Midwest PHP Call For Papers Open

htmlspecialchars

(PHP 4, PHP 5, PHP 7)

htmlspecialchars特殊文字を HTML エンティティに変換する

説明

htmlspecialchars ( string $string [, int $flags = ENT_COMPAT | ENT_HTML401 [, string $encoding = ini_get("default_charset") [, bool $double_encode = TRUE ]]] ) : string

文字の中には HTML において特殊な意味を持つものがあり、 それらの本来の値を表示したければ HTML の表現形式に変換してやらなければなりません。 この関数は、これらの変換を行った結果の文字列を返します。 入力文字列の中で名前付きエンティティに関連づけられたものを すべて変換する必要がある場合には、代わりに htmlentities() を使用してください。

この関数への入力文字列と最終的なドキュメントとの間で文字セットが一致している場合は、 この関数を使えば HTML ドキュメントに組み込むための準備として十分です。 しかし、この関数への入力の中に、最終的なドキュメントの文字セットでは符号化できない文字がある場合もあります。 そんな文字も (数値エンティティや名前つきエンティティなどで) 残したい場合は、 この関数と htmlentities() (これは、名前付きエンティティに対応する文字しか置換しません) を使うだけでは不十分です。そんな場合は mb_encode_numericentity() を使う必要があります。

変換対象となる文字
変換前 変換語
& (アンパサンド) &
" (ダブルクォート) ENT_NOQUOTES が指定されていない場合、"
' (シングルクォート) ' (ENT_HTML401 の場合) あるいは ' ( ENT_XML1ENT_XHTMLENT_HTML5 の場合)。ただし ENT_QUOTES が指定されている場合に限る
< (小なり) &lt;
> (大なり) &gt;

パラメータ

string

変換される文字列。

flags

以下のフラグを組み合わせたビットマスクです。 クォートや無効な符号単位シーケンス、そして文書型の扱いを指定します。 デフォルトは ENT_COMPAT | ENT_HTML401 です。

利用可能な flags 定数
定数名 説明
ENT_COMPAT ダブルクオートは変換しますがシングルクオートは変換しません。
ENT_QUOTES シングルクオートとダブルクオートを共に変換します。
ENT_NOQUOTES シングルクオートとダブルクオートは共に変換されません。
ENT_IGNORE 無効な符号単位シーケンスを含む文字列を渡したときに、 空の文字列を返すのではなく無効な部分を切り捨てるようになります。 このフラグは使わないようにしましょう。 » セキュリティの問題が発生する可能性があります
ENT_SUBSTITUTE 無効な符号単位シーケンスを含む文字列を渡したときに、 空の文字列を返すのではなく Unicode の置換文字に置き換えます。 UTF-8 の場合は U+FFFD、それ以外の場合は &#xFFFD; となります。
ENT_DISALLOWED 指定した文書型において無効な符号位置を、Unicode の代替文字である U+FFFD (UTF-8) あるいは &#xFFFD; で置き換えます。 これを設定しなければ、無効な符号位置をそのまま残します。 これは、外部コンテンツを埋め込んだ XML 文書を整形式に保つために有用です。
ENT_HTML401 コードを HTML 4.01 として処理します。
ENT_XML1 コードを XML 1 として処理します。
ENT_XHTML コードを XHTML として処理します。
ENT_HTML5 コードを HTML 5 として処理します。

encoding

オプションの引数。文字を変換するときに使うエンコーディングを定義します。

省略した場の encoding のデフォルト値は、varies PHP のバージョンによって異なります。 PHP 5.6.0 以降では、デフォルト値として default_charset の値を使います。PHP 5.4 と PHP 5.5 のデフォルト値は、 UTF-8 で、それより前のバージョンの PHP のデフォルト値は ISO-8859-1 でした。

技術的にはこの引数を省略可能ですが、PHP 5.5 以前のバージョンを使っている場合や、 default_charset の指定が入力とは違う文字セットになっている場合は、 適切な値を指定しておくことを強く推奨します。

この関数を使ううえでは ISO-8859-1ISO-8859-15UTF-8cp866cp1251cp1252 そして KOI8-R は事実上同等です。 string 自体がそのエンコーディングにおける有効な文字列である限り、 これらのエンコーディングでは htmlspecialchars() の影響が及ぶ文字がみな同じ位置にあるからです。

