+// Copyright 2011 Google Inc. All Rights Reserved.\r
+//\r
+// Licensed under the Apache License, Version 2.0 (the "License");\r
+// you may not use this file except in compliance with the License.\r
+// You may obtain a copy of the License at\r
+//\r
+// http://www.apache.org/licenses/LICENSE-2.0\r
+//\r
+// Unless required by applicable law or agreed to in writing, software\r
+// distributed under the License is distributed on an "AS IS" BASIS,\r
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\r
+// See the License for the specific language governing permissions and\r
+// limitations under the License.\r
+\r
package com.hughes.android.dictionary.engine;\r
\r
-import java.util.Comparator;\r
+import java.util.ArrayList;\r
import java.util.LinkedHashMap;\r
import java.util.Locale;\r
import java.util.Map;\r
+import java.util.regex.Pattern;\r
\r
+import com.hughes.android.dictionary.R;\r
import com.ibm.icu.text.Collator;\r
\r
public class Language {\r
\r
- static final Map<String, Language> symbolToLangauge = new LinkedHashMap<String, Language>();\r
+ public static final Map<String,Integer> isoCodeToResourceId = new LinkedHashMap<String,Integer>();\r
+ static {\r
+ isoCodeToResourceId.put("AF", R.string.AF);\r
+ isoCodeToResourceId.put("SQ", R.string.SQ);\r
+ isoCodeToResourceId.put("AR", R.string.AR);\r
+ isoCodeToResourceId.put("HY", R.string.HY);\r
+ isoCodeToResourceId.put("BE", R.string.BE);\r
+ isoCodeToResourceId.put("BN", R.string.BN);\r
+ isoCodeToResourceId.put("BS", R.string.BS);\r
+ isoCodeToResourceId.put("BG", R.string.BG);\r
+ isoCodeToResourceId.put("CA", R.string.CA);\r
+ isoCodeToResourceId.put("HR", R.string.HR);\r
+ isoCodeToResourceId.put("CS", R.string.CS);\r
+ isoCodeToResourceId.put("ZH", R.string.ZH);\r
+ isoCodeToResourceId.put("DA", R.string.DA);\r
+ isoCodeToResourceId.put("NL", R.string.NL);\r
+ isoCodeToResourceId.put("EN", R.string.EN);\r
+ isoCodeToResourceId.put("EO", R.string.EO);\r
+ isoCodeToResourceId.put("ET", R.string.ET);\r
+ isoCodeToResourceId.put("FI", R.string.FI);\r
+ isoCodeToResourceId.put("FR", R.string.FR);\r
+ isoCodeToResourceId.put("DE", R.string.DE);\r
+ isoCodeToResourceId.put("EL", R.string.EL);\r
+ isoCodeToResourceId.put("haw", R.string.haw);\r
+ isoCodeToResourceId.put("HE", R.string.HE);\r
+ isoCodeToResourceId.put("HI", R.string.HI);\r
+ isoCodeToResourceId.put("HU", R.string.HU);\r
+ isoCodeToResourceId.put("IS", R.string.IS);\r
+ isoCodeToResourceId.put("ID", R.string.ID);\r
+ isoCodeToResourceId.put("GA", R.string.GA);\r
+ isoCodeToResourceId.put("IT", R.string.IT);\r
+ isoCodeToResourceId.put("LA", R.string.LA);\r
+ isoCodeToResourceId.put("LV", R.string.LV);\r
+ isoCodeToResourceId.put("LT", R.string.LT);\r
+ isoCodeToResourceId.put("JA", R.string.JA);\r
+ isoCodeToResourceId.put("KO", R.string.KO);\r
+ isoCodeToResourceId.put("KU", R.string.KU);\r
+ isoCodeToResourceId.put("MS", R.string.MS);\r
+ isoCodeToResourceId.put("MI", R.string.MI);\r
+ isoCodeToResourceId.put("MN", R.string.MN);\r
+ isoCodeToResourceId.put("NE", R.string.NE);\r
+ isoCodeToResourceId.put("NO", R.string.NO);\r
+ isoCodeToResourceId.put("FA", R.string.FA);\r
+ isoCodeToResourceId.put("PL", R.string.PL);\r
+ isoCodeToResourceId.put("PT", R.string.PT);\r
+ isoCodeToResourceId.put("PA", R.string.PA);\r
+ isoCodeToResourceId.put("RO", R.string.RO);\r
+ isoCodeToResourceId.put("RU", R.string.RU);\r
+ isoCodeToResourceId.put("SA", R.string.SA);\r
+ isoCodeToResourceId.put("SR", R.string.SR);\r
