; break; // COMBINING CIRCUMFLEX ACCENT | |
| 353 | |
| 354 case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE | |
| 355 case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE | |
| 356 case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX | |
| 357 case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS | |
| 358 case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E | |
| 359 case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA | |
| 360 case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE | |
| 361 case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE | |
| 362 case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX | |
| 363 case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS | |
| 364 case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; | |
| 365 case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE | |
| 366 case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX | |
| 367 case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS | |
| 368 case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE | |
| 369 case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE | |
| 370 case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX | |
| 371 case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS | |
| 372 case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE | |
| 373 case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE | |
| 374 case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX | |
| 375 case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS | |
| 376 case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE | |
| 377 case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE | |
| 378 case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX | |
| 379 case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS | |
| 380 case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E | |
| 381 case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA | |
| 382 case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE | |
| 383 case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE | |
| 384 case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX | |
| 385 case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS | |
| 386 case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE | |
| 387 case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE | |
| 388 case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX | |
| 389 case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS | |
| 390 case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE | |
| 391 case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE | |
| 392 case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX | |
| 393 case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS | |
| 394 case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE | |
| 395 case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE | |
| 396 case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX | |
| 397 case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS | |
| 398 case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON | |
| 399 case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON | |
| 400 case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE | |
| 401 case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE | |
| 402 case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON | |
| 403 case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON | |
| 404 case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE | |
| 405 case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE | |
| 406 case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK | |
| 407 case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK | |
| 408 case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON | |
| 409 case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON | |
| 410 case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE | |
| 411 case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE | |
| 412 case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON | |
| 413 case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON | |
| 414 case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE | |
| 415 case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE | |
| 416 case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E | |
| 417 case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E | |
| 418 case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON | |
| 419 case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON | |
| 420 case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE | |
| 421 case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE | |
| 422 case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S | |
| 423 case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S | |
| 424 // new in MPDL project by J. Willenborg | |
| 425 case '\u1e8d': replace = "e"; break; // LATIN SMALL LETTER E WITH TILDE | |
| 426 // by Malcolm | |
| 427 case '\u00ad': break; // soft hyphen | |
| 428 case '\u2329': break; // BRA | |
| 429 case '\u232a': break; // KET | |
| 430 default: replace += c; break; | |
| 431 } | |
| 432 buf.append(replace); | |
| 433 // update offsets if replacement is a different length | |
| 434 if (offsets != null) { | |
| 435 int r = replace.length(); | |