以下の文字セットをサポートします。

サポートする文字セット
文字セット エイリアス 説明
ISO-8859-1 ISO8859-1 西欧、Latin-1
ISO-8859-5 ISO8859-5 ほとんど使われないキリル文字セット (Latin/Cyrillic)。
ISO-8859-15 ISO8859-15 西欧、Latin-9 。Latin-1(ISO-8859-1) に欠けている ユーロ記号やフランス・フィンランドの文字を追加したもの。
UTF-8   ASCII 互換のマルチバイト 8 ビット Unicode 。
cp866 ibm866, 866 DOS 固有のキリル文字セット。
cp1251 Windows-1251, win-1251, 1251 Windows 固有のキリル文字セット。
cp1252 Windows-1252, 1252 西欧のための Windows 固有の文字セット。
KOI8-R koi8-ru, koi8r ロシア語。
BIG5 950 繁体字中国語。主に台湾で使用されます。
GB2312 936 簡体字中国語。国の標準文字セットです。
BIG5-HKSCS   Big5 に香港の拡張を含めたもの。繁体字中国語。
Shift_JIS SJIS, SJIS-win, cp932, 932 日本語。
EUC-JP EUCJP, eucJP-win 日本語。
MacRoman   Mac OS で使われる文字セット。
''   空文字列を指定すると、 スクリプトのエンコーディング (Zend multibyte)、 default_charset、 そして現在のロケール (nl_langinfo() および setlocale() を参照ください) の順でエンコーディングを検出します。 この方法はおすすめしません。

注意: これら以外の文字セットは理解できません。 かわりにデフォルトのエンコーディングを使用し、警告を発生させます。

double_encode

double_encode をオフにすると、PHP は既存の html エンティティをエンコードしません。 デフォルトでは、既存のエンティティも含めてすべてを変換します。

返り値

変換後の文字列を返します。

入力の string の中に エンコーディング encoding における無効なコードユニットシーケンスが含まれており、 かつ ENT_IGNORE フラグが設定されていなければ、 htmlspecialchars() は空文字列を返します。

変更履歴

バージョン 説明
5.6.0 encoding パラメータのデフォルト値が、 default_charset の設定値に変わりました。
5.4.0 encoding パラメータのデフォルトが UTF-8 に変わりました。
5.4.0 定数 ENT_SUBSTITUTEENT_DISALLOWEDENT_HTML401ENT_XML1ENT_XHTML および ENT_HTML5 が追加されました。
5.3.0 定数 ENT_IGNORE が追加されました。
5.2.3 double_encode パラメータが追加されました。

例1 htmlspecialchars() の例

<?php
$new 
htmlspecialchars("<a href='test'>Test</a>"ENT_QUOTES);
echo 
$new// &lt;a href=&#039;test&#039;&gt;Test&lt;/a&gt;
?>

注意

注意:

この関数は上記のあげたもの以外に関しては一切の変換を行わないことに注意してください。 すべての変換を行うには htmlentities() を参照してください。

注意:

In case of an ambiguous flags value, the following rules apply:

  • When neither of ENT_COMPAT, ENT_QUOTES, ENT_NOQUOTES is present, the default is ENT_COMPAT.
  • When more than one of ENT_COMPAT, ENT_QUOTES, ENT_NOQUOTES is present, ENT_QUOTES takes the highest precedence, followed by ENT_COMPAT.
  • When neither of ENT_HTML401, ENT_HTML5, ENT_XHTML, ENT_XML1 is present, the default is ENT_HTML401.
  • When more than one of ENT_HTML401, ENT_HTML5, ENT_XHTML, ENT_XML1 is present, ENT_HTML5 takes the highest precedence, followed by ENT_XHTML, ENT_HTML401.
  • When more than one of ENT_DISALLOWED, ENT_IGNORE, ENT_SUBSTITUTE are present, ENT_IGNORE takes the highest precedence, followed by ENT_SUBSTITUTE.

