Why doesn't the html_entity_decode() function convert entities without the last semicolon (like A or A) to characters?
---
<?php
echo 'like A or A';
---
Browser displays fine:
----
like A or A
(PHP 4 >= 4.3.0, PHP 5, PHP 7, PHP 8)
html_entity_decode — Convert HTML entities to their corresponding characters
$string
, int $flags = ENT_QUOTES | ENT_SUBSTITUTE | ENT_HTML401
, string|null $encoding = null
) : string
html_entity_decode() is the opposite of
htmlentities() in that it converts HTML entities
in the string to their corresponding characters.
More precisely, this function decodes all the entities (including all numeric entities) that a) are necessarily valid for the chosen document type — i.e., for XML, this function does not decode named entities that might be defined in some DTD — and b) whose character or characters are in the coded character set associated with the chosen encoding and are permitted in the chosen document type. All other entities are left as is.
stringThe input string.
flags
A bitmask of one or more of the following flags, which specify how to handle quotes and
which document type to use. The default is ENT_QUOTES | ENT_SUBSTITUTE | ENT_HTML401.
| Constant Name | Description |
|---|---|
ENT_COMPAT |
Will convert double-quotes and leave single-quotes alone. |
ENT_QUOTES |
Will convert both double and single quotes. |
ENT_NOQUOTES |
Will leave both double and single quotes unconverted. |
ENT_SUBSTITUTE |
Replace invalid code unit sequences with a Unicode Replacement Character U+FFFD (UTF-8) or � (otherwise) instead of returning an empty string. |
ENT_HTML401 |
Handle code as HTML 4.01. |
ENT_XML1 |
Handle code as XML 1. |
ENT_XHTML |
Handle code as XHTML. |
ENT_HTML5 |
Handle code as HTML 5. |
encodingAn optional argument defining the encoding used when converting characters.
If omitted, encoding defaults to the value of the
default_charset configuration
option.
Although this argument is technically optional, you are highly encouraged to specify the correct value for your code if the default_charset configuration option may be set incorrectly for the given input.
支持以下字符集：
| 字符集 | 别名 | 描述 |
|---|---|---|
| ISO-8859-1 | ISO8859-1 | 西欧，Latin-1 |
| ISO-8859-5 | ISO8859-5 | Little used cyrillic charset (Latin/Cyrillic). |
| ISO-8859-15 | ISO8859-15 | 西欧，Latin-9。增加欧元符号，法语和芬兰语字母在 Latin-1(ISO-8859-1) 中缺失。 |
| UTF-8 | ASCII 兼容的多字节 8 位 Unicode。 | |
| cp866 | ibm866, 866 | DOS 特有的西里尔编码。本字符集在 4.3.2 版本中得到支持。 |
| cp1251 | Windows-1251, win-1251, 1251 | Windows 特有的西里尔编码。本字符集在 4.3.2 版本中得到支持。 |
| cp1252 | Windows-1252, 1252 | Windows 特有的西欧编码。 |
| KOI8-R | koi8-ru, koi8r | 俄语。本字符集在 4.3.2 版本中得到支持。 |
| BIG5 | 950 | 繁体中文，主要用于中国台湾省。 |
| GB2312 | 936 | 简体中文，中国国家标准字符集。 |
| BIG5-HKSCS | 繁体中文，附带香港扩展的 Big5 字符集。 | |
| Shift_JIS | SJIS, 932 | 日语 |
| EUC-JP | EUCJP | 日语 |
| MacRoman | Mac OS 使用的字符串。 | |
'' |
An empty string activates detection from script encoding (Zend multibyte), default_charset and current locale (see nl_langinfo() and setlocale()), in this order. Not recommended. |
Note: 其他字符集没有认可。将会使用默认编码并抛出异常。
Returns the decoded string.
| 版本 | 说明 |
|---|---|
| 8.1.0 |
flags changed from ENT_COMPAT to ENT_QUOTES | ENT_SUBSTITUTE | ENT_HTML401.
|
| 8.0.0 |
encoding is nullable now.
|
Example #1 Decoding HTML entities
<?php
$orig = "I'll \"walk\" the <b>dog</b> now";
$a = htmlentities($orig);
$b = html_entity_decode($a);
echo $a; // I'll "walk" the <b>dog</b> now
echo $b; // I'll "walk" the <b>dog</b> now
?>
Note:
You might wonder why trim(html_entity_decode(' ')); doesn't reduce the string to an empty string, that's because the ' ' entity is not ASCII code 32 (which is stripped by trim()) but ASCII code 160 (0xa0) in the default ISO 8859-1 encoding.
