advanced code snippet search

Posted By

Pazuzu on 04/06/11


Tagged

php xml json parser unicode utf-8


Versions (?)

Who likes this?

1 person have marked this snippet as a favorite

vitor_np


XML to JSON (Simple) Parser

 / Published in: PHP
 

The unicode decoding functions were adapted from the Decoder.php source from the Zend Framework.

Expand | Embed | Plain Text 
  1. <?php
  2. 	header("Content-type: text/html; charset=utf-8");
  3.  
  4. 	//http://framework.zend.com/svn/framework/standard/trunk/library/Zend/Json/Decoder.php
  5.     function utf162utf8($utf16) {
  6.         // Check for mb extension otherwise do by hand.
  7.         if( function_exists('mb_convert_encoding') ) {
  8.             return mb_convert_encoding($utf16, 'UTF-8', 'UTF-16');
  9.         }
  10.  
  11.         $bytes = (ord($utf16{0}) << 8) | ord($utf16{1});
  12.  
  13.         switch (true) {
  14.             case ((0x7F & $bytes) == $bytes):
  15.                 // this case should never be reached, because we are in ASCII range
  16.                 // see: http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  17.                 return chr(0x7F & $bytes);
  18.  
  19.             case (0x07FF & $bytes) == $bytes:
  20.                 // return a 2-byte UTF-8 character
  21.                 // see: http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  22.                 return chr(0xC0 | (($bytes >> 6) & 0x1F))
  23.                      . chr(0x80 | ($bytes & 0x3F));
  24.  
  25.             case (0xFFFF & $bytes) == $bytes:
  26.                 // return a 3-byte UTF-8 character
  27.                 // see: http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  28.                 return chr(0xE0 | (($bytes >> 12) & 0x0F))
  29.                      . chr(0x80 | (($bytes >> 6) & 0x3F))
  30.                      . chr(0x80 | ($bytes & 0x3F));
  31.         }
  32.  
  33.         // ignoring UTF-32 for now, sorry
  34.         return '';
  35.     }
  36.  
  37.    	function decodeUnicodeString($chrs)
  38.     {
  39.         $delim       = substr($chrs, 0, 1);
  40.         $utf8        = '';
  41.         $strlen_chrs = strlen($chrs);
  42.  
  43.         for($i = 0; $i < $strlen_chrs; $i++) {
  44.  
  45.             $substr_chrs_c_2 = substr($chrs, $i, 2);
  46.             $ord_chrs_c = ord($chrs[$i]);
  47.  
  48.             switch (true) {
  49.                 case preg_match('/\\\u[0-9A-F]{4}/i', substr($chrs, $i, 6)):
  50.                     // single, escaped unicode character
  51.                     $utf16 = chr(hexdec(substr($chrs, ($i + 2), 2)))
  52.                            . chr(hexdec(substr($chrs, ($i + 4), 2)));
  53.                     $utf8 .= utf162utf8($utf16);
  54.                     $i += 5;
  55.                     break;
  56.                 case ($ord_chrs_c >= 0x20) && ($ord_chrs_c <= 0x7F):
  57.                     $utf8 .= $chrs{$i};
  58.                     break;
  59.                 case ($ord_chrs_c & 0xE0) == 0xC0:
  60.                     // characters U-00000080 - U-000007FF, mask 110XXXXX
  61.                     //see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  62.                     $utf8 .= substr($chrs, $i, 2);
  63.                     ++$i;
  64.                     break;
  65.                 case ($ord_chrs_c & 0xF0) == 0xE0:
  66.                     // characters U-00000800 - U-0000FFFF, mask 1110XXXX
  67.                     // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  68.                     $utf8 .= substr($chrs, $i, 3);
  69.                     $i += 2;
  70.                     break;
  71.                 case ($ord_chrs_c & 0xF8) == 0xF0:
  72.                     // characters U-00010000 - U-001FFFFF, mask 11110XXX
  73.                     // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  74.                     $utf8 .= substr($chrs, $i, 4);
  75.                     $i += 3;
  76.                     break;
  77.                 case ($ord_chrs_c & 0xFC) == 0xF8:
  78.                     // characters U-00200000 - U-03FFFFFF, mask 111110XX
  79.                     // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  80.                     $utf8 .= substr($chrs, $i, 5);
  81.                     $i += 4;
  82.                     break;
  83.                 case ($ord_chrs_c & 0xFE) == 0xFC:
  84.                     // characters U-04000000 - U-7FFFFFFF, mask 1111110X
  85.                     // see http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8
  86.                     $utf8 .= substr($chrs, $i, 6);
  87.                     $i += 5;
  88.                     break;
  89.             }
  90.         }
  91.  
  92.         return $utf8;
  93.     }	
  94.  
  95.  
  96. 	$json = json_encode(simplexml_load_file("the_file.xml")); 
  97. 	echo decodeUnicodeString( str_replace("\t", "", $json) );
  98.  
  99. ?>

Report this snippet  

You need to login to post a comment.