| 436 if (r == 0) | |
| 437 this.offsets = arrayKill(this.offsets, i - n); | |
| 438 else if (r == 2) | |
| 439 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 440 n += 1 - r; | |
| 441 } | |
| 442 } | |
| 443 return buf.toString(); | |
| 444 } else if (language.equals("fr")) { | |
| 445 // new Mpdl code: added by J. Willenborg: some of Malcolms code did not work without errors so it has to be taken away | |
| 446 StringBuffer buf = new StringBuffer(); | |
| 447 int n = 0; | |
| 448 for (int i = 0; i < s.length(); i++) { | |
| 449 char c = s.charAt(i); | |
| 450 String replace = new String(); | |
| 451 switch (c) { | |
| 452 case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E | |
| 453 case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S | |
| 454 case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S | |
| 455 case '\u00ad': break; // soft hyphen | |
| 456 case '-': break; | |
| 457 default: replace += c; break; | |
| 458 } | |
| 459 buf.append(replace); | |
| 460 // update offsets if replacement is a different length | |
| 461 if (offsets != null) { | |
| 462 int r = replace.length(); | |
| 463 if (r == 0) | |
| 464 this.offsets = arrayKill(this.offsets, i - n); | |
| 465 else if (r == 2) | |
| 466 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 467 n += 1 - r; | |
| 468 } | |
| 469 } | |
| 470 return buf.toString(); | |
| 471 } else if (language.equals("de")) { | |
| 472 StringBuffer buf = new StringBuffer(); | |
| 473 int n = 0; | |
| 474 for (int i = 0; i < s.length(); i++) { | |
| 475 char c = s.charAt(i); | |
| 476 String replace = new String(); | |
| 477 switch (c) { | |
| 478 case '\u00c4': replace = "Ae"; break; | |
| 479 case '\u00d6': replace = "Oe"; break; | |
| 480 case '\u00dc': replace = "Ue"; break; | |
| 481 case '\u00df': replace = "ss"; break; | |
| 482 case '\u00e4': replace = "ae"; break; | |
| 483 case '\u00f6': replace = "oe"; break; | |
| 484 case '\u00fc': replace = "ue"; break; | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
485 case '\u00ad': break; // soft hyphen |
| 0 | 486 case '\u00e9': replace = "e"; break; |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
487 // new in MPDL project by J. Willenborg |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
488 case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
489 // case '-': break; |
| 0 | 490 default: replace += c; break; |
| 491 } | |
| 492 buf.append(replace); | |
| 493 // update offsets if replacement is a different length | |
| 494 if (offsets != null) { | |
| 495 int r = replace.length(); | |
| 496 if (r == 0) | |
| 497 this.offsets = arrayKill(this.offsets, i - n); | |
| 498 else if (r == 2) | |
| 499 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 500 n += 1 - r; | |
| 501 } | |
| 502 } | |
| 503 return buf.toString(); | |
| 504 } else if (language.equals("zh")) { | |
| 505 StringBuffer buf = new StringBuffer(); | |
| 506 int n = 0; | |
| 507 for (int i = 0; i < s.length(); i++) { | |
| 508 char c = s.charAt(i); | |
| 509 String replace = new String(); | |
| 510 switch (c) { | |
| 511 case '\u00b9': replace = "1"; break; | |
| 512 case '\u00b2': replace = "2"; break; | |
| 513 case '\u00b3': replace = "3"; break; | |
| 514 case '\u2074': replace = "4"; break; | |
| 515 case '\u2075': replace = "5"; break; | |
| 516 // original by Malcolm Hyman: with the following replacements // TODO uncomment these 3 lines | |
| 517 // case '\u3000': replace = " "; break; | |
| 518 // case '\u3001': replace = ","; break; | |
| 519 // case '\u3002': replace = "."; break; | |
| 520 // case '\u200b': break; // BREAKS EVERYTHING! | |
| 521 default: replace += c; break; | |
| 522 } | |
| 523 buf.append(replace); | |
| 524 // update offsets if replacement is a different length | |
| 525 if (offsets != null) { | |
| 526 int r = replace.length(); | |
| 527 if (r == 0) | |
| 528 this.offsets = arrayKill(this.offsets, i - n); | |
| 529 else if (r == 2) | |
| 530 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 531 n += 1 - r; | |
| 532 } | |
| 533 } | |
| 534 return buf.toString(); | |
| 535 } else if (language.equals("akk") || | |
| 536 language.equals("qam") || | |
| 537 language.equals("qpc") || | |
| 538 language.equals("elx") || | |
| 539 language.equals("sux") || | |
| 540 language.equals("hit") || | |
| 541 language.equals("qhu") || | |
| 542 language.equals("peo") || | |
| 543 language.equals("uga") || | |
| 544 language.equals("ura") || | |
| 545 language.equals("qcu")) { | |
| 546 StringBuffer buf = new StringBuffer(); | |
| 547 int n = 0; | |
| 548 char last = '\u0000'; | |
| 549 for (int i = 0; i < s.length(); i++) { | |
| 550 char c = s.charAt(i); | |
| 551 c = Character.toLowerCase(c); | |
| 552 String replace = new String(); | |
| 553 switch (c) { | |
| 554 case '{': replace += "-"; break; | |
| 555 case '}': replace += "-"; break; | |
| 556 // These are from PSD::ATF::Unicode by Steve Tinney | |
| 557 case '\u0161': replace += "sz"; break; | |
| 558 case '\u1e63': replace += "s,"; break; | |
| 559 case '\u1e6d': replace += "t,"; break; | |
| 560 case '\u014b': replace += "j"; break; | |
| 561 case '\u015b': replace += "s'"; break; | |
| 562 case '\u2080': replace += "0"; break; | |
| 563 case '\u2081': replace += "1"; break; | |
| 564 case '\u2082': replace += "2"; break; | |
| 565 case '\u2083': replace += "3"; break; | |
| 566 case '\u2084': replace += "4"; break; | |
