Mercurial > hg > extraction-interface

diff develop/config/Lib_mb_utf8.php @ 6:63e08b98032f

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
rewrite extraction interface into PHP MVC architecture. (Although js hasn't been rewritten into MVC, it's fitted into the current PHP MVC architecture.) - The root of the new PHP MVC is at 'develop/'. - extraction interface is called "Extractapp" with several action, eg TaggingText, EditWordlist, EditTaglist, ExportTable.
author Zoe Hong <zhong@mpiwg-berlin.mpg.de>
date Thu, 05 Feb 2015 16:07:53 +0100
parents
children
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/develop/config/Lib_mb_utf8.php	Thu Feb 05 16:07:53 2015 +0100
@@ -0,0 +1,76 @@
+<?php
+   mb_regex_encoding('UTF-8');
+   mb_internal_encoding('UTF-8');
+
+   // ******************************************************************************
+
+   function mb($s) {
+      return mb8($s);
+   }
+   
+   function mb16($s) {
+      //return mb_convert_encoding($s, 'UTF-16LE');
+      return big5_to_utf16($s);
+   }
+   
+   function mb8($s) {
+      //return mb_convert_encoding($s, 'UTF-16LE');
+      return big5_to_utf8($s);
+   }
+   
+   function u8b5($s) {
+      return utf8_to_big5($s);
+   }
+   
+   function b5u8($s) {
+      return big5_to_utf8($s);
+   }
+   
+   function u8u16($s) {
+      return utf8_to_utf16($s);
+   }
+   
+   function u16u8($s) {
+      return utf16_to_utf8($s);
+   }
+   
+   function b5u16($s) {
+      return big5_to_utf16($s);
+   }
+   
+   function u16b5($s) {
+      return utf16_to_big5($s);
+   }
+   
+   function big5_to_utf16($s) {
+      if (!$s) return '';
+      return iconv("BIG-5", "UTF-16LE", $s);
+   }
+   
+   function utf16_to_big5($s) {
+      if (!$s) return '';
+      return iconv("UTF-16LE", "BIG-5//IGNORE", $s);
+   }
+   
+   function utf16_to_utf8($s) {
+      if (!$s) return '';
+      return iconv("UTF-16LE", "UTF-8", $s);
+   }
+   
+   function utf8_to_utf16($s) {
+      if (!$s) return '';
+      return iconv("UTF-8", "UTF-16LE", $s);
+   }
+   
+   function big5_to_utf8($s) {
+      if (!$s) return '';
+      return iconv("CP950", "UTF-8//IGNORE", $s);
+   }
+   
+   function utf8_to_big5($s) {
+      if (!$s) return '';
+      return iconv("UTF-8", "CP950//IGNORE", $s);
+   }
+   
+   // ******************************************************************************
+?>