Mercurial > hg > mpdl-group
view software/eXist/mpdl-modules/src/de/mpg/mpiwg/berlin/mpdl/lt/analyzer/MpdlNormalizer.java @ 14:5df60f24e997
diverse Fehlerbehebungen
| author | Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de> |
|---|---|
| date | Mon, 29 Aug 2011 17:40:02 +0200 |
| parents | 1ec29fdd0db8 |
| children | 257f67be5c00 |
line wrap: on
line source
package de.mpg.mpiwg.berlin.mpdl.lt.analyzer; import java.io.IOException; import java.io.StringReader; import java.util.ArrayList; import de.mpg.mpiwg.berlin.mpdl.exception.ApplicationException; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexAR; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexDE; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexEL; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexEN; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexFR; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexIT; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexLA; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexNL; import de.mpg.mpiwg.berlin.mpdl.lt.analyzer.lang.MpdlNormalizerLexZH; import de.mpg.mpiwg.berlin.mpdl.lt.doc.regularization.Regularization; import de.mpg.mpiwg.berlin.mpdl.lt.doc.regularization.RegularizationManager; import de.mpg.mpiwg.berlin.mpdl.lt.general.Language; public class MpdlNormalizer { public static int DISPLAY = 1; // normalization in DISPLAY mode public static int DICTIONARY = 2; // normalization in DICTIONARY mode public static int SEARCH = 3; // normalization in SEARCH mode; never used so far in indexing because it does not support the morph. lexicons such as CELEX (e.g. eingeschränkt would not be stemmed to eingeschraenkt) private int normMode = DICTIONARY; // Default e.g. for indexing and querying private String[] normFunctionsToUse = {"reg", "norm"}; // default is to use all of these normalization functions private String language; private int[] offsets; public MpdlNormalizer(String[] normFunctionsToUse, String lang) { this.normFunctionsToUse = normFunctionsToUse; String language = Language.getInstance().getLanguageId(lang); this.language = language; } public MpdlNormalizer(String language) { this.language = language; } public String getLanguage() { return language; } public void setNormMode(int normMode) { this.normMode = normMode; } /** * Applies the normalization rules in <code>language</code> to * <code>s</code>, without offset tracking. * * @param s source string * @return normalized string */ public String normalize(String s) throws ApplicationException { String normStr = s; if (useRegFunction()) { // try to regularize the string to the norm form over predefined regularizations RegularizationManager regManager = RegularizationManager.getInstance(); ArrayList<Regularization> regs = regManager.findRegsByOrig(language, s); if (regs != null && regs.size() > 0) { Regularization reg = regs.get(0); // only one: the first one String regNormStr = reg.getNorm(); normStr = regNormStr; } } if (useNormFunction()) { // normalize the string by string replacements if (normMode == DICTIONARY) { normStr = normalize(normStr, DICTIONARY); } else if (normMode == DISPLAY) normStr = normalize(normStr, DISPLAY); else if (normMode == SEARCH) normStr = normalize(normStr, SEARCH); } return normStr; } private boolean useRegFunction() { boolean useReg = false; for (int i=0; i< normFunctionsToUse.length; i++) { String function = normFunctionsToUse[i]; if (function.equals("reg")) return true; } return useReg; } private boolean useNormFunction() { boolean useNorm = false; for (int i=0; i< normFunctionsToUse.length; i++) { String function = normFunctionsToUse[i]; if (function.equals("norm")) return true; } return useNorm; } public String deNormalizeToRegExpr(String s) { // TODO all characters in all languages if (language.equals("la") || language.equals("lat")) { StringBuffer buf = new StringBuffer(); if (s.indexOf("ae") != -1) { String str1 = s; str1 = str1.replaceAll("ae", "\u0119"); String str2 = s; str2 = str2.replaceAll("ae", "\u00c6"); String str3 = s; str3 = str3.replaceAll("ae", "\u00e6"); buf.append(str1 + "|" + str2 + "|" + str3 + "|"); } if (s.indexOf("oe") != -1) { String str1 = s; str1 = str1.replaceAll("oe", "\u0152"); String str2 = s; str2 = str2.replaceAll("oe", "\u0153"); buf.append(str1 + "|" + str2 + "|"); } if (s.indexOf("ss") != -1) { String str1 = s; str1 = str1.replaceAll("ss", "\u00df"); buf.append(str1 + "|"); } boolean beginWord = true; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); if (! beginWord) c = Character.toLowerCase(c); beginWord = Character.isWhitespace(c); String replace = new String(); switch (c) { case 'a': replace = "[a\u00c0\u00c1\u00c2\u00c4\u00e0\u00e1\u00e2\u00e4]"; break; case 