| 567 case '\u2085': replace += "5"; break; | |
| 568 case '\u2086': replace += "6"; break; | |
| 569 case '\u2087': replace += "7"; break; | |
| 570 case '\u2088': replace += "8"; break; | |
| 571 case '\u2089': replace += "9"; break; | |
| 572 | |
| 573 case 'c': // shin (except where used as modifier) | |
| 574 if ((i > 0) && ((last == '~') || (last == '@'))) | |
| 575 replace += "c"; | |
| 576 else replace += "sz"; | |
| 577 break; | |
| 578 default: replace += c; break; | |
| 579 } | |
| 580 // suppress grapheme boundary before or after word boundary | |
| 581 if (replace.equals("-")) { | |
| 582 if ((i + 1 == s.length()) || (s.charAt(i + 1) == ' ') || (i == 0) || (s.charAt(i - 1) == ' ')) | |
| 583 replace = ""; | |
| 584 } | |
| 585 last = c; | |
| 586 buf.append(replace); | |
| 587 // update offsets if replacement is a different length | |
| 588 if (offsets != null) { | |
| 589 int r = replace.length(); | |
| 590 if (r == 0) | |
| 591 this.offsets = arrayKill(this.offsets, i - n); | |
| 592 else if (r == 2) | |
| 593 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 594 n += 1 - r; | |
| 595 } | |
| 596 } | |
| 597 return buf.toString(); | |
| 598 } else if (language.equals("el") || language.equals("grc")) { | |
| 599 StringBuffer buf = new StringBuffer(); | |
| 600 int n = 0; | |
| 601 for (int i = 0; i < s.length(); i++) { | |
| 602 char c = s.charAt(i); | |
| 603 String replace = new String(); | |
| 604 switch (c) { | |
| 605 case '\u03c2': replace = "\u03c3"; break; // GREEK SMALL LETTER FINAL SIGMA | |
| 606 case '<': break; | |
| 607 case '>': break; | |
| 608 case '[': break; | |
| 609 case ']': break; | |
| 610 case '1': break; | |
| 611 case '2': break; | |
| 612 case '\u03ac': replace = "\u1f71"; break; | |
| 613 case '\u03ad': replace = "\u1f73"; break; | |
| 614 case '\u03ae': replace = "\u1f75"; break; | |
| 615 case '\u03af': replace = "\u1f77"; break; | |
| 616 case '\u03cc': replace = "\u1f79"; break; | |
| 617 case '\u03cd': replace = "\u1f7b"; break; | |
| 618 case '\u03ce': replace = "\u1f7d"; break; | |
| 619 case '-': break; // same treatment as soft hyphen | |
| 620 case '\u00ad': break; // soft hyphen | |
| 621 default: replace += c; break; | |
| 622 } | |
| 623 buf.append(replace); | |
| 624 // update offsets if replacement is a different length | |
| 625 if (offsets != null) { | |
| 626 int r = replace.length(); | |
| 627 if (r == 0) | |
| 628 this.offsets = arrayKill(this.offsets, i - n); | |
| 629 else if (r == 2) | |
| 630 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 631 n += 1 - r; | |
| 632 } | |
| 633 } | |
| 634 return buf.toString(); | |
| 635 } else if (language.equals("el_atonic")) { | |
| 636 StringBuffer buf = new StringBuffer(); | |
| 637 int n = 0; | |
| 638 for (int i = 0; i < s.length(); i++) { | |
| 639 char c = s.charAt(i); | |
| 640 String replace = new String(); | |
| 641 switch (c) { | |
| 642 case '\u03c2': replace = "\u03c3"; break; // GREEK SMALL LETTER FINAL SIGMA | |
| 643 // map characters with diacritics to their plain equivalent | |
| 644 // cf. <code>BetaCode.java</code> | |
| 645 case '\u03aa': replace = "\u0399"; break; | |
| 646 case '\u03ab': replace = "\u03a5"; break; | |
| 647 case '\u03ac': replace = "\u0381"; break; | |
| 648 case '\u03ad': replace = "\u0385"; break; | |
| 649 case '\u03ae': replace = "\u0387"; break; | |
| 650 case '\u03af': replace = "\u0389"; break; | |
| 651 case '\u03ca': replace = "\u03b9"; break; | |
| 652 case '\u03cb': replace = "\u03c5"; break; | |
| 653 case '\u03cc': replace = "\u03bf"; break; | |
| 654 case '\u03cd': replace = "\u03c5"; break; | |
| 655 case '\u03ce': replace = "\u03c9"; break; | |
| 656 case '\u1f00': replace = "\u03b1"; break; | |
| 657 case '\u1f01': replace = "\u03b1"; break; | |
| 658 case '\u1f02': replace = "\u03b1"; break; | |
| 659 case '\u1f03': replace = "\u03b1"; break; | |
| 660 case '\u1f04': replace = "\u03b1"; break; | |
| 661 case '\u1f05': replace = "\u03b1"; break; | |
| 662 case '\u1f06': replace = "\u03b1"; break; | |
| 663 case '\u1f07': replace = "\u03b1"; break; | |
| 664 case '\u1f08': replace = "\u0391"; break; | |
| 665 case '\u1f09': replace = "\u0391"; break; | |
| 666 case '\u1f0a': replace = "\u0391"; break; | |
| 667 case '\u1f0b': replace = "\u0391"; break; | |
| 668 case '\u1f0c': replace = "\u0391"; break; | |
| 669 case '\u1f0d': replace = "\u0391"; break; | |
| 670 case '\u1f0e': replace = "\u0391"; break; | |
| 671 case '\u1f0f': replace = "\u0391"; break; | |
| 672 case '\u1f10': replace = "\u03b5"; break; | |
| 673 case '\u1f11': replace = "\u03b5"; break; | |
| 674 case '\u1f12': replace = "\u03b5"; break; | |
| 675 case '\u1f13': replace = "\u03b5"; break; | |
| 676 case '\u1f14': replace = "\u03b5"; break; | |
| 677 case '\u1f15': replace = "\u03b5"; break; | |
| 678 case '\u1f18': replace = "\u0395"; break; | |
| 679 case '\u1f19': replace = "\u0395"; break; | |
| 680 case '\u1f1a': replace = "\u0395"; break; | |
| 681 case '\u1f1b': replace = "\u0395"; break; | |
| 682 case '\u1f1c': replace = "\u0395"; break; | |
| 683 case '\u1f1d': replace = "\u0395"; break; | |
| 684 case '\u1f20': replace = "\u03b7"; break; | |
| 685 case '\u1f21': replace = "\u03b7"; break; | |
| 686 case '\u1f22': replace = "\u03b7"; break; | |
| 687 case '\u1f23': replace = "\u03b7"; break; | |