+ isoCodeToResourceId.put("SK", R.string.SK);\r
+ isoCodeToResourceId.put("SO", R.string.SO);\r
+ isoCodeToResourceId.put("ES", R.string.ES);\r
+ isoCodeToResourceId.put("SW", R.string.SW);\r
+ isoCodeToResourceId.put("SV", R.string.SV);\r
+ isoCodeToResourceId.put("TL", R.string.TL);\r
+ isoCodeToResourceId.put("TG", R.string.TG);\r
+ isoCodeToResourceId.put("TH", R.string.TH);\r
+ isoCodeToResourceId.put("BO", R.string.BO);\r
+ isoCodeToResourceId.put("TR", R.string.TR);\r
+ isoCodeToResourceId.put("UK", R.string.UK);\r
+ isoCodeToResourceId.put("UR", R.string.UR);\r
+ isoCodeToResourceId.put("VI", R.string.VI);\r
+ isoCodeToResourceId.put("CI", R.string.CI);\r
+ isoCodeToResourceId.put("YI", R.string.YI);\r
+ isoCodeToResourceId.put("ZU", R.string.ZU);\r
+\r
+ isoCodeToResourceId.put("AZ", R.string.AZ);\r
+ isoCodeToResourceId.put("EU", R.string.EU);\r
+ isoCodeToResourceId.put("BR", R.string.BR);\r
+ isoCodeToResourceId.put("MR", R.string.MR);\r
+ isoCodeToResourceId.put("FO", R.string.FO);\r
+ isoCodeToResourceId.put("GL", R.string.GL);\r
+ isoCodeToResourceId.put("HT", R.string.HT);\r
+ isoCodeToResourceId.put("LB", R.string.LB);\r
+ isoCodeToResourceId.put("MK", R.string.MK);\r
+\r
+ // Hack to allow lower-case ISO codes to work:\r
+ for (final String isoCode : new ArrayList<String>(isoCodeToResourceId.keySet())) {\r
+ isoCodeToResourceId.put(isoCode.toLowerCase(), isoCodeToResourceId.get(isoCode));\r
+ }\r
\r
- final String symbol;\r
- final Locale locale;\r
+ }\r
\r
- Collator sortCollator;\r
- final Comparator<String> sortComparator;\r
\r
- private Collator findCollator;\r
- final Comparator<String> findComparator;\r
+ private static final Map<String, Language> registry = new LinkedHashMap<String, Language>();\r
\r
- public Language(final Locale locale) {\r
- this.symbol = locale.getLanguage();\r
- this.locale = locale;\r
+ final String isoCode;\r
+ final Locale locale;\r
+ \r
+ private Collator collator;\r
\r
- this.sortComparator = new Comparator<String>() {\r
- public int compare(final String s1, final String s2) {\r
- return getSortCollator().compare(textNorm(s1, false), textNorm(s2, false));\r
- }\r
- };\r
-\r
- this.findComparator = new Comparator<String>() {\r
- public int compare(final String s1, final String s2) {\r
- return getFindCollator().compare(textNorm(s1, false), textNorm(s2, false));\r
- }\r
- };\r
- \r
- symbolToLangauge.put(symbol.toLowerCase(), this);\r
- }\r
+ private Language(final Locale locale, final String isoCode) {\r
+ this.locale = locale;\r
+ this.isoCode = isoCode;\r
\r
- public String textNorm(final String s, final boolean toLower) {\r
- return toLower ? s.toLowerCase() : s;\r
+ registry.put(isoCode.toLowerCase(), this);\r
}\r
\r
@Override\r
return locale.toString();\r
}\r
\r
- public String getSymbol() {\r
- return symbol;\r
+ public String getIsoCode() {\r
+ return isoCode;\r
}\r
\r
- public synchronized Collator getFindCollator() {\r
- if (findCollator == null) {\r
- findCollator = Collator.getInstance(locale);\r
- findCollator.setDecomposition(Collator.CANONICAL_DECOMPOSITION);\r
- findCollator.setStrength(Collator.SECONDARY);\r
+ public synchronized Collator getCollator() {\r
+ if (collator == null) {\r
+ this.collator = Collator.getInstance(locale);\r
+ this.collator.setStrength(Collator.IDENTICAL);\r
}\r
- return findCollator;\r
+ return collator;\r
}\r
-\r
- public synchronized Collator getSortCollator() {\r
- if (sortCollator == null) {\r
- sortCollator = Collator.getInstance(locale);\r