参考

add a note add a note

User Contributed Notes 18 notes

up
71
Dave
6 years ago
As of PHP 5.4 they changed default encoding from "ISO-8859-1" to "UTF-8". So if you get null from htmlspecialchars or htmlentities

where you have only set
<?php
echo htmlspecialchars($string);
echo
htmlentities($string);
?>

you can fix it by
<?php
echo htmlspecialchars($string, ENT_COMPAT,'ISO-8859-1', true);
echo
htmlentities($string, ENT_COMPAT,'ISO-8859-1', true);
?>

On linux you can find the scripts you need to fix by

grep -Rl "htmlspecialchars\\|htmlentities" /path/to/php/scripts/
up
49
Mike Robinson
6 years ago
Unfortunately, as far as I can tell, the PHP devs did not provide ANY way to set the default encoding used by htmlspecialchars() or htmlentities(), even though they changed the default encoding in PHP 5.4 (*golf clap for PHP devs*). To save someone the time of trying it, this does not work:

<?php
ini_set
('default_charset', $charset); // doesn't work.
?>

Unfortunately, the only way to not have to explicitly provide the second and third parameter every single time this function is called (which gets extremely tedious) is to write your own function as a wrapper:

<?php
define
('CHARSET', 'ISO-8859-1');
define('REPLACE_FLAGS', ENT_COMPAT | ENT_XHTML);

function
html($string) {
    return
htmlspecialchars($string, REPLACE_FLAGS, CHARSET);
}

echo
html("ñ"); // works
?>

You can do the same for htmlentities()
up
19
Thomasvdbulk at gmail dot com
8 years ago
i searched for a while for a script, that could see the difference between an html tag and just < and > placed in the text,
the reason is that i recieve text from a database,
wich is inserted by an html form, and contains text and html tags,
the text can contain < and >, so does the tags,
with htmlspecialchars you can validate your text to XHTML,
but you'll also change the tags, like <b> to &lt;b&gt;,
so i needed a script that could see the difference between those two...
but i couldn't find one so i made my own one,
i havent fully tested it, but the parts i tested worked perfect!
just for people that were searching for something like this,
it may looks big, could be done easier, but it works for me, so im happy.

<?php
function fixtags($text){
$text = htmlspecialchars($text);
$text = preg_replace("/=/", "=\"\"", $text);
$text = preg_replace("/&quot;/", "&quot;\"", $text);
$tags = "/&lt;(\/|)(\w*)(\ |)(\w*)([\\\=]*)(?|(\")\"&quot;\"|)(?|(.*)?&quot;(\")|)([\ ]?)(\/|)&gt;/i";
$replacement = "<$1$2$3$4$5$6$7$8$9$10>";
$text = preg_replace($tags, $replacement, $text);
$text = preg_replace("/=\"\"/", "=", $text);
return
$text;
}
?>

an example:

<?php
$string
= "
this is smaller < than this<br />
this is greater > than this<br />
this is the same = as this<br />
<a href=\"http://www.example.com/example.php?test=test\">This is a link</a><br />
<b>Bold</b> <i>italic</i> etc..."
;
echo
fixtags($string);
?>

will echo:
this is smaller &lt; than this<br />
this is greater &gt; than this<br />
this is the same = as this<br />
<a href="http://www.example.com/example.php?test=test">This is a link</a><br />
<b>Bold</b> <i>italic</i> etc...

I hope its helpfull!!
up
8
Kenneth Kin Lum
11 years ago
if your goal is just to protect your page from Cross Site Scripting (XSS) attack, or just to show HTML tags on a web page (showing <body> on the page, for example), then using htmlspecialchars() is good enough and better than using htmlentities().  A minor point is htmlspecialchars() is faster than htmlentities().  A more important point is, when we use  htmlspecialchars($s) in our code, it is automatically compatible with UTF-8 string.  Otherwise, if we use htmlentities($s), and there happens to be foreign characters in the string $s in UTF-8 encoding, then htmlentities() is going to mess it up, as it modifies the byte 0x80 to 0xFF in the string to entities like &eacute;.  (unless you specifically provide a second argument and a third argument to htmlentities(), with the third argument being "UTF-8").