| 688 case '\u1f24': replace = "\u03b7"; break; | |
| 689 case '\u1f25': replace = "\u03b7"; break; | |
| 690 case '\u1f26': replace = "\u03b7"; break; | |
| 691 case '\u1f27': replace = "\u03b7"; break; | |
| 692 case '\u1f28': replace = "\u0397"; break; | |
| 693 case '\u1f29': replace = "\u0397"; break; | |
| 694 case '\u1f2a': replace = "\u0397"; break; | |
| 695 case '\u1f2b': replace = "\u0397"; break; | |
| 696 case '\u1f2c': replace = "\u0397"; break; | |
| 697 case '\u1f2d': replace = "\u0397"; break; | |
| 698 case '\u1f2e': replace = "\u0397"; break; | |
| 699 case '\u1f2f': replace = "\u0397"; break; | |
| 700 case '\u1f30': replace = "\u03b9"; break; | |
| 701 case '\u1f31': replace = "\u03b9"; break; | |
| 702 case '\u1f32': replace = "\u03b9"; break; | |
| 703 case '\u1f33': replace = "\u03b9"; break; | |
| 704 case '\u1f34': replace = "\u03b9"; break; | |
| 705 case '\u1f35': replace = "\u03b9"; break; | |
| 706 case '\u1f36': replace = "\u03b9"; break; | |
| 707 case '\u1f37': replace = "\u03b9"; break; | |
| 708 case '\u1f38': replace = "\u0399"; break; | |
| 709 case '\u1f39': replace = "\u0399"; break; | |
| 710 case '\u1f3a': replace = "\u0399"; break; | |
| 711 case '\u1f3b': replace = "\u0399"; break; | |
| 712 case '\u1f3c': replace = "\u0399"; break; | |
| 713 case '\u1f3d': replace = "\u0399"; break; | |
| 714 case '\u1f3e': replace = "\u0399"; break; | |
| 715 case '\u1f3f': replace = "\u0399"; break; | |
| 716 case '\u1f40': replace = "\u03bf"; break; | |
| 717 case '\u1f41': replace = "\u03bf"; break; | |
| 718 case '\u1f42': replace = "\u03bf"; break; | |
| 719 case '\u1f43': replace = "\u03bf"; break; | |
| 720 case '\u1f44': replace = "\u03bf"; break; | |
| 721 case '\u1f45': replace = "\u03bf"; break; | |
| 722 case '\u1f48': replace = "\u039f"; break; | |
| 723 case '\u1f49': replace = "\u039f"; break; | |
| 724 case '\u1f4a': replace = "\u039f"; break; | |
| 725 case '\u1f4b': replace = "\u039f"; break; | |
| 726 case '\u1f4c': replace = "\u039f"; break; | |
| 727 case '\u1f4d': replace = "\u039f"; break; | |
| 728 case '\u1f50': replace = "\u03c5"; break; | |
| 729 case '\u1f51': replace = "\u03c5"; break; | |
| 730 case '\u1f52': replace = "\u03c5"; break; | |
| 731 case '\u1f53': replace = "\u03c5"; break; | |
| 732 case '\u1f54': replace = "\u03c5"; break; | |
| 733 case '\u1f55': replace = "\u03c5"; break; | |
| 734 case '\u1f56': replace = "\u03c5"; break; | |
| 735 case '\u1f57': replace = "\u03c5"; break; | |
| 736 case '\u1f58': replace = "\u03a5"; break; | |
| 737 case '\u1f59': replace = "\u03a5"; break; | |
| 738 case '\u1f5a': replace = "\u03a5"; break; | |
| 739 case '\u1f5b': replace = "\u03a5"; break; | |
| 740 case '\u1f5c': replace = "\u03a5"; break; | |
| 741 case '\u1f5d': replace = "\u03a5"; break; | |
| 742 case '\u1f5e': replace = "\u03a5"; break; | |
| 743 case '\u1f5f': replace = "\u03a5"; break; | |
| 744 case '\u1f60': replace = "\u03c9"; break; | |
| 745 case '\u1f61': replace = "\u03c9"; break; | |
| 746 case '\u1f62': replace = "\u03c9"; break; | |
| 747 case '\u1f63': replace = "\u03c9"; break; | |
| 748 case '\u1f64': replace = "\u03c9"; break; | |
| 749 case '\u1f65': replace = "\u03c9"; break; | |
| 750 case '\u1f66': replace = "\u03c9"; break; | |
| 751 case '\u1f67': replace = "\u03c9"; break; | |
| 752 case '\u1f68': replace = "\u03a9"; break; | |
| 753 case '\u1f69': replace = "\u03a9"; break; | |
| 754 case '\u1f6a': replace = "\u03a9"; break; | |
| 755 case '\u1f6b': replace = "\u03a9"; break; | |
| 756 case '\u1f6c': replace = "\u03a9"; break; | |
| 757 case '\u1f6d': replace = "\u03a9"; break; | |
| 758 case '\u1f6e': replace = "\u03a9"; break; | |
| 759 case '\u1f6f': replace = "\u03a9"; break; | |
| 760 case '\u1f70': replace = "\u03b1"; break; | |
| 761 case '\u1f71': replace = "\u03b1"; break; | |
| 762 case '\u1f72': replace = "\u03b5"; break; | |
| 763 case '\u1f73': replace = "\u03b5"; break; | |
| 764 case '\u1f74': replace = "\u03b7"; break; | |
| 765 case '\u1f75': replace = "\u03b7"; break; | |
| 766 case '\u1f76': replace = "\u03b9"; break; | |
| 767 case '\u1f77': replace = "\u03b9"; break; | |
| 768 case '\u1f78': replace = "\u03bf"; break; | |
| 769 case '\u1f79': replace = "\u03bf"; break; | |
| 770 case '\u1f7a': replace = "\u03c5"; break; | |
| 771 case '\u1f7b': replace = "\u03c5"; break; | |
| 772 case '\u1f7c': replace = "\u03c9"; break; | |
| 773 case '\u1f7d': replace = "\u03c9"; break; | |
| 774 case '\u1f80': replace = "\u03b1"; break; | |
| 775 case '\u1f81': replace = "\u03b1"; break; | |
| 776 case '\u1f82': replace = "\u03b1"; break; | |
| 777 case '\u1f83': replace = "\u03b1"; break; | |
| 778 case '\u1f84': replace = "\u03b1"; break; | |
| 779 case '\u1f85': replace = "\u03b1"; break; | |
| 780 case '\u1f86': replace = "\u03b1"; break; | |
| 781 case '\u1f87': replace = "\u03b1"; break; | |
| 782 case '\u1f88': replace = "\u0391"; break; | |
| 783 case '\u1f89': replace = "\u0391"; break; | |
| 784 case '\u1f8a': replace = "\u0391"; break; | |
| 785 case '\u1f8b': replace = "\u0391"; break; | |
| 786 case '\u1f8c': replace = "\u0391"; break; | |
| 787 case '\u1f8d': replace = "\u0391"; break; | |
| 788 case '\u1f8e': replace = "\u0391"; break; | |
| 789 case '\u1f8f': replace = "\u0391"; break; | |
| 790 case '\u1f90': replace = "\u03b7"; break; | |
| 791 case '\u1f91': replace = "\u03b7"; break; | |