- sortCollator.setDecomposition(Collator.CANONICAL_DECOMPOSITION);\r
- sortCollator.setStrength(Collator.IDENTICAL);\r
- }\r
- return sortCollator;\r
+ \r
+ public String getDefaultNormalizerRules() {\r
+ return ":: Any-Latin; ' ' > ; :: Lower; :: NFD; :: [:Nonspacing Mark:] Remove; :: NFC ;";\r
}\r
-\r
+ \r
+ /**\r
+ * A practical pattern to identify strong RTL characters. This pattern is not\r
+ * completely correct according to the Unicode standard. It is simplified for\r
+ * performance and small code size.\r
+ */\r
+ private static final String rtlChars =\r
+ "\u0591-\u07FF\uFB1D-\uFDFD\uFE70-\uFEFC";\r
+\r
+ private static final String puncChars =\r
+ "\\[\\]\\(\\)\\{\\}\\=";\r
+\r
+ private static final Pattern RTL_LEFT_BOUNDARY = Pattern.compile("(["+ puncChars +"])([" + rtlChars + "])");\r
+ private static final Pattern RTL_RIGHT_BOUNDARY = Pattern.compile("([" + rtlChars + "])(["+ puncChars +"])");\r
+ \r
+ public static String fixBidiText(String text) {\r
+// text = RTL_LEFT_BOUNDARY.matcher(text).replaceAll("$1\u200e $2");\r
+// text = RTL_RIGHT_BOUNDARY.matcher(text).replaceAll("$1 \u200e$2");\r
+ return text;\r
+ }\r
+ \r
// ----------------------------------------------------------------\r
\r
- public static final Language en = new Language(Locale.ENGLISH);\r
- public static final Language fr = new Language(Locale.FRENCH);\r
- public static final Language it = new Language(Locale.ITALIAN);\r
+ public static final Language en = new Language(Locale.ENGLISH, "EN");\r
+ public static final Language fr = new Language(Locale.FRENCH, "FR");\r
+ public static final Language it = new Language(Locale.ITALIAN, "IT");\r
\r
- public static final Language de = new Language(Locale.GERMAN) {\r
+ public static final Language de = new Language(Locale.GERMAN, "DE") {\r
@Override\r
- public String textNorm(String token, final boolean toLower) {\r
- if (toLower) {\r
- token = token.toLowerCase();\r
- }\r
- boolean sub = false;\r
- // This is meant to be fast: occurrences of ae, oe, ue are probably rare.\r
- for (int ePos = token.indexOf('e', 1); ePos != -1; ePos = token.indexOf(\r
- 'e', ePos + 1)) {\r
- final char pre = Character.toLowerCase(token.charAt(ePos - 1));\r
- if (pre == 'a' || pre == 'o' || pre == 'u') {\r
- sub = true;\r
- break;\r
- }\r
- }\r
- if (!sub) {\r
- return token;\r
- }\r
- \r
- token = token.replaceAll("ae", "ä");\r
- token = token.replaceAll("oe", "ö");\r
- token = token.replaceAll("ue", "ü");\r
-\r
- token = token.replaceAll("Ae", "Ä");\r
- token = token.replaceAll("Oe", "Ö");\r
- token = token.replaceAll("Ue", "Ü");\r
-\r
- token = token.replaceAll("AE", "Ä");\r
- token = token.replaceAll("OE", "Ö");\r
- token = token.replaceAll("UE", "Ü");\r
- \r
- return token; \r
+ public String getDefaultNormalizerRules() {\r
+ return ":: Lower; 'ae' > 'ä'; 'oe' > 'ö'; 'ue' > 'ü'; 'ß' > 'ss'; ";\r
}\r
};\r
\r
- static {\r
- for (final String lang : Locale.getISOLanguages()) {\r
- if (lookup(lang) == null) {\r
- new Language(new Locale(lang));\r
- }\r
- }\r
- }\r
-\r
// ----------------------------------------------------------------\r
\r
- public static Language lookup(final String symbol) {\r
- return symbolToLangauge.get(symbol.toLowerCase());\r
+ public static synchronized Language lookup(final String isoCode) {\r
+ Language lang = registry.get(isoCode.toLowerCase());\r
+ if (lang == null) {\r
+ lang = new Language(new Locale(isoCode), isoCode);\r
+ }\r
+ return lang;\r
}\r
\r
}\r
projects / Dictionary.git / blobdiff