The reason htmlspecialchars($s) already works with UTF-8 string is that, it changes bytes that are in the range 0x00 to 0x7F to &lt; etc, while leaving bytes in the range 0x80 to 0xFF unchanged.  We may wonder whether htmlspecialchars() may accidentally change any byte in a 2 to 4 byte UTF-8 character to &lt; etc.  The answer is, it won't.  When a UTF-8 character is 2 to 4 bytes long, all the bytes in this character is in the 0x80 to 0xFF range. None can be in the 0x00 to 0x7F range.  When a UTF-8 character is 1 byte long, it is just the same as ASCII, which is 7 bit, from 0x00 to 0x7F.  As a result, when a UTF-8 character is 1 byte long, htmlspecialchars($s) will do its job, and when the UTF-8 character is 2 to 4 bytes long, htmlspecialchars($s) will just pass those bytes unchanged.  So htmlspecialchars($s) will do the same job no matter whether $s is in ASCII, ISO-8859-1 (Latin-1), or UTF-8.
up
6
ryan at ryano dot net
18 years ago
Actually, if you're using >= 4.0.5, this should theoretically be quicker (less overhead anyway):

$text = str_replace(array("&gt;", "&lt;", "&quot;", "&amp;"), array(">", "<", "\"", "&"), $text);
up
9
ivan at lutrov dot com
8 years ago
Be careful, the "charset" argument IS case sensitive. This is counter-intuitive and serves no practical purpose because the HTML spec actually has the opposite.
up
5
php dot net at orakio dot net
11 years ago
I was recently exploring some code when I saw this being used to make data safe for "SQL".

This function should not be used to make data SQL safe (although to prevent phishing it is perfectly good).

Here is an example of how NOT to use this function:

<?php
$username
= htmlspecialchars(trim("$_POST[username]"));

$uniqueuser = $realm_db->query("SELECT `login` FROM `accounts` WHERE `login` = '$username'");
?>

(Only other check on $_POST['username'] is to make sure it isn't empty which it is after trim on a white space only name)

The problem here is that it is left to default which allows single quote marks which are used in the sql query. Turning on magic quotes might fix it but you should not rely on magic quotes, in fact you should never use it and fix the code instead. There are also problems with \ not being escaped. Even if magic quotes were used there would be the problem of allowing usernames longer than the limit and having some really weird usernames given they are to be used outside of html, this just provide a front end for registering to another system using mysql. Of course using it on the output wouldn;t cause that problem.

Another way to make something of a fix would be to use ENT_QUOTE or do:

<?php
$uniqueuser
= $realm_db->query('SELECT `login` FROM `accounts` WHERE `login` = "'.$username.'";');
?>

Eitherway none of these solutions are good practice and are not entirely unflawed. This function should simply never be used in such a fashion.

I hope this will prevent newbies using this function incorrectly (as they apparently do).
up
6
minder at ufive dot unibe dot ch
6 years ago
Problem

In many PHP legacy products the function htmlspecialchars($string) is used to convert characters like < and > and quotes a.s.o to HTML-entities. That avoids the interpretation of HTML Tags and asymmetric quote situations.

Since PHP 5.4 for $string in htmlspecialchars($string) utf8 characters are expected if no charset is defined explicitly as third parameter in the function. Legacy products are mostly in Latin1 (alias iso-8859-1) what makes the functions htmlspecialchars(), htmlentites() and html_entity_decode() to return empty strings if a special character, e. g. a German Umlaut, is present in $string:

PHP<5.4

echo htmlspecialchars('<b>Woermann</b>') //Output: &lt;b&gt;Woermann&lt;b&gt;
echo htmlspecialchars('Wörmann') //Output: &lt;b&gt;Wörmann&lt;b&gt;

PHP=5.4

echo htmlspecialchars('<b>Woermann</b>') //Output: &lt;b&gt;Woermann&lt;b&gt;
echo htmlspecialchars('<b>Wörmann</b>') //Output: empty

Three alternative solutions

a) Not runnig legacy products on PHP 5.4
b) Change all find spots in your code from
htmlspecialchars($string) and *** to
htmlspecialchars($string, ENT_COMPAT | ENT_HTML401, 'ISO-8859-1')
c) Replace all htmlspecialchars() and *** with a new self-made function

*** The same is true for htmlentities() and html_entity_decode();

Solution c

1 Make Search and Replace in the concerned legacy project:
Search for:        htmlspecialchars
Replace with:   htmlXspecialchars
Search for:        htmlentities
Replace with:   htmlXentities
Search for:        html_entity_decode
Replace with:   htmlX_entity_decode
2a Copy and paste the following three functions into an existing already everywhere included PHP-file in your legacy project. (of course that PHP-file must be included only once per request, otherwise you will get a Redeclare Function Fatal Error).