| 792 case '\u1f92': replace = "\u03b7"; break; | |
| 793 case '\u1f93': replace = "\u03b7"; break; | |
| 794 case '\u1f94': replace = "\u03b7"; break; | |
| 795 case '\u1f95': replace = "\u03b7"; break; | |
| 796 case '\u1f96': replace = "\u03b7"; break; | |
| 797 case '\u1f97': replace = "\u03b7"; break; | |
| 798 case '\u1f98': replace = "\u0397"; break; | |
| 799 case '\u1f99': replace = "\u0397"; break; | |
| 800 case '\u1f9a': replace = "\u0397"; break; | |
| 801 case '\u1f9b': replace = "\u0397"; break; | |
| 802 case '\u1f9c': replace = "\u0397"; break; | |
| 803 case '\u1f9d': replace = "\u0397"; break; | |
| 804 case '\u1f9e': replace = "\u0397"; break; | |
| 805 case '\u1f9f': replace = "\u0397"; break; | |
| 806 case '\u1fa0': replace = "\u03c9"; break; | |
| 807 case '\u1fa1': replace = "\u03c9"; break; | |
| 808 case '\u1fa2': replace = "\u03c9"; break; | |
| 809 case '\u1fa3': replace = "\u03c9"; break; | |
| 810 case '\u1fa4': replace = "\u03c9"; break; | |
| 811 case '\u1fa5': replace = "\u03c9"; break; | |
| 812 case '\u1fa6': replace = "\u03c9"; break; | |
| 813 case '\u1fa7': replace = "\u03c9"; break; | |
| 814 case '\u1fa8': replace = "\u03a9"; break; | |
| 815 case '\u1fa9': replace = "\u03a9"; break; | |
| 816 case '\u1faa': replace = "\u03a9"; break; | |
| 817 case '\u1fab': replace = "\u03a9"; break; | |
| 818 case '\u1fac': replace = "\u03a9"; break; | |
| 819 case '\u1fad': replace = "\u03a9"; break; | |
| 820 case '\u1fae': replace = "\u03a9"; break; | |
| 821 case '\u1faf': replace = "\u03a9"; break; | |
| 822 case '\u1fb2': replace = "\u03b1"; break; | |
| 823 case '\u1fb3': replace = "\u03b1"; break; | |
| 824 case '\u1fb4': replace = "\u03b1"; break; | |
| 825 case '\u1fb6': replace = "\u03b1"; break; | |
| 826 case '\u1fb7': replace = "\u03b1"; break; | |
| 827 case '\u1fba': replace = "\u0391"; break; | |
| 828 case '\u1fbb': replace = "\u0391"; break; | |
| 829 case '\u1fbc': replace = "\u0391"; break; | |
| 830 case '\u1fc2': replace = "\u03b7"; break; | |
| 831 case '\u1fc3': replace = "\u03b7"; break; | |
| 832 case '\u1fc4': replace = "\u03b7"; break; | |
| 833 case '\u1fc6': replace = "\u03b7"; break; | |
| 834 case '\u1fc7': replace = "\u03b7"; break; | |
| 835 case '\u1fca': replace = "\u0397"; break; | |
| 836 case '\u1fcb': replace = "\u0397"; break; | |
| 837 case '\u1fcc': replace = "\u0397"; break; | |
| 838 case '\u1fd2': replace = "\u03b9"; break; | |
| 839 case '\u1fd3': replace = "\u03b9"; break; | |
| 840 case '\u1fd6': replace = "\u03b9"; break; | |
| 841 case '\u1fd7': replace = "\u03b9"; break; | |
| 842 case '\u1fda': replace = "\u0399"; break; | |
| 843 case '\u1fdb': replace = "\u039f"; break; | |
| 844 case '\u1fe2': replace = "\u03c5"; break; | |
| 845 case '\u1fe3': replace = "\u03c5"; break; | |
| 846 case '\u1fe4': replace = "\u03c1"; break; | |
| 847 case '\u1fe5': replace = "\u03c1"; break; | |
| 848 case '\u1fe6': replace = "\u03c5"; break; | |
| 849 case '\u1fe7': replace = "\u03c5"; break; | |
| 850 case '\u1fea': replace = "\u03a5"; break; | |
| 851 case '\u1feb': replace = "\u03a5"; break; | |
| 852 case '\u1fec': replace = "\u03a1"; break; | |
| 853 case '\u1ff2': replace = "\u03c9"; break; | |
| 854 case '\u1ff3': replace = "\u03c9"; break; | |
| 855 case '\u1ff4': replace = "\u03c9"; break; | |
| 856 case '\u1ff6': replace = "\u03c9"; break; | |
| 857 case '\u1ff7': replace = "\u03c9"; break; | |
| 858 case '\u1ff8': replace = "\u039f"; break; | |
| 859 case '\u1ff9': replace = "\u039f"; break; | |
| 860 case '\u1ffa': replace = "\u03a9"; break; | |
| 861 case '\u1ffb': replace = "\u03a9"; break; | |
| 862 case '\u1ffc': replace = "\u03a9"; break; | |
| 863 | |
| 864 case '<': break; | |
| 865 case '>': break; | |
| 866 case '-': break; // same treatment as soft hyphen | |
| 867 case '\u00ad': break; // soft hyphen | |
| 868 default: replace += c; break; | |
| 869 } | |
| 870 buf.append(replace); | |
| 871 // update offsets if replacement is a different length | |
| 872 if (offsets != null) { | |
| 873 int r = replace.length(); | |
| 874 if (r == 0) | |
| 875 this.offsets = arrayKill(this.offsets, i - n); | |
| 876 else if (r == 2) | |
| 877 this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); | |
| 878 n += 1 - r; | |
| 879 } | |
| 880 } | |
| 881 return buf.toString(); | |
| 882 } else { // unknown or no language | |
| 883 return s; | |
| 884 } | |
| 885 } | |
| 886 | |
| 887 public String deNormalizeToRegExpr(String s) { | |
| 888 // TODO all characters in all languages | |
| 889 if (language.equals("la") || language.equals("lat")) { | |
| 890 StringBuffer buf = new StringBuffer(); | |
| 891 if (s.indexOf("ae") != -1) { | |
| 892 String str1 = s; | |
| 893 str1 = str1.replaceAll("ae", "\u0119"); | |
| 894 String str2 = s; | |
| 895 str2 = str2.replaceAll("ae", "\u00c6"); | |
| 896 String str3 = s; | |
| 897 str3 = str3.replaceAll("ae", "\u00e6"); | |
| 898 buf.append(str1 + "|" + str2 + "|" + str3 + "|"); | |
| 899 } | |
| 900 if (s.indexOf("oe") != -1) { | |
| 901 String str1 = s; | |
| 902 str1 = str1.replaceAll("oe", "\u0152"); | |
| 903 String str2 = s; | |
| 904 str2 = str2.replaceAll("oe", "\u0153"); | |
| 905 buf.append(str1 + "|" + str2 + "|"); | |
| 906 } | |
| 907 if (s.indexOf("ss") != -1) { | |
| 908 String str1 = s; | |
| 909 str1 = str1.replaceAll("ss", "\u00df"); | |
| 910 buf.append(str1 + "|"); | |
