\r
package com.hughes.android.dictionary.engine;\r
\r
-import java.util.Arrays;\r
+import com.hughes.android.dictionary.DictionaryApplication;\r
+import java.text.Collator;\r
+\r
+import java.util.Comparator;\r
import java.util.LinkedHashMap;\r
-import java.util.List;\r
import java.util.Locale;\r
import java.util.Map;\r
-\r
-import com.ibm.icu.text.Collator;\r
+import java.util.regex.Pattern;\r
\r
public class Language {\r
\r
- public static final Map<String,String> isoCodeToWikiName = new LinkedHashMap<String,String>();\r
- static {\r
-// Albanian\r
-// Armenian\r
-// Belarusian\r
-// Bengali\r
-// Bosnian\r
-// Bulgarian\r
-// Catalan\r
-// Esperanto\r
-// Estonian\r
-// Hungarian\r
-// Indonesian\r
-// Kurdish\r
-// Latin\r
-// Lithuanian\r
-// Nepali\r
-// Punjabi\r
-// Swahili\r
- isoCodeToWikiName.put("AF", "Afrikaans");\r
- isoCodeToWikiName.put("AR", "Arabic");\r
- isoCodeToWikiName.put("HY", "Armenian");\r
- isoCodeToWikiName.put("HR", "Croation");\r
- isoCodeToWikiName.put("CS", "Czech");\r
- isoCodeToWikiName.put("ZH", "Chinese|Mandarin|Cantonese");\r
- isoCodeToWikiName.put("DA", "Danish");\r
- isoCodeToWikiName.put("NL", "Dutch");\r
- isoCodeToWikiName.put("EN", "English");\r
- isoCodeToWikiName.put("FI", "Finish");\r
- isoCodeToWikiName.put("FR", "French");\r
- isoCodeToWikiName.put("DE", "German");\r
- isoCodeToWikiName.put("EL", "Greek");\r
- isoCodeToWikiName.put("haw", "Hawaiian");\r
- isoCodeToWikiName.put("HE", "Hewbrew");\r
- isoCodeToWikiName.put("HI", "Hindi");\r
- isoCodeToWikiName.put("IS", "Icelandic");\r
- isoCodeToWikiName.put("GA", "Irish");\r
- isoCodeToWikiName.put("IT", "Italian");\r
- isoCodeToWikiName.put("LT", "Lithuanian");\r
- isoCodeToWikiName.put("JA", "Japanese");\r
- isoCodeToWikiName.put("KO", "Korean");\r
- isoCodeToWikiName.put("KU", "Kurdish");\r
- isoCodeToWikiName.put("MS", "Malay");\r
- isoCodeToWikiName.put("MI", "Maori");\r
- isoCodeToWikiName.put("MN", "Mongolian");\r
- isoCodeToWikiName.put("NO", "Norwegian");\r
- isoCodeToWikiName.put("FA", "Persian");\r
- isoCodeToWikiName.put("PT", "Portuguese");\r
- isoCodeToWikiName.put("RO", "Romanian");\r
- isoCodeToWikiName.put("RU", "Russian");\r
- isoCodeToWikiName.put("SA", "Sanskrit");\r
- isoCodeToWikiName.put("SR", "Serbian");\r
- isoCodeToWikiName.put("SO", "Somali");\r
- isoCodeToWikiName.put("ES", "Spanish");\r
- isoCodeToWikiName.put("SU", "Sudanese");\r
- isoCodeToWikiName.put("SV", "Swedish");\r
- isoCodeToWikiName.put("TG", "Tajik");\r
- isoCodeToWikiName.put("TH", "Thai");\r
- isoCodeToWikiName.put("BO", "Tibetan");\r
- isoCodeToWikiName.put("TR", "Turkish");\r
- isoCodeToWikiName.put("UK", "Ukranian");\r
- isoCodeToWikiName.put("VI", "Vietnamese");\r
- isoCodeToWikiName.put("CI", "Welsh");\r
- isoCodeToWikiName.put("YI", "Yiddish");\r
- isoCodeToWikiName.put("ZU", "Zulu");\r
- }\r
-\r
- static final List<String> ISO_CODES_WITH_DICTS = Arrays.asList();\r
-\r
- static final Map<String, Language> symbolToLangauge = new LinkedHashMap<String, Language>();\r
-\r
- final String symbol;\r
- final Locale locale;\r
- \r
- private Collator collator;\r
-\r
- public Language(final Locale locale) {\r
- this.symbol = locale.getLanguage();\r
- this.locale = locale;\r
-\r
- symbolToLangauge.put(symbol.toLowerCase(), this);\r