Why doesn't the html_entity_decode() function convert entities without the last semicolon (like A or A) to characters?
---
<?php
echo 'like A or A';
---
Browser displays fine:
----
like A or A
I wanted to use this function today and I found the documentation, especially about the flags, not particularly helpful.
Running the code below, for example, failed because the flag I used was the wrong one...
$string = 'Donna's Bakery';
$title = html_entity_decode($string, ENT_HTML401, 'UTF-8');
echo $title;
The correct flag to use in this case is ENT_QUOTES.
My understanding of the flag to use is the one that would correspond to the expected, converted outcome. So, ENT_QUOTES for a character that would be a single or double quote when converted... and so on.
Please help make the documentation a bit clearer.
When using this function, it's a good idea to pay attention when it says that leaving the charset parameter empty is "not recommended".
I had an issue where I was storing text files, with entities converted, into a database. When I retrieved them later and ran
$text_file = html_entity_decode($text_data);
the entities were NOT decoded.
Once I was aware of the problem, I changed the decode call to fully specify all of the parameters:
$text_file = html_entity_decode($text_data, ENT_COMPAT | ENT_HTML5,'utf-8');
This converted the entities as expected.
Use the following to decode all entities:
<?php html_entity_decode($string, ENT_QUOTES | ENT_XML1, 'UTF-8') ?>
I've checked these special entities:
- double quotes (")
- single quotes (' and ')
- non printable chars (e.g. )
With other $flags some or all won't be decoded.
It seems that ENT_XML1 and ENT_XHTML are identical when decoding.
The following function decodes named and numeric HTML entities and works on UTF-8. Requires iconv.
function decodeHtmlEnt($str) {
$ret = html_entity_decode($str, ENT_COMPAT, 'UTF-8');
$p2 = -1;
for(;;) {
$p = strpos($ret, '&#', $p2+1);
if ($p === FALSE)
break;
$p2 = strpos($ret, ';', $p);
if ($p2 === FALSE)
break;
if (substr($ret, $p+2, 1) == 'x')
$char = hexdec(substr($ret, $p+3, $p2-$p-3));
else
$char = intval(substr($ret, $p+2, $p2-$p-2));
//echo "$char\n";
$newchar = iconv(
'UCS-4', 'UTF-8',
chr(($char>>24)&0xFF).chr(($char>>16)&0xFF).chr(($char>>8)&0xFF).chr($char&0xFF)
);
//echo "$newchar<$p<$p2<<\n";
$ret = substr_replace($ret, $newchar, $p, 1+$p2-$p);
$p2 = $p + strlen($newchar);
}
return $ret;
}
We were having very peculiar behavior regarding foreign characters such as e-acute.
However, it was only showing up as a problem when extracting those characters out of our mysql database and when being displayed through a proxy server of ours that handles dns issues.
As other users have made a note of, the default character setting wasn't what they were expecting it to be when they left theirs blank.
When we changed our default_charset to "UTF-8", our problems and needs for using functions like these were no longer necessary in handling foreign characters such as e-acute. Good enough for us!
If you need something that converts &#[0-9]+ entities to UTF-8, this is simple and works:
<?php
/* Entity crap. /
$input = "Fovič";
$output = preg_replace_callback("/(&#[0-9]+;)/", function($m) { return mb_convert_encoding($m[1], "UTF-8", "HTML-ENTITIES"); }, $input);
/* Plain UTF-8. */
echo $output;
?>
This is a safe rawurldecode with utf8 detection:
<?php
function utf8_rawurldecode($raw_url_encoded){
$enc = rawurldecode($raw_url_encoded);
if(utf8_encode(utf8_decode($enc))==$enc){;
return rawurldecode($raw_url_encoded);
}else{
return utf8_encode(rawurldecode($raw_url_encoded));
}
}
?>
Handy function to convert remaining HTML-entities into human readable chars (for entities which do not exist in target charset):
<?php
function cleanString($in,$offset=null)
{
$out = trim($in);
if (!empty($out))
{
$entity_start = strpos($out,'&',$offset);
if ($entity_start === false)
{
// ideal
return $out;
}
else
{
$entity_end = strpos($out,';',$entity_start);
if ($entity_end === false)
{
return $out;
}
// zu lang um eine entity zu sein
else if ($entity_end > $entity_start+7)
{
// und weiter gehts
$out = cleanString($out,$entity_start+1);
}
// gottcha!
else
{
$clean = substr($out,0,$entity_start);
$subst = substr($out,$entity_start+1,1);
// š => "s" / š => "_"
$clean .= ($subst != "#") ? $subst : "_";
$clean .= substr($out,$entity_end+1);
// und weiter gehts
$out = cleanString($clean,$entity_start+1);
}
}
}
return $out;
}
?>
I wrote in a previous comment that html_entity_decode() only handled about 100 characters. That's not quite true; it only handles entities that exist in the output character set (the third argument). If you want to get ALL HTML entities, make sure you use ENT_QUOTES and set the third argument to 'UTF-8'.