| 911 } | |
| 912 boolean beginWord = true; | |
| 913 for (int i = 0; i < s.length(); i++) { | |
| 914 char c = s.charAt(i); | |
| 915 if (! beginWord) | |
| 916 c = Character.toLowerCase(c); | |
| 917 beginWord = Character.isWhitespace(c); | |
| 918 String replace = new String(); | |
| 919 switch (c) { | |
| 920 case 'a': replace = "[a\u00c0\u00c1\u00c2\u00c4\u00e0\u00e1\u00e2\u00e4]"; break; | |
| 921 case 'c': replace = "[c\u00c7\u00e7]"; break; | |
| 922 case 'e': replace = "[e\u00c8\u00c9\u00ca\u00cb\u00e8\u00e9\u00ea\u00eb\u0113\u0115\u1ebd]"; break; | |
| 923 case 'i': replace = "[ij\u00cc\u00cd\u00ce\u00cf\u00ec\u00ed\u00ee\u00ef\u012a\u012b\u012c\u012d]"; break; | |
| 924 case 'o': replace = "[o\u00d2\u00d3\u00d4\u00d6\u00f2\u00f3\u00f4\u00f6\u014c\u014d\u014e\u014f]"; break; | |
| 925 case 'u': replace = "[uv\u00d9\u00da\u00db\u00dc\u00f9\u00fa\u00fb\u00fc\u016a\u016b\u016c\u016d]"; break; | |
| 926 case 's': replace = "[s\u017f]"; break; | |
| 927 default: replace += c; break; | |
| 928 } | |
| 929 buf.append(replace); | |
| 930 } | |
| 931 return buf.toString(); | |
| 932 } else if (language.equals("en")) { | |
| 933 StringBuffer buf = new StringBuffer(); | |
| 934 if (s.indexOf("ae") != -1) { | |
| 935 String str1 = s; | |
| 936 str1 = str1.replaceAll("ae", "\u0119"); | |
| 937 String str2 = s; | |
| 938 str2 = str2.replaceAll("ae", "\u00c6"); | |
| 939 String str3 = s; | |
| 940 str3 = str3.replaceAll("ae", "\u00e6"); | |
| 941 buf.append(str1 + "|" + str2 + "|" + str3 + "|"); | |
| 942 } | |
| 943 if (s.indexOf("oe") != -1) { | |
| 944 String str1 = s; | |
| 945 str1 = str1.replaceAll("oe", "\u0152"); | |
| 946 String str2 = s; | |
| 947 str2 = str2.replaceAll("oe", "\u0153"); | |
| 948 buf.append(str1 + "|" + str2 + "|"); | |
| 949 } | |
| 950 if (s.indexOf("ss") != -1) { | |
| 951 String str1 = s; | |
| 952 str1 = str1.replaceAll("ss", "\u00df"); | |
| 953 buf.append(str1 + "|"); | |
| 954 } | |
| 955 boolean beginWord = true; | |
| 956 for (int i = 0; i < s.length(); i++) { | |
| 957 char c = s.charAt(i); | |
| 958 if (! beginWord) | |
| 959 c = Character.toLowerCase(c); | |
| 960 beginWord = Character.isWhitespace(c); | |
| 961 String replace = new String(); | |
| 962 switch (c) { | |
| 963 case 'a': replace = "[a\u00c0\u00c1\u00c2\u00c4\u00e0\u00e1\u00e2\u00e4]"; break; | |
| 964 case 'c': replace = "[c\u00c7\u00e7]"; break; | |
| 965 case 'e': replace = "[e\u00c8\u00c9\u00ca\u00cb\u00e8\u00e9\u00ea\u00eb\u0113\u0115\u1e8d]"; break; | |
| 966 case 'i': replace = "[i\u00cc\u00cd\u00ce\u00cf\u00ec\u00ed\u00ee\u00ef\u012a\u012b\u012c\u012d]"; break; | |
| 967 case 'o': replace = "[o\u00d2\u00d3\u00d4\u00d6\u00f2\u00f3\u00f4\u00f6\u014c\u014d\u014e\u014f]‚"; break; | |
| 968 case 'u': replace = "[u\u00d9\u00da\u00db\u00dc\u00f9\u00fa\u00fb\u00fc\u016a\u016b\u016c\u016d]"; break; | |
| 969 case 's': replace = "[s\u017f]"; break; | |
| 970 default: replace += c; break; | |
| 971 } | |
| 972 buf.append(replace); | |
| 973 } | |
| 974 return buf.toString(); | |
| 975 } else if (language.equals("de")) { | |
| 976 StringBuffer buf = new StringBuffer(); | |
| 977 if (s.indexOf("ss") != -1) { | |
| 978 String str1 = s; | |
| 979 str1 = str1.replaceAll("ss", "\u00df"); | |
| 980 buf.append(str1 + "|"); | |
| 981 } | |
| 982 if (s.indexOf("ae") != -1) { | |
| 983 String str1 = s; | |
| 984 str1 = str1.replaceAll("ae", "\u00e4"); | |
| 985 buf.append(str1 + "|"); | |
| 986 } | |
| 987 if (s.indexOf("oe") != -1) { | |
| 988 String str1 = s; | |
| 989 str1 = str1.replaceAll("oe", "\u00f6"); | |
| 990 buf.append(str1 + "|"); | |
| 991 } | |
| 992 if (s.indexOf("ue") != -1) { | |
| 993 String str1 = s; | |
| 994 str1 = str1.replaceAll("ue", "\u00fc"); | |
| 995 buf.append(str1 + "|"); | |
| 996 } | |
| 997 boolean beginWord = true; | |
| 998 for (int i = 0; i < s.length(); i++) { | |
| 999 char c = s.charAt(i); | |
| 1000 if (! beginWord) | |
| 1001 c = Character.toLowerCase(c); | |
| 1002 beginWord = Character.isWhitespace(c); | |
| 1003 String replace = new String(); | |
| 1004 switch (c) { | |
| 1005 case 'e': replace = "[e\u00e9]"; break; | |
| 1006 default: replace += c; break; | |
| 1007 } | |
| 1008 buf.append(replace); | |
| 1009 } | |
| 1010 return buf.toString(); | |
| 1011 } else { // unknown or no language | |
| 1012 return s; | |
| 1013 } | |
| 1014 } | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1015 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1016 private String normalize4HumanReaders(String s) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1017 String normStr = s; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1018 StringReader strReader = new StringReader(normStr + "\n"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1019 MpdlNormalizerLexAll mpdlNormalizerLexAll = new MpdlNormalizerLexAll(strReader); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1020 if (Language.getInstance().isLatin(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1021 mpdlNormalizerLexAll.yybegin(MpdlNormalizerLexAll.LA); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1022 } else if (Language.getInstance().isChinese(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1023 mpdlNormalizerLexAll.yybegin(MpdlNormalizerLexAll.ZH); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1024 } else { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1025 // TODO