Author: sback
Date: 2007-07-23 13:08:36 +0000 (Mon, 23 Jul 2007)
New Revision: 14274

Modified:
   trunk/freenet/test/freenet/utils/UTFUtil.java
Log:
UTF representation of HTML entities added

Modified: trunk/freenet/test/freenet/utils/UTFUtil.java
===================================================================
--- trunk/freenet/test/freenet/utils/UTFUtil.java       2007-07-23 11:54:06 UTC 
(rev 14273)
+++ trunk/freenet/test/freenet/utils/UTFUtil.java       2007-07-23 13:08:36 UTC 
(rev 14274)
@@ -57,4 +57,45 @@
                
'\u0173','\u00de','\u00f7','\u0110','\u0129','\u0142','\u015b','\u0174','\u00df','\u00f8',
                
'\u0111','\u012a','\u0143','\u015c','\u0175','\u00e0','\u00f9','\u0112','\u012b','\u0144',
                
'\u015d','\u0176','\u00e1','\u00fa','\u0113','\u012c','\u0145','\u015e','\u0177'};
+       
+       /*
+        * HTML 4.01 entities as referred by W3Schools 
+        * http://www.w3schools.com/tags/ref_entities.asp
+        */
+       public static final String HTML_ENTITIES_UTF[][] = {
+               //ISO 8859-1 Symbol Entities
+               
{"\u00a1","¡"},{"\u00a4","¤"},{"\u00a2","¢"},{"\u00a3","£"},
+               
{"\u00a5","¥"},{"\u00a6","¦"},{"\u00a7","§"},{"\u00a8","¨"},{"\u00a9","©"},
+               
{"\u00aa","ª"},{"\u00ab","«"},{"\u00ac","¬"},{"\u00ad","­"},{"\u00ae","®"},
+               
{"\u2122","™"},{"\u00af","¯"},{"\u00b0","°"},{"\u00b1","±"},
+               
{"\u00b2","²"},{"\u00b3","³"},{"\u00b4","´"},{"\u00b5","µ"},
+               
{"\u00b6","¶"},{"\u00b7","·"},{"\u00b8","¸"},{"\u00b9","¹"},
+               
{"\u00ba","º"},{"\u00bb","»"},{"\u00bc","¼"},{"\u00bd","½"},
+               
{"\u00be","¾"},{"\u00bf","¿"},{"\u00d7","×"},{"\u00f7","÷"},
+               //ISO 8859-1 Character Entities
+               
{"\u00c0","À"},{"\u00c1","Á"},{"\u00c2","Â"},{"\u00c3","Ã"},
+               
{"\u00c4","Ä"},{"\u00c5","Å"},{"\u00c6","Æ"},{"\u00c7","Ç"},
+               
{"\u00c8","È"},{"\u00c9","É"},{"\u00ca","Ê"},{"\u00cb","Ë"},
+               
{"\u00cc","Ì"},{"\u00cd","Í"},{"\u00ce","Î"},{"\u00cf","Ë"},
+               
{"\u00d0","Ð"},{"\u00d1","Ñ"},{"\u00d2","Ò"},{"\u00d3","Ó"},
+               
{"\u00d4","Ô"},{"\u00d5","Õ"},{"\u00d6","Ö"},{"\u00d8","Ø"},
+               
{"\u00d9","Ù"},{"\u00da","Ú"},{"\u00db","Ú"},{"\u00dc","Ü"},
+               {"\u00dd","Ý"},{"\u00de","Þ"},{"\u00df","ß"},
+               
{"\u00e0","à"},{"\u00e1","á"},{"\u00e2","â"},{"\u00e3","ã"},
+               
{"\u00e4","ä"},{"\u00e5","å"},{"\u00e6","æ"},{"\u00e7","ç"},
+               
{"\u00e8","è"},{"\u00e9","é"},{"\u00ea","ê"},{"\u00eb","ë"},
+               
{"\u00ec","ì"},{"\u00ed","í"},{"\u00ee","î"},{"\u00ef","ï"},
+               {"\u00f0","ð"},{"\u00f1","ñ"},
+               
{"\u00f2","ò"},{"\u00f3","ó"},{"\u00f4","ô"},{"\u00f5","õ"},
+               {"\u00f6","ö"},{"\u00f8","ø"},
+               
{"\u00f9","ù"},{"\u00fa","ú"},{"\u00fb","û"},{"\u00fc","ü"},
+               {"\u00fd","ý"},{"\u00fe","þ"},{"\u00ff","ÿ"},
+               //Some Other Entities supported by HTML
+               
{"\u0152","Œ"},{"\u0153","œ"},{"\u0160","Š"},{"\u0161","š"},
+               
{"\u0178","&Ytml;"},{"\u02c6","ˆ"},{"\u02dc","˜"},{"\u2013","–"},
+               
{"\u2014","—"},{"\u2018","‘"},{"\u2019","’"},{"\u201a","‚"},
+               
{"\u201c","“"},{"\u201d","”"},{"\u201e","„"},{"\u2020","†"},
+               
{"\u2021","‡"},{"\u2026","…"},{"\u2030","‰"},{"\u2039","‹"},
+               {"\u203a","›"},{"\u20ac","€"}
+       };
 }


Reply via email to