'c': replace = "[c\u00c7\u00e7]"; break; case 'e': replace = "[e\u00c8\u00c9\u00ca\u00cb\u00e8\u00e9\u00ea\u00eb\u0113\u0115\u1ebd]"; break; case 'i': replace = "[ij\u00cc\u00cd\u00ce\u00cf\u00ec\u00ed\u00ee\u00ef\u012a\u012b\u012c\u012d]"; break; case 'o': replace = "[o\u00d2\u00d3\u00d4\u00d6\u00f2\u00f3\u00f4\u00f6\u014c\u014d\u014e\u014f]"; break; case 'u': replace = "[uv\u00d9\u00da\u00db\u00dc\u00f9\u00fa\u00fb\u00fc\u016a\u016b\u016c\u016d]"; break; case 's': replace = "[s\u017f]"; break; default: replace += c; break; } buf.append(replace); } return buf.toString(); } else if (language.equals("en")) { StringBuffer buf = new StringBuffer(); if (s.indexOf("ae") != -1) { String str1 = s; str1 = str1.replaceAll("ae", "\u0119"); String str2 = s; str2 = str2.replaceAll("ae", "\u00c6"); String str3 = s; str3 = str3.replaceAll("ae", "\u00e6"); buf.append(str1 + "|" + str2 + "|" + str3 + "|"); } if (s.indexOf("oe") != -1) { String str1 = s; str1 = str1.replaceAll("oe", "\u0152"); String str2 = s; str2 = str2.replaceAll("oe", "\u0153"); buf.append(str1 + "|" + str2 + "|"); } if (s.indexOf("ss") != -1) { String str1 = s; str1 = str1.replaceAll("ss", "\u00df"); buf.append(str1 + "|"); } boolean beginWord = true; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); if (! beginWord) c = Character.toLowerCase(c); beginWord = Character.isWhitespace(c); String replace = new String(); switch (c) { case 'a': replace = "[a\u00c0\u00c1\u00c2\u00c4\u00e0\u00e1\u00e2\u00e4]"; break; case 'c': replace = "[c\u00c7\u00e7]"; break; case 'e': replace = "[e\u00c8\u00c9\u00ca\u00cb\u00e8\u00e9\u00ea\u00eb\u0113\u0115\u1e8d]"; break; case 'i': replace = "[i\u00cc\u00cd\u00ce\u00cf\u00ec\u00ed\u00ee\u00ef\u012a\u012b\u012c\u012d]"; break; case 'o': replace = "[o\u00d2\u00d3\u00d4\u00d6\u00f2\u00f3\u00f4\u00f6\u014c\u014d\u014e\u014f]‚"; break; case 'u': replace = "[u\u00d9\u00da\u00db\u00dc\u00f9\u00fa\u00fb\u00fc\u016a\u016b\u016c\u016d]"; break; case 's': replace = "[s\u017f]"; break; default: replace += c; break; } buf.append(replace); } return buf.toString(); } else if (language.equals("de")) { StringBuffer buf = new StringBuffer(); if (s.indexOf("ss") != -1) { String str1 = s; str1 = str1.replaceAll("ss", "\u00df"); buf.append(str1 + "|"); } if (s.indexOf("ae") != -1) { String str1 = s; str1 = str1.replaceAll("ae", "\u00e4"); buf.append(str1 + "|"); } if (s.indexOf("oe") != -1) { String str1 = s; str1 = str1.replaceAll("oe", "\u00f6"); buf.append(str1 + "|"); } if (s.indexOf("ue") != -1) { String str1 = s; str1 = str1.replaceAll("ue", "\u00fc"); buf.append(str1 + "|"); } boolean beginWord = true; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); if (! beginWord) c = Character.toLowerCase(c); beginWord = Character.isWhitespace(c); String replace = new String(); switch (c) { case 'e': replace = "[e\u00e9]"; break; default: replace += c; break; } buf.append(replace); } return buf.toString(); } else { // unknown or no language return s; } } private String normalize(String s, int mode) { String inputStr = s; StringReader strReader = new StringReader(inputStr + "\n"); String retStr = ""; String token = ""; try { if (Language.getInstance().isLatin(language)) { MpdlNormalizerLexLA mpdlNormalizerLex = new MpdlNormalizerLexLA(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexLA.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexLA.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexLA.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isArabic(language)) { MpdlNormalizerLexAR mpdlNormalizerLex = new MpdlNormalizerLexAR(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexAR.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexAR.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexAR.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isGerman(language)) { MpdlNormalizerLexDE mpdlNormalizerLex = new MpdlNormalizerLexDE(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexDE.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexDE.CELEX); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexDE.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isGreek(language)) { MpdlNormalizerLexEL mpdlNormalizerLex = new MpdlNormalizerLexEL(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEL.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEL.SIGMA); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEL.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isEnglish(language)) { MpdlNormalizerLexEN mpdlNormalizerLex = new MpdlNormalizerLexEN(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEN.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEN.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexEN.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isFrench(language)) { MpdlNormalizerLexFR mpdlNormalizerLex = new MpdlNormalizerLexFR(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexFR.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexFR.CELEX); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexFR.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isItalian(language)) { MpdlNormalizerLexIT mpdlNormalizerLex = new MpdlNormalizerLexIT(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexIT.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexIT.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexIT.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isDutch(language)) { MpdlNormalizerLexNL mpdlNormalizerLex = new MpdlNormalizerLexNL(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexNL.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexNL.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexNL.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else if (Language.getInstance().isChinese(language)) { MpdlNormalizerLexZH mpdlNormalizerLex = new MpdlNormalizerLexZH(strReader); if (mode == DISPLAY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexZH.DISP); else if (mode == DICTIONARY) mpdlNormalizerLex.yybegin(MpdlNormalizerLexZH.DICT); else if (mode == SEARCH) mpdlNormalizerLex.yybegin(MpdlNormalizerLexZH.SEARCH); while (token != null) { token = mpdlNormalizerLex.yylex(); if (token != null) retStr += token; } } else { retStr = s; // return the string unchanged } } catch (IOException e ) { // nothing cause IOException is not needed for a StringReader } return retStr; } /** * Old code from Arboreal (Malcolm Hyman) * Applies the normalization rules in <code>language</code> to * <code>s</code>, with offset tracking.<p> * * <strong>WARNING:</strong> * Arboreal will not work properly if a normalization substitution * replaces a source character with more than two target characters! * This is simply a BUG, and should be fixed. Fortunately, however, * one does not often need such a replacement.<p> * * @param s source string * @param offsets character offset table * @return normalized string */ private String normalize4Lexica(String s, int[] offsets) { this.offsets = offsets; if (language.equals("la") || language.equals("lat")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case 'j': replace = "i"; break; case 'v': replace = "u"; break; /* * Linguistic note: /u/ and /v/ are rarely phonemic * in Latin, as in alui 's/he nourished' vs. * alvi 'of a belly', volui 's/he wished' or 'it rolled' * vs. volvi 'to be rolled', (in)seruit 's/he joined * together' vs. (in)servit 's/he serves'. */ case 'q': if ((i < s.length() - 1) && (s.charAt(i + 1) == ';')) replace = "qu"; else replace = "q"; break; case ';': if ((i > 0) && (s.charAt(i - 1) == 'q')) replace = "e"; else if ((i == 0) || ! Character.isLetter(s.charAt(i - 1))) replace = ";"; else replace = ""; break; case '\u0300': replace = ""; break; // COMBINING GRAVE ACCENT case '\u0301': replace = ""; break; // COMBINING ACCUTE ACCENT case '\u0302': replace = ""; break; // COMBINING CIRCUMFLEX ACCENT case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S case '\u00ad': break; // soft hyphen // new in MPDL project by J. Willenborg case '\u1e14': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e15': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e16': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e17': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e18': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e19': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e1a': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e1b': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e1c': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1e1d': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1eb8': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1eb9': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1eba': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ebb': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ebc': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ebd': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ebe': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ebf': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec0': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec1': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec2': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec3': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec4': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec5': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec6': replace = "e"; break; // LATIN ... LETTER E WITH ... case '\u1ec7': replace = "e"; break; // LATIN ... LETTER E WITH ... // by Malcolm case '\u2329': break; // BRA case '\u232a': break; // KET default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("it")) { // 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 StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S // new in MPDL project by J. Willenborg case '\u1e8d': replace = "e"; break; // LATIN SMALL LETTER E WITH TILDE default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); // new Mpdl code: added by J. Willenborg: most of the latin replacements also in english } else if (language.equals("en")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u0300': replace = ""; break; // COMBINING GRAVE ACCENT case '\u0301': replace = ""; break; // COMBINING ACCUTE ACCENT case '\u0302': replace = ""; break; // COMBINING CIRCUMFLEX ACCENT case '\u00c0': replace = "A"; break; // LATIN CAPITAL LETTER A GRAVE case '\u00c1': replace = "A"; break; // LATIN CAPITAL LETTER A ACUTE case '\u00c2': replace = "A"; break; // LATIN CAPITAL LETTER A CIRCUMFLEX case '\u00c4': replace = "A"; break; // LATIN CAPITAL LETTER A DIAERESIS case '\u00c6': replace = "Ae"; break; // LATIN CAPITAL LETTER A E case '\u00c7': replace = "C"; break; // LATIN CAPITAL LETTER C CEDILLA case '\u00c8': replace = "E"; break; // LATIN CAPITAL LETTER E GRAVE case '\u00c9': replace = "E"; break; // LATIN CAPITAL LETTER E ACUTE case '\u00ca': replace = "E"; break; // LATIN CAPITAL LETTER E CIRCUMFLEX case '\u00cb': replace = "E"; break; // LATIN CAPITAL LETTER E DIAERESIS case '\u00cc': replace = "I"; break; // LATIN CAPITAL LETTER I GRAVE; case '\u00cd': replace = "I"; break; // LATIN CAPITAL LETTER I ACUTE case '\u00ce': replace = "I"; break; // LATIN CAPITAL LETTER I CIRCUMFLEX case '\u00cf': replace = "I"; break; // LATIN CAPITAL LETTER I DIAERESIS case '\u00d2': replace = "O"; break; // LATIN CAPITAL LETTER O GRAVE case '\u00d3': replace = "O"; break; // LATIN CAPITAL LETTER O ACUTE case '\u00d4': replace = "O"; break; // LATIN CAPITAL LETTER O CIRCUMFLEX case '\u00d6': replace = "O"; break; // LATIN CAPITAL LETTER O DIAERESIS case '\u00d9': replace = "U"; break; // LATIN CAPITAL LETTER U GRAVE case '\u00da': replace = "U"; break; // LATIN CAPITAL LETTER U ACUTE case '\u00db': replace = "U"; break; // LATIN CAPITAL LETTER U CIRCUMFLEX case '\u00dc': replace = "U"; break; // LATIN CAPITAL LETTER U DIAERESIS case '\u00e0': replace = "a"; break; // LATIN SMALL LETTER A GRAVE case '\u00e1': replace = "a"; break; // LATIN SMALL LETTER A ACUTE case '\u00e2': replace = "a"; break; // LATIN SMALL LETTER A CIRCUMFLEX case '\u00e4': replace = "a"; break; // LATIN SMALL LETTER A DIAERESIS case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E case '\u00e7': replace = "c"; break; // LATIN SMALL LETTER C CEDILLA case '\u00e8': replace = "e"; break; // LATIN SMALL LETTER E GRAVE case '\u00e9': replace = "e"; break; // LATIN SMALL LETTER E ACUTE case '\u00ea': replace = "e"; break; // LATIN SMALL LETTER E CIRCUMFLEX case '\u00eb': replace = "e"; break; // LATIN SMALL LETTER E DIAERESIS case '\u00ec': replace = "i"; break; // LATIN SMALL LETTER I GRAVE case '\u00ed': replace = "i"; break; // LATIN SMALL LETTER I ACUTE case '\u00ee': replace = "i"; break; // LATIN SMALL LETTER I CIRCUMFLEX case '\u00ef': replace = "i"; break; // LATIN SMALL LETTER I DIAERESIS case '\u00f2': replace = "o"; break; // LATIN SMALL LETTER O GRAVE case '\u00f3': replace = "o"; break; // LATIN SMALL LETTER O ACUTE case '\u00f4': replace = "o"; break; // LATIN SMALL LETTER O CIRCUMFLEX case '\u00f6': replace = "o"; break; // LATIN SMALL LETTER O DIAERESIS case '\u00f9': replace = "u"; break; // LATIN SMALL LETTER U GRAVE case '\u00fa': replace = "u"; break; // LATIN SMALL LETTER U ACUTE case '\u00fb': replace = "u"; break; // LATIN SMALL LETTER U CIRCUMFLEX case '\u00fc': replace = "u"; break; // LATIN SMALL LETTER U DIAERESIS case '\u0100': replace = "A"; break; // LATIN CAPITAL LETTER A MACRON case '\u0101': replace = "a"; break; // LATIN SMALL LETTER A MACRON case '\u0102': replace = "A"; break; // LATIN CAPITAL LETTER A BREVE case '\u0103': replace = "a"; break; // LATIN SMALL LETTER A BREVE case '\u0112': replace = "E"; break; // LATIN CAPITAL LETTER E MACRON case '\u0113': replace = "e"; break; // LATIN SMALL LETTER E MACRON case '\u0114': replace = "E"; break; // LATIN CAPITAL LETTER E BREVE case '\u0115': replace = "e"; break; // LATIN SMALL LETTER E BREVE case '\u0118': replace = "Ae"; break; // LATIN CAPITAL LETTER E OGONEK case '\u0119': replace = "ae"; break; // LATIN SMALL LETTER E OGONEK case '\u012a': replace = "I"; break; // LATIN CAPITAL LETTER I MACRON case '\u012b': replace = "i"; break; // LATIN SMALL LETTER I MACRON case '\u012c': replace = "I"; break; // LATIN CAPITAL LETTER I BREVE case '\u012d': replace = "i"; break; // LATIN SMALL LETTER I BREVE case '\u014c': replace = "O"; break; // LATIN CAPITAL LETTER O MACRON case '\u014d': replace = "o"; break; // LATIN SMALL LETTER O MACRON case '\u014e': replace = "O"; break; // LATIN CAPITAL LETTER O BREVE case '\u014f': replace = "o"; break; // LATIN SMALL LETTER O BREVE case '\u0152': replace = "Oe"; break; // LATIN CAPITAL LETTER O E case '\u0153': replace = "oe"; break; // LATIN SMALL LETTER O E case '\u016a': replace = "U"; break; // LATIN CAPITAL LETTER U MACRON case '\u016b': replace = "u"; break; // LATIN SMALL LETTER U MACRON case '\u016c': replace = "U"; break; // LATIN CAPITAL LETTER U BREVE case '\u016d': replace = "u"; break; // LATIN SMALL LETTER U BREVE case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S // new in MPDL project by J. Willenborg case '\u1e8d': replace = "e"; break; // LATIN SMALL LETTER E WITH TILDE // by Malcolm case '\u00ad': break; // soft hyphen case '\u2329': break; // BRA case '\u232a': break; // KET default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("fr")) { // new Mpdl code: added by J. Willenborg: some of Malcolms code did not work without errors so it has to be taken away StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u00e6': replace = "ae"; break; // LATIN SMALL LETTER A E case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S case '\u00df': replace = "ss"; break; // LATIN SMALL LETTER SHARP S case '\u00ad': break; // soft hyphen case '-': break; default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("de")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u00c4': replace = "Ae"; break; case '\u00d6': replace = "Oe"; break; case '\u00dc': replace = "Ue"; break; case '\u00df': replace = "ss"; break; case '\u00e4': replace = "ae"; break; case '\u00f6': replace = "oe"; break; case '\u00fc': replace = "ue"; break; case '\u00ad': break; // soft hyphen case '\u00e9': replace = "e"; break; // new in MPDL project by J. Willenborg case '\u017f': replace = "s"; break; // LATIN SMALL LETTER LONG S // case '-': break; default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("zh")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u00b9': replace = "1"; break; case '\u00b2': replace = "2"; break; case '\u00b3': replace = "3"; break; case '\u2074': replace = "4"; break; case '\u2075': replace = "5"; break; // original by Malcolm Hyman: with the following replacements // case '\u3000': replace = " "; break; // case '\u3001': replace = ","; break; // case '\u3002': replace = "."; break; // case '\u200b': break; // BREAKS EVERYTHING! default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("akk") || language.equals("qam") || language.equals("qpc") || language.equals("elx") || language.equals("sux") || language.equals("hit") || language.equals("qhu") || language.equals("peo") || language.equals("uga") || language.equals("ura") || language.equals("qcu")) { StringBuffer buf = new StringBuffer(); int n = 0; char last = '\u0000'; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); c = Character.toLowerCase(c); String replace = new String(); switch (c) { case '{': replace += "-"; break; case '}': replace += "-"; break; // These are from PSD::ATF::Unicode by Steve Tinney case '\u0161': replace += "sz"; break; case '\u1e63': replace += "s,"; break; case '\u1e6d': replace += "t,"; break; case '\u014b': replace += "j"; break; case '\u015b': replace += "s'"; break; case '\u2080': replace += "0"; break; case '\u2081': replace += "1"; break; case '\u2082': replace += "2"; break; case '\u2083': replace += "3"; break; case '\u2084': replace += "4"; break; case '\u2085': replace += "5"; break; case '\u2086': replace += "6"; break; case '\u2087': replace += "7"; break; case '\u2088': replace += "8"; break; case '\u2089': replace += "9"; break; case 'c': // shin (except where used as modifier) if ((i > 0) && ((last == '~') || (last == '@'))) replace += "c"; else replace += "sz"; break; default: replace += c; break; } // suppress grapheme boundary before or after word boundary if (replace.equals("-")) { if ((i + 1 == s.length()) || (s.charAt(i + 1) == ' ') || (i == 0) || (s.charAt(i - 1) == ' ')) replace = ""; } last = c; buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("el") || language.equals("grc")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u03c2': replace = "\u03c3"; break; // GREEK SMALL LETTER FINAL SIGMA case '<': break; case '>': break; case '[': break; case ']': break; case '1': break; case '2': break; case '\u03ac': replace = "\u1f71"; break; case '\u03ad': replace = "\u1f73"; break; case '\u03ae': replace = "\u1f75"; break; case '\u03af': replace = "\u1f77"; break; case '\u03cc': replace = "\u1f79"; break; case '\u03cd': replace = "\u1f7b"; break; case '\u03ce': replace = "\u1f7d"; break; case '-': break; // same treatment as soft hyphen case '\u00ad': break; // soft hyphen default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else if (language.equals("el_atonic")) { StringBuffer buf = new StringBuffer(); int n = 0; for (int i = 0; i < s.length(); i++) { char c = s.charAt(i); String replace = new String(); switch (c) { case '\u03c2': replace = "\u03c3"; break; // GREEK SMALL LETTER FINAL SIGMA // map characters with diacritics to their plain equivalent // cf. <code>BetaCode.java</code> case '\u03aa': replace = "\u0399"; break; case '\u03ab': replace = "\u03a5"; break; case '\u03ac': replace = "\u0381"; break; case '\u03ad': replace = "\u0385"; break; case '\u03ae': replace = "\u0387"; break; case '\u03af': replace = "\u0389"; break; case '\u03ca': replace = "\u03b9"; break; case '\u03cb': replace = "\u03c5"; break; case '\u03cc': replace = "\u03bf"; break; case '\u03cd': replace = "\u03c5"; break; case '\u03ce': replace = "\u03c9"; break; case '\u1f00': replace = "\u03b1"; break; case '\u1f01': replace = "\u03b1"; break; case '\u1f02': replace = "\u03b1"; break; case '\u1f03': replace = "\u03b1"; break; case '\u1f04': replace = "\u03b1"; break; case '\u1f05': replace = "\u03b1"; break; case '\u1f06': replace = "\u03b1"; break; case '\u1f07': replace = "\u03b1"; break; case '\u1f08': replace = "\u0391"; break; case '\u1f09': replace = "\u0391"; break; case '\u1f0a': replace = "\u0391"; break; case '\u1f0b': replace = "\u0391"; break; case '\u1f0c': replace = "\u0391"; break; case '\u1f0d': replace = "\u0391"; break; case '\u1f0e': replace = "\u0391"; break; case '\u1f0f': replace = "\u0391"; break; case '\u1f10': replace = "\u03b5"; break; case '\u1f11': replace = "\u03b5"; break; case '\u1f12': replace = "\u03b5"; break; case '\u1f13': replace = "\u03b5"; break; case '\u1f14': replace = "\u03b5"; break; case '\u1f15': replace = "\u03b5"; break; case '\u1f18': replace = "\u0395"; break; case '\u1f19': replace = "\u0395"; break; case '\u1f1a': replace = "\u0395"; break; case '\u1f1b': replace = "\u0395"; break; case '\u1f1c': replace = "\u0395"; break; case '\u1f1d': replace = "\u0395"; break; case '\u1f20': replace = "\u03b7"; break; case '\u1f21': replace = "\u03b7"; break; case '\u1f22': replace = "\u03b7"; break; case '\u1f23': replace = "\u03b7"; break; case '\u1f24': replace = "\u03b7"; break; case '\u1f25': replace = "\u03b7"; break; case '\u1f26': replace = "\u03b7"; break; case '\u1f27': replace = "\u03b7"; break; case '\u1f28': replace = "\u0397"; break; case '\u1f29': replace = "\u0397"; break; case '\u1f2a': replace = "\u0397"; break; case '\u1f2b': replace = "\u0397"; break; case '\u1f2c': replace = "\u0397"; break; case '\u1f2d': replace = "\u0397"; break; case '\u1f2e': replace = "\u0397"; break; case '\u1f2f': replace = "\u0397"; break; case '\u1f30': replace = "\u03b9"; break; case '\u1f31': replace = "\u03b9"; break; case '\u1f32': replace = "\u03b9"; break; case '\u1f33': replace = "\u03b9"; break; case '\u1f34': replace = "\u03b9"; break; case '\u1f35': replace = "\u03b9"; break; case '\u1f36': replace = "\u03b9"; break; case '\u1f37': replace = "\u03b9"; break; case '\u1f38': replace = "\u0399"; break; case '\u1f39': replace = "\u0399"; break; case '\u1f3a': replace = "\u0399"; break; case '\u1f3b': replace = "\u0399"; break; case '\u1f3c': replace = "\u0399"; break; case '\u1f3d': replace = "\u0399"; break; case '\u1f3e': replace = "\u0399"; break; case '\u1f3f': replace = "\u0399"; break; case '\u1f40': replace = "\u03bf"; break; case '\u1f41': replace = "\u03bf"; break; case '\u1f42': replace = "\u03bf"; break; case '\u1f43': replace = "\u03bf"; break; case '\u1f44': replace = "\u03bf"; break; case '\u1f45': replace = "\u03bf"; break; case '\u1f48': replace = "\u039f"; break; case '\u1f49': replace = "\u039f"; break; case '\u1f4a': replace = "\u039f"; break; case '\u1f4b': replace = "\u039f"; break; case '\u1f4c': replace = "\u039f"; break; case '\u1f4d': replace = "\u039f"; break; case '\u1f50': replace = "\u03c5"; break; case '\u1f51': replace = "\u03c5"; break; case '\u1f52': replace = "\u03c5"; break; case '\u1f53': replace = "\u03c5"; break; case '\u1f54': replace = "\u03c5"; break; case '\u1f55': replace = "\u03c5"; break; case '\u1f56': replace = "\u03c5"; break; case '\u1f57': replace = "\u03c5"; break; case '\u1f58': replace = "\u03a5"; break; case '\u1f59': replace = "\u03a5"; break; case '\u1f5a': replace = "\u03a5"; break; case '\u1f5b': replace = "\u03a5"; break; case '\u1f5c': replace = "\u03a5"; break; case '\u1f5d': replace = "\u03a5"; break; case '\u1f5e': replace = "\u03a5"; break; case '\u1f5f': replace = "\u03a5"; break; case '\u1f60': replace = "\u03c9"; break; case '\u1f61': replace = "\u03c9"; break; case '\u1f62': replace = "\u03c9"; break; case '\u1f63': replace = "\u03c9"; break; case '\u1f64': replace = "\u03c9"; break; case '\u1f65': replace = "\u03c9"; break; case '\u1f66': replace = "\u03c9"; break; case '\u1f67': replace = "\u03c9"; break; case '\u1f68': replace = "\u03a9"; break; case '\u1f69': replace = "\u03a9"; break; case '\u1f6a': replace = "\u03a9"; break; case '\u1f6b': replace = "\u03a9"; break; case '\u1f6c': replace = "\u03a9"; break; case '\u1f6d': replace = "\u03a9"; break; case '\u1f6e': replace = "\u03a9"; break; case '\u1f6f': replace = "\u03a9"; break; case '\u1f70': replace = "\u03b1"; break; case '\u1f71': replace = "\u03b1"; break; case '\u1f72': replace = "\u03b5"; break; case '\u1f73': replace = "\u03b5"; break; case '\u1f74': replace = "\u03b7"; break; case '\u1f75': replace = "\u03b7"; break; case '\u1f76': replace = "\u03b9"; break; case '\u1f77': replace = "\u03b9"; break; case '\u1f78': replace = "\u03bf"; break; case '\u1f79': replace = "\u03bf"; break; case '\u1f7a': replace = "\u03c5"; break; case '\u1f7b': replace = "\u03c5"; break; case '\u1f7c': replace = "\u03c9"; break; case '\u1f7d': replace = "\u03c9"; break; case '\u1f80': replace = "\u03b1"; break; case '\u1f81': replace = "\u03b1"; break; case '\u1f82': replace = "\u03b1"; break; case '\u1f83': replace = "\u03b1"; break; case '\u1f84': replace = "\u03b1"; break; case '\u1f85': replace = "\u03b1"; break; case '\u1f86': replace = "\u03b1"; break; case '\u1f87': replace = "\u03b1"; break; case '\u1f88': replace = "\u0391"; break; case '\u1f89': replace = "\u0391"; break; case '\u1f8a': replace = "\u0391"; break; case '\u1f8b': replace = "\u0391"; break; case '\u1f8c': replace = "\u0391"; break; case '\u1f8d': replace = "\u0391"; break; case '\u1f8e': replace = "\u0391"; break; case '\u1f8f': replace = "\u0391"; break; case '\u1f90': replace = "\u03b7"; break; case '\u1f91': replace = "\u03b7"; break; case '\u1f92': replace = "\u03b7"; break; case '\u1f93': replace = "\u03b7"; break; case '\u1f94': replace = "\u03b7"; break; case '\u1f95': replace = "\u03b7"; break; case '\u1f96': replace = "\u03b7"; break; case '\u1f97': replace = "\u03b7"; break; case '\u1f98': replace = "\u0397"; break; case '\u1f99': replace = "\u0397"; break; case '\u1f9a': replace = "\u0397"; break; case '\u1f9b': replace = "\u0397"; break; case '\u1f9c': replace = "\u0397"; break; case '\u1f9d': replace = "\u0397"; break; case '\u1f9e': replace = "\u0397"; break; case '\u1f9f': replace = "\u0397"; break; case '\u1fa0': replace = "\u03c9"; break; case '\u1fa1': replace = "\u03c9"; break; case '\u1fa2': replace = "\u03c9"; break; case '\u1fa3': replace = "\u03c9"; break; case '\u1fa4': replace = "\u03c9"; break; case '\u1fa5': replace = "\u03c9"; break; case '\u1fa6': replace = "\u03c9"; break; case '\u1fa7': replace = "\u03c9"; break; case '\u1fa8': replace = "\u03a9"; break; case '\u1fa9': replace = "\u03a9"; break; case '\u1faa': replace = "\u03a9"; break; case '\u1fab': replace = "\u03a9"; break; case '\u1fac': replace = "\u03a9"; break; case '\u1fad': replace = "\u03a9"; break; case '\u1fae': replace = "\u03a9"; break; case '\u1faf': replace = "\u03a9"; break; case '\u1fb2': replace = "\u03b1"; break; case '\u1fb3': replace = "\u03b1"; break; case '\u1fb4': replace = "\u03b1"; break; case '\u1fb6': replace = "\u03b1"; break; case '\u1fb7': replace = "\u03b1"; break; case '\u1fba': replace = "\u0391"; break; case '\u1fbb': replace = "\u0391"; break; case '\u1fbc': replace = "\u0391"; break; case '\u1fc2': replace = "\u03b7"; break; case '\u1fc3': replace = "\u03b7"; break; case '\u1fc4': replace = "\u03b7"; break; case '\u1fc6': replace = "\u03b7"; break; case '\u1fc7': replace = "\u03b7"; break; case '\u1fca': replace = "\u0397"; break; case '\u1fcb': replace = "\u0397"; break; case '\u1fcc': replace = "\u0397"; break; case '\u1fd2': replace = "\u03b9"; break; case '\u1fd3': replace = "\u03b9"; break; case '\u1fd6': replace = "\u03b9"; break; case '\u1fd7': replace = "\u03b9"; break; case '\u1fda': replace = "\u0399"; break; case '\u1fdb': replace = "\u039f"; break; case '\u1fe2': replace = "\u03c5"; break; case '\u1fe3': replace = "\u03c5"; break; case '\u1fe4': replace = "\u03c1"; break; case '\u1fe5': replace = "\u03c1"; break; case '\u1fe6': replace = "\u03c5"; break; case '\u1fe7': replace = "\u03c5"; break; case '\u1fea': replace = "\u03a5"; break; case '\u1feb': replace = "\u03a5"; break; case '\u1fec': replace = "\u03a1"; break; case '\u1ff2': replace = "\u03c9"; break; case '\u1ff3': replace = "\u03c9"; break; case '\u1ff4': replace = "\u03c9"; break; case '\u1ff6': replace = "\u03c9"; break; case '\u1ff7': replace = "\u03c9"; break; case '\u1ff8': replace = "\u039f"; break; case '\u1ff9': replace = "\u039f"; break; case '\u1ffa': replace = "\u03a9"; break; case '\u1ffb': replace = "\u03a9"; break; case '\u1ffc': replace = "\u03a9"; break; case '<': break; case '>': break; case '-': break; // same treatment as soft hyphen case '\u00ad': break; // soft hyphen default: replace += c; break; } buf.append(replace); // update offsets if replacement is a different length if (offsets != null) { int r = replace.length(); if (r == 0) this.offsets = arrayKill(this.offsets, i - n); else if (r == 2) this.offsets = arrayInsert(this.offsets, i - n + 1, this.offsets[i - n], r - 1); n += 1 - r; } } return buf.toString(); } else { // unknown or no language return s; } } /* // explicit words normStr = normStr.replaceAll("aliàs", "alias"); normStr = normStr.replaceAll("hîc", "hic"); normStr = normStr.replaceAll("quòd", "quod"); normStr = normStr.replaceAll("Quòd", "Quod"); normStr = normStr.replaceAll("QVòd", "Quod"); normStr = normStr.replaceAll("Cùmque", "Cumque"); normStr = normStr.replaceAll("aër", "aer"); // ij normStr = normStr.replaceAll("ij", "ii"); // qu/qv normStr = normStr.replaceAll("qv", "qu"); // normStr = normStr.replaceAll("qV", "qU"); normStr = normStr.replaceAll("Qv", "Qu"); normStr = normStr.replaceAll("QV", "QU"); // u/v String vowels = getVowels(); String consonants = getConsonants(); normStr = normStr.replaceAll("([" + vowels + "])([-]*)u([" + vowels +"])", "$1$2v$3"); // vowel + u + vowel --> vowel + v + vowel normStr = normStr.replaceAll("([" + vowels + "])([-]*)U([" + vowels +"])", "$1$2V$3"); // vowel + U + vowel --> vowel + V + vowel normStr = normStr.replaceAll("([" + consonants + "])([-]*)v([" + consonants +"])", "$1$2u$3"); // consonant + v + consonant --> consonant + u + consonant normStr = normStr.replaceAll("([" + consonants + "])([-]*)V([" + consonants +"])", "$1$2U$3"); // consonant + V + consonant --> consonant + U + consonant normStr = normStr.replaceAll("^v([" + consonants +"])", "u$1"); // v + consonant --> u + consonant normStr = normStr.replaceAll("^V([" + consonants +"])", "U$1"); // V + consonant --> U + consonant // end of word: diacritica normStr = normStr.replaceAll("à$", "a"); normStr = normStr.replaceAll("è$", "e"); normStr = normStr.replaceAll("ò$", "o"); normStr = normStr.replaceAll("àm$", "am"); normStr = normStr.replaceAll("ùm$", "um"); String normStrTmp = normStr; normStr = ""; for (int i = 0; i < normStrTmp.length(); i++) { char c = normStrTmp.charAt(i); String replace = ""; switch (c) { case 'ſ': replace = "s"; break; case 'ß': replace = "ss"; break; case 'æ': replace = "ae"; break; case 'Æ': replace = "AE"; break; case 'ę': replace = "ae"; break; case 'œ': replace = "oe"; break; default: replace += c; break; } normStr = normStr + replace; } private String getVowels() { String retStr = null; if (Language.getInstance().isItalian(language)) { retStr = "AEIOUaeiou" + "\u00c6\u00e6" + // AE ligatures "\u0152\u0153"; // OE ligatures } else if (Language.getInstance().isLatin(language)) { retStr = "AEIOUaeiouÆœęàèòù"; } return retStr; } private String getConsonants() { String retStr = null; if (Language.getInstance().isItalian(language)) { retStr = "BCDFGHKLMNPQRSTVWXZ" + "bcdfghklmnpqrstvwxz" + "ſß"; // long/sharp S } else if (Language.getInstance().isLatin(language)) { retStr = "BCDFGHKLMNPQRSTVWXZ" + "bcdfghklmnpqrstvwxz" + "ſß"; // long/sharp S } return retStr; } * * * * */ /** * Returns a copy of an integer array with the element at * <code>index</code> removed ("killed"). * * @param array integer array * @param index index of element to remove */ private int[] arrayKill(int[] array, int index) { int[] newArray = new int[array.length - 1]; System.arraycopy(array, 0, newArray, 0, index); System.arraycopy(array, index + 1, newArray, index, array.length - index - 1); return newArray; } /** * Returns a copy of an integer array with <code>count</code> elements * inserted at <code>index</code>. * * @param array integer array * @param index index to insert new elements * @param value value to insert into new slots * @param count number of new slots to insert */ private int[] arrayInsert(int[] array, int index, int value, int count) { int[] newArray = new int[array.length + count]; System.arraycopy(array, 0, newArray, 0, index); for (int i = 0; i < count; i++) newArray[index + i] = value; System.arraycopy(array, index, newArray, index + count, array.length - index); return newArray; } }