- }\r
-\r
- @Override\r
- public String toString() {\r
- return locale.toString();\r
- }\r
- \r
- public String getSymbol() {\r
- return symbol;\r
- }\r
- \r
- public synchronized Collator getCollator() {\r
- if (collator == null) {\r
- this.collator = Collator.getInstance(locale);\r
- this.collator.setStrength(Collator.IDENTICAL);\r
+ public static final class LanguageResources {\r
+ public final String englishName;\r
+ public final int nameId;\r
+ public final int flagId;\r
+\r
+ public LanguageResources(final String englishName, int nameId, int flagId) {\r
+ this.englishName = englishName;\r
+ this.nameId = nameId;\r
+ this.flagId = flagId;\r
+ }\r
+\r
+ public LanguageResources(final String englishName, int nameId) {\r
+ this(englishName, nameId, 0);\r
+ }\r
+ }\r
+\r
+ private static final Map<String, Language> registry = new LinkedHashMap<String, Language>();\r
+\r
+ final String isoCode;\r
+ final Locale locale;\r
+\r
+ private Collator collator;\r
+\r
+ private Language(final Locale locale, final String isoCode) {\r
+ this.locale = locale;\r
+ this.isoCode = isoCode;\r
+\r
+ registry.put(isoCode.toLowerCase(), this);\r
}\r
- return collator;\r
- }\r
- \r
- public String getDefaultNormalizerRules() {\r
- return ":: Any-Latin; :: Lower; :: NFD; :: [:Nonspacing Mark:] Remove; :: NFC ;";\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
-\r
- public static final Language de = new Language(Locale.GERMAN) {\r
+\r
@Override\r
+ public String toString() {\r
+ return locale.toString();\r
+ }\r
+\r
+ public String getIsoCode() {\r
+ return isoCode;\r
+ }\r
+\r
+ public synchronized Comparator getCollator() {\r
+ if (!DictionaryApplication.USE_COLLATOR)\r
+ return String.CASE_INSENSITIVE_ORDER;\r
+ // Don't think this is thread-safe...\r
+ // if (collator == null) {\r
+ this.collator = Collator.getInstance(locale);\r
+ this.collator.setStrength(Collator.IDENTICAL);\r
+ // }\r
+ return collator;\r
+ }\r
+\r
public String getDefaultNormalizerRules() {\r
- return ":: Lower; 'ae' > 'ä'; 'oe' > 'ö'; 'ue' > 'ü'; 'ß' > 'ss'; ";\r
+ return ":: Any-Latin; ' ' > ; :: Lower; :: NFD; :: [:Nonspacing Mark:] Remove; :: NFC ;";\r
}\r
- };\r
- \r
- // ----------------------------------------------------------------\r
-\r
- public static synchronized Language lookup(final String symbol) {\r
- Language lang = symbolToLangauge.get(symbol.toLowerCase());\r
- if (lang == null) {\r
- lang = new Language(new Locale(symbol));\r
+\r
+ /**\r
+ * A practical pattern to identify strong RTL characters. This pattern is\r
+ * not completely correct according to the Unicode standard. It is\r
+ * simplified for 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 + "])(["\r
+ + rtlChars + "])");\r
+ private static final Pattern RTL_RIGHT_BOUNDARY = Pattern.compile("([" + rtlChars + "])(["\r
+ + 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, "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, "DE") {\r
+ @Override\r
+ public String getDefaultNormalizerRules() {\r
+ return ":: Lower; 'ae' > 'ä'; 'oe' > 'ö'; 'ue' > 'ü'; 'ß' > 'ss'; ";\r
+ }\r
+ };\r
+\r
+ // ----------------------------------------------------------------\r
+\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
- return lang;\r
- }\r
\r
}\r