normalization for all languages |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1026 return normalize4Lexica(s, null); // old function |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1027 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1028 String retStr = ""; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1029 String token = ""; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1030 while (token != null) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1031 try { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1032 token = mpdlNormalizerLexAll.yylex(); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1033 if (token != null) |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1034 retStr += token; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1035 } catch (IOException e ) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1036 // nothing cause IOException is not needed for a StringReader |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1037 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1038 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1039 normStr = retStr; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1040 return normStr; |
| 0 | 1041 } |
| 1042 | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1043 /* |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1044 // explicit words |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1045 normStr = normStr.replaceAll("aliàs", "alias"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1046 normStr = normStr.replaceAll("hîc", "hic"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1047 normStr = normStr.replaceAll("quòd", "quod"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1048 normStr = normStr.replaceAll("Quòd", "Quod"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1049 normStr = normStr.replaceAll("QVòd", "Quod"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1050 normStr = normStr.replaceAll("Cùmque", "Cumque"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1051 normStr = normStr.replaceAll("aër", "aer"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1052 // ij |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1053 normStr = normStr.replaceAll("ij", "ii"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1054 // qu/qv |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1055 normStr = normStr.replaceAll("qv", "qu"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1056 // normStr = normStr.replaceAll("qV", "qU"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1057 normStr = normStr.replaceAll("Qv", "Qu"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1058 normStr = normStr.replaceAll("QV", "QU"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1059 // u/v |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1060 String vowels = getVowels(); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1061 String consonants = getConsonants(); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1062 normStr = normStr.replaceAll("([" + vowels + "])([-]*)u([" + vowels +"])", "$1$2v$3"); // vowel + u + vowel --> vowel + v + vowel |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1063 normStr = normStr.replaceAll("([" + vowels + "])([-]*)U([" + vowels +"])", "$1$2V$3"); // vowel + U + vowel --> vowel + V + vowel |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1064 normStr = normStr.replaceAll("([" + consonants + "])([-]*)v([" + consonants +"])", "$1$2u$3"); // consonant + v + consonant --> consonant + u + consonant |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1065 normStr = normStr.replaceAll("([" + consonants + "])([-]*)V([" + consonants +"])", "$1$2U$3"); // consonant + V + consonant --> consonant + U + consonant |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1066 normStr = normStr.replaceAll("^v([" + consonants +"])", "u$1"); // v + consonant --> u + consonant |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1067 normStr = normStr.replaceAll("^V([" + consonants +"])", "U$1"); // V + consonant --> U + consonant |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1068 // end of word: diacritica |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1069 normStr = normStr.replaceAll("à$", "a"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1070 normStr = normStr.replaceAll("è$", "e"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1071 normStr = normStr.replaceAll("ò$", "o"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1072 normStr = normStr.replaceAll("àm$", "am"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1073 normStr = normStr.replaceAll("ùm$", "um"); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1074 String normStrTmp = normStr; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1075 normStr = ""; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1076 for (int i = 0; i < normStrTmp.length(); i++) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1077 char c = normStrTmp.charAt(i); |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1078 String replace = ""; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1079 switch (c) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1080 case 'ſ': replace = "s"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1081 case 'ß': replace = "ss"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1082 case 'æ': replace = "ae"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1083 case 'Æ': replace = "AE"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1084 case 'ę': replace = "ae"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1085 case 'œ': replace = "oe"; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1086 default: replace += c; break; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1087 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1088 normStr = normStr + replace; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1089 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1090 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1091 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1092 private String getVowels() { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1093 String retStr = null; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1094 if (Language.getInstance().isItalian(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1095 retStr = "AEIOUaeiou" + |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1096 "\u00c6\u00e6" + // AE ligatures |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1097 "\u0152\u0153"; // OE ligatures |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1098 } else if (Language.getInstance().isLatin(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1099 retStr = "AEIOUaeiouÆœęàèòù"; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1100 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1101 // TODO all languages |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1102 return retStr; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1103 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1104 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1105 private String getConsonants() { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1106 String retStr = null; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1107 if (Language.getInstance().isItalian(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1108 retStr = "BCDFGHKLMNPQRSTVWXZ" + |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1109 "bcdfghklmnpqrstvwxz" + |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1110 "ſß"; // long/sharp S |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1111 } else if (Language.getInstance().isLatin(language)) { |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1112 retStr = "BCDFGHKLMNPQRSTVWXZ" + |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1113 "bcdfghklmnpqrstvwxz" + |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1114 "ſß"; // long/sharp S |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1115 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1116 // TODO all languages |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1117 return retStr; |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1118 } |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1119 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1120 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1121 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1122 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1123 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1124 * |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1125 * |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1126 * |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1127 * |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1128 */ |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1129 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1130 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1131 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1132 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1133 |
|
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1134 |
| 0 | 1135 /** |
| 1136 * Returns a copy of an integer array with the element at | |
| 1137 * <code>index</code> removed ("killed"). | |
| 1138 * | |
| 1139 * @param array integer array | |
| 1140 * @param index index of element to remove | |
| 1141 */ | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1142 private int[] arrayKill(int[] array, int index) { |
| 0 | 1143 int[] newArray = new int[array.length - 1]; |
| 1144 System.arraycopy(array, 0, newArray, 0, index); | |
| 1145 System.arraycopy(array, index + 1, newArray, index, array.length - index - 1); | |
| 1146 return newArray; | |
| 1147 } | |
| 1148 | |
| 1149 /** | |
| 1150 * Returns a copy of an integer array with <code>count</code> elements | |
| 1151 * inserted at <code>index</code>. | |
| 1152 * | |
| 1153 * @param array integer array | |
| 1154 * @param index index to insert new elements | |
| 1155 * @param value value to insert into new slots | |
| 1156 * @param count number of new slots to insert | |
| 1157 */ | |
|
6
2396a569e446
new functions: externalObjects, normalizer, Unicode2Betacode
Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de>
parents:
0
diff
changeset
|
1158 private int[] arrayInsert(int[] array, int index, int value, int count) { |
| 0 | 1159 int[] newArray = new int[array.length + count]; |
| 1160 System.arraycopy(array, 0, newArray, 0, index); | |
| 1161 for (int i = 0; i < count; i++) newArray[index + i] = value; | |
| 1162 System.arraycopy(array, index, newArray, index + count, array.length - index); | |
| 1163 return newArray; | |
| 1164 } | |
| 1165 | |
| 1166 } |