Multibyte String 函数
在线手册：中文 英文
PHP手册

mb_decode_numericentity

(PHP 4 >= 4.0.6, PHP 5)

mb_decode_numericentity — Decode HTML numeric string reference to character

说明

string mb_decode_numericentity ( string $str , array $convmap , string $encoding )

Convert numeric string reference of string str in a specified block to character.

参数

str

The string being decoded.

convmap

convmap is an array that specifies the code area to convert.

encoding

encoding 参数为字符编码。如果省略，则使用内部字符编码。

返回值

The converted string.

范例

Example #1 convmap example

<?php
$convmap = array (
   int start_code1, int end_code1, int offset1, int mask1,
   int start_code2, int end_code2, int offset2, int mask2,
   ........
   int start_codeN, int end_codeN, int offsetN, int maskN );
// Specify Unicode value for start_codeN and end_codeN
// Add offsetN to value and take bit-wise 'AND' with maskN, 
// then convert value to numeric string reference.
?>

参见

Multibyte String 函数
在线手册：中文 英文
PHP手册
PHP手册 - N: Decode HTML numeric string reference to character

用户评论:

tom (13-Apr-2011 10:46)

When I use this function, I had found an error.
Example)
------------------------
INPUT STRING1 : abc&
INPUT STRING2 : abc&#
------------------------
OUTPUT STRING : abc
------------------------
<?php
    $input = 'abc&';
    $convmap = array (0x0, 0xffff, 0, 0xffff);
    $output = mb_decode_numericentity($intput, $convmap, 'UTF-8');
    echo $output;
?>
result : abc

If an input string is finished with some characters such the beginning of NCR-form, this function remove that characters.

So, I use an trick.

<?php
function decode_numericentity($string){
    $string = $string.chr(32);
    $string = mb_decode_numericentity($string, $convmap, 'UTF-8');
    $pos = strlen($string)-1;
    //if(ord($string[$pos]) == 32){
        $string = substr($string,0,$pos);
    //}
    return $string;
}
?>

Navi (01-Apr-2009 09:00)

Manual entity => utf8 conversion:
<?php
        // parse entities
        $raw = preg_replace_callback
        (
            "/&#(\\d+);/u",
            "_pcreEntityToUtf",
            $raw
        );

    function _pcreEntityToUtf($matches)
    {
        $char = intval(is_array($matches) ? $matches[1] : $matches);

        if ($char < 0x80)
        {
            // to prevent insertion of control characters
            if ($char >= 0x20) return htmlspecialchars(chr($char));
            else return "&#$char;";
        }
        else if ($char < 0x8000)
        {
            return chr(0xc0 | (0x1f & ($char >> 6))) . chr(0x80 | (0x3f & $char));
        }
        else
        {
            return chr(0xe0 | (0x0f & ($char >> 12))) . chr(0x80 | (0x3f & ($char >> 6))). chr(0x80 | (0x3f & $char));
        }
    }
?>

donovan at conduit it (19-Apr-2006 05:05)

note that at this time it seems that mb_decode_numericentity() only works with decimal entities and not hexadecimal entities.  This fact would have saved me a good hour of time in debugging.

For those who need to convert hex entities try first converting them all to decimal entities with a combination of the preg_replace() and hexdec() functions.

dirk at camindo de (30-Jan-2005 05:51)

By use of function utf8_decode you'll get a problem with all extended chars above ISO-8859-1 charset. You can solve this problem by using the

function mb_encode_numericentity before:

  // convert $text from UTF-8 to ISO-8859-1
  $convmap = array(0xFF, 0x2FFFF, 0, 0xFFFF);
  $text = mb_encode_numericentity($text, $convmap, "UTF-8");
  $text = utf8_decode($text);

The second line encodes all extended chars below 0xFF, the third line converts the rest: 0x80 - 0xFF

Andrew Simpson (11-Dec-2004 01:29)

Many web browsers will tend upload high order characters as UTF-8 encoded entities.

Here is some simple code to convert UTF-8 HTML entities within a block of text into proper characters:

<?php
   //decode decimal HTML entities added by web browser
  $body = preg_replace('/&#\d{2,5};/ue', "utf8_entity_decode('$0')", $body );
  //decode hex HTML entities added by web browser
  $body = preg_replace('/&#x([a-fA-F0-7]{2,8});/ue', "utf8_entity_decode('&#'.hexdec('$1').';')", $body );

//callback function for the regex
function utf8_entity_decode($entity){
 $convmap = array(0x0, 0x10000, 0, 0xfffff);
 return mb_decode_numericentity($entity, $convmap, 'UTF-8');
}
?>

php at cNhOiSpPpAlMe dot org (31-Mar-2004 08:55)

Here are functions to convert hankaku to zenkaku characters (and vice-versa) in Japanese text.

<?php

// Supported characters:
//    (space)
//     !#$%&()*+,./0123456789:;<=>?@
//    ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`
//    abcdefghijklmnopqrstuvwxyz{|}
// (Katakana isn't supported.)

function f_han2zen ($string,$encoding = null) {
  if (is_null($encoding)) $encoding = mb_internal_encoding();
  $convmap = array(
     0x20,0x20,0x3000-0x20,0xffff,   // Space
     0x21,0x7e,0xff01-0x21,0xffff);
  $temp = mb_encode_numericentity($string,$convmap,$encoding);
  $convmap = array(0,0xffff,0,0xffff);
  return mb_decode_numericentity($temp,$convmap,$encoding);
}
function f_zen2han ($string,$encoding = null) {
  if (is_null($encoding)) $encoding = mb_internal_encoding();
  $convmap = array(
     0x3000,0x3000,-(0x3000-0x20),0xffff,   // Space
     0xff01,0xff5e,-(0xff01-0x21),0xffff);
  $temp = mb_encode_numericentity($string,$convmap,$encoding);
  $convmap = array(0,0xffff,0,0xffff);
  return mb_decode_numericentity($temp,$convmap,$encoding);
}

// Sample usage:
f_han2zen("test","shift_jis");
f_han2zen("test","utf-8");

?>

dev at glossword info (19-Nov-2003 03:43)

Just two great functions for daily use:

/* Converts any HTML-entities into characters */
function my_numeric2character($t)
{
    $convmap = array(0x0, 0x2FFFF, 0, 0xFFFF);
    return mb_decode_numericentity($t, $convmap, 'UTF-8');
}
/* Converts any characters into HTML-entities */
function my_character2numeric($t)
{
    $convmap = array(0x0, 0x2FFFF, 0, 0xFFFF);
    return mb_encode_numericentity($t, $convmap, 'UTF-8');
}
print my_numeric2character('&#8217; &#7936; &#226;');
print my_character2numeric(' ');