If you don't want a UTF-8 string, you'll need to convert it afterward with something like utf8_decode(), iconv(), or mb_convert_encoding().
If you're producing XML, which doesn't recognise most HTML entities:
When producing a UTF-8 document (the default), then htmlspecialchars(html_entity_decode($string, ENT_QUOTES, 'UTF-8'), ENT_NOQUOTES, 'UTF-8') (because you only need to escape < and > and & unless you're printing inside the XML tags themselves).
Otherwise, either convert all the named entities to numeric ones, or declare the named entities in the document's DTD. The full list of 252 entities can be found in the HTML 4.01 Spec, or you can cut and paste the function from my site (http://inanimatt.com/php-convert-entities.php).
I just ran into the:
Bug #27626 html_entity_decode bug - cannot yet handle MBCS in html_entity_decode()!
The simple solution if you're still running PHP 4 is to wrap the html_entity_decode() function with the utf8_decode() function.
<?php
$string = ' ';
$utf8_encode = utf8_encode(html_entity_decode($string));
?>
By default html_entity_decode() returns the ISO-8859-1 character set, and by default utf8_decode()...
http://us.php.net/manual/en/function.utf8-decode.php
"Converts a string with ISO-8859-1 characters encoded with UTF-8 to single-byte ISO-8859-1"
I created this function to filter all the text that goes in or comes out of the database.
<?php
function filter_string($string, $nohtml='', $save='') {
if(!empty($nohtml)) {
$string = trim($string);
if(!empty($save)) $string = htmlentities(trim($string), ENT_QUOTES, 'ISO-8859-15');
else $string = html_entity_decode($string, ENT_QUOTES, 'ISO-8859-15');
}
if(!empty($save)) $string = mysql_real_escape_string($string);
else $string = stripslashes($string);
return($string);
}
?>
the references to 'chr()' in the example unhtmlentities() function should be changed to unichr, using the example unichr() function described in the 'chr' reference (http://php.net/chr).
the reason for this is characters such as € which do not break down into an ASCII number (that's the Euro, by the way).
I had a problem getting the 'TM' trademark symbol to display correctly in an email subject line. Using html_entity_decode() with different charsets didn't work, but directly replacing the entity with it's ASCII equivalent did:
$subject = str_replace('™', chr(153), $subject);
The decipherment does the character encoded by the escape function of JavaScript.
When the multi byte is used on the page, it is effective.
javascript escape('aaああaa') ..... 'aa%u3042%u3042aa'
php jsEscape_decode('aa%u3042%u3042aa')..'aaああaa'
<?php
function jsEscape_decode($jsEscaped,$outCharCode='SJIS'){
$arrMojis = explode("%u",$jsEscaped);
for ($i = 1;$i < count($arrMojis);$i++){
$c = substr($arrMojis[$i],0,4);
$cc = mb_convert_encoding(pack('H*',$c),$outCharCode,'UTF-16');
$arrMojis[$i] = substr_replace($arrMojis[$i],$cc,0,4);
}
return implode('',$arrMojis);
}
?>
Here is the ultimate functions to convert HTML entities to UTF-8 :
The main function is htmlentities2utf8
Others are helper functions
<?php
function chr_utf8($code)
{
if ($code < 0) return false;
elseif ($code < 128) return chr($code);
elseif ($code < 160) // Remove Windows Illegals Cars
{
if ($code==128) $code=8364;
elseif ($code==129) $code=160; // not affected
elseif ($code==130) $code=8218;
elseif ($code==131) $code=402;
elseif ($code==132) $code=8222;
elseif ($code==133) $code=8230;
elseif ($code==134) $code=8224;
elseif ($code==135) $code=8225;
elseif ($code==136) $code=710;
elseif ($code==137) $code=8240;
elseif ($code==138) $code=352;
elseif ($code==139) $code=8249;
elseif ($code==140) $code=338;
elseif ($code==141) $code=160; // not affected
elseif ($code==142) $code=381;
elseif ($code==143) $code=160; // not affected
elseif ($code==144) $code=160; // not affected
elseif ($code==145) $code=8216;
elseif ($code==146) $code=8217;
elseif ($code==147) $code=8220;
elseif ($code==148) $code=8221;
elseif ($code==149) $code=8226;
elseif ($code==150) $code=8211;
elseif ($code==151) $code=8212;
elseif ($code==152) $code=732;
elseif ($code==153) $code=8482;
elseif ($code==154) $code=353;
elseif ($code==155) $code=8250;
elseif ($code==156) $code=339;
elseif ($code==157) $code=160; // not affected
elseif ($code==158) $code=382;
elseif ($code==159) $code=376;
}
if ($code < 2048) return chr(192 | ($code >> 6)) . chr(128 | ($code & 63));
elseif ($code < 65536) return chr(224 | ($code >> 12)) . chr(128 | (($code >> 6) & 63)) . chr(128 | ($code & 63));
else return chr(240 | ($code >> 18)) . chr(128 | (($code >> 12) & 63)) . chr(128 | (($code >> 6) & 63)) . chr(128 | ($code & 63));
}
// Callback for preg_replace_callback('~&(#(x?))?([^;]+);~', 'html_entity_replace', $str);
function html_entity_replace($matches)
{
if ($matches[2])
{
return chr_utf8(hexdec($matches[3]));
} elseif ($matches[1])
{
return chr_utf8($matches[3]);
}
switch ($matches[3])
{
case "nbsp": return chr_utf8(160);
case "iexcl": return chr_utf8(161);
case "cent": return chr_utf8(162);
case "pound": return chr_utf8(163);
case "curren": return chr_utf8(164);
case "yen": return chr_utf8(165);
//... etc with all named HTML entities
}
return false;
}
function htmlentities2utf8 ($string) // because of the html_entity_decode() bug with UTF-8
{
$string = preg_replace_callback('~&(#(x?))?([^;]+);~', 'html_entity_replace', $string);
return $string;
}
?>
Note that
<?php
echo urlencode(html_entity_decode(" "));
?>
will output "%A0" instead of "+".
Quick & dirty code that translates numeric entities to UTF-8.
<?php
function replace_num_entity($ord)
{
$ord = $ord[1];
if (preg_match('/^x([0-9a-f]+)$/i', $ord, $match))
{
$ord = hexdec($match[1]);
}
else
{
$ord = intval($ord);
}
$no_bytes = 0;
$byte = array();
if ($ord < 128)
{
return chr($ord);
}
elseif ($ord < 2048)
{
$no_bytes = 2;
}
elseif ($ord < 65536)
{
$no_bytes = 3;
}
elseif ($ord < 1114112)
{
$no_bytes = 4;
}
else
{
return;
}
switch($no_bytes)
{
case 2:
{
$prefix = array(31, 192);
break;
}
case 3:
{
$prefix = array(15, 224);
break;
}
case 4:
{
$prefix = array(7, 240);
}
}
for ($i = 0; $i < $no_bytes; $i++)
{
$byte[$no_bytes - $i - 1] = (($ord & (63 * pow(2, 6 * $i))) / pow(2, 6 * $i)) & 63 | 128;
}
$byte[0] = ($byte[0] & $prefix[0]) | $prefix[1];
$ret = '';
for ($i = 0; $i < $no_bytes; $i++)
{
$ret .= chr($byte[$i]);
}
return $ret;
}
$test = 'This is a čא test'';
echo $test . "<br />\n";
echo preg_replace_callback('/&#([0-9a-fx]+);/mi', 'replace_num_entity', $test);
?>
This function seems to have to have two limitations (at least in PHP 4.3.8):
a) it does not work with multibyte character codings, such as UTF-8
b) it does not decode numeric entity references
a) can be solved by using iconv to convert to ISO-8859-1, then decoding the entities, than convert to UTF-8 again. But that's quite ugly and detroys all characters not present in Latin-1.
b) can be solved rather nicely using the following code:
<?php
function decode_entities($text) {
$text= html_entity_decode($text,ENT_QUOTES,"ISO-8859-1"); #NOTE: UTF-8 does not work!
$text= preg_replace('/&#(\d+);/me',"chr(\\1)",$text); #decimal notation
$text= preg_replace('/&#x([a-f0-9]+);/mei',"chr(0x\\1)",$text); #hex notation
return $text;
}
?>
HTH
This functionality is now implemented in the PEAR package PHP_Compat.
More information about using this function without upgrading your version of PHP can be found on the below link:
http://pear.php.net/package/PHP_Compat