function htmlXspecialchars($string, $ent=ENT_COMPAT, $charset='ISO-8859-1') {
return htmlspecialchars($string, $ent, $charset);
}

function htmlXentities($string, $ent=ENT_COMPAT, $charset='ISO-8859-1') {
return htmlentities($string, $ent, $charset);
}

function htmlX_entity_decode($string, $ent=ENT_COMPAT, $charset='ISO-8859-1') {
return html_entity_decode($string, $ent, $charset);
}

or 2b crate a new PHP-file containing the three functions mentioned above, let's say, z. B. htmlXfunctions.inc.php and include it on the first line of every PHP-file in your legacy product like this: require_once('htmlXfunctions.inc.php').
up
6
Anonymous
10 years ago
Just a few notes on how one can use htmlspecialchars() and htmlentities() to filter user input on forms for later display and/or database storage...

1. Use htmlspecialchars() to filter text input values for html input tags.  i.e.,

echo '<input name=userdata type=text value="'.htmlspecialchars($data).'" />';


2. Use htmlentities() to filter the same data values for most other kinds of html tags, i.e.,

echo '<p>'.htmlentities($data).'</p>';

3. Use your database escape string function to filter the data for database updates & insertions, for instance, using postgresql,

pg_query($connection,"UPDATE datatable SET datavalue='".pg_escape_string($data)."'");


This strategy seems to work well and consistently, without restricting anything the user might like to type and display, while still providing a good deal of protection against a wide variety of html and database escape sequence injections, which might otherwise be introduced through deliberate and/or accidental input of such character sequences by users submitting their input data via html forms.
up
5
solar-energy
12 years ago
also see function "urlencode()", useful for passing text with ampersand and other special chars through url

(i.e. the text is encoded as if sent from form using GET method)

e.g.

<?php
echo "<a href='foo.php?text=".urlencode("foo?&bar!")."'>link</a>";
?>

produces

<a href='foo.php?text=foo%3F%26bar%21'>link</a>

and if the link is followed, the $_GET["text"] in foo.php will contain "foo?&bar!"
up
2
Felix D.
5 years ago
Another thing important to mention is that
htmlspecialchars(NULL)
returnes an empty string and not NULL!
up
3
Anonymous
10 years ago
This may seem obvious, but it caused me some frustration. If you try and use htmlspecialchars with the $charset argument set and the string you run it on is not actually the same charset you specify, you get any empty string returned without any notice/warning/error.

<?php

$ok_utf8
= "A valid UTF-8 string";
$bad_utf8 = "An invalid UTF-8 string";

var_dump(htmlspecialchars($bad_utf8, ENT_NOQUOTES, 'UTF-8'));  // string(0) ""

var_dump(htmlspecialchars($ok_utf8, ENT_NOQUOTES, 'UTF-8'));  // string(20) "A valid UTF-8 string"

?>

So make sure your charsets are consistent

<?php

$bad_utf8
= "An invalid UTF-8 string";

// make sure it's really UTF-8
$bad_utf8 = mb_convert_encoding($bad_utf8, 'UTF-8', mb_detect_encoding($bad_utf8));

var_dump(htmlspecialchars($bad_utf8, ENT_NOQUOTES, 'UTF-8'));  // string(23) "An invalid UTF-8 string"

?>

I had this problem because a Mac user was submitting posts copy/pasted from a program and it contained weird chars in it.
up
3
PoV
4 years ago
Be aware of the encoding of your source files!!!

Some of the suggestions here make reference to workarounds where you hard-code an encoding.

<?php
 
echo htmlspecialchars('<b>Wörmann</b>');  // Why isn't this working?
?>

As it turns out, it may actually be your text editor that is to blame.

As of PHP 5.4, htmlspecialchars now defaults to the UTF-8 encoding. That said, many text editors default to non-UTF encodings like ISO-8859-1 (i.e. Latin-1) or WIN-1252. If you change the encoding of the file to UTF-8, the code above will now work (i.e. the ö is encoded differently in UTF-8 and ISO-8859-1, and you need the UTF-8 version).

Make sure you are editing in UTF-8 Unicode mode! Check your UI or manual for how to convert files to Unicode. It's also a good idea to figure out where to look in your UI to see what the current file encoding is.
up
2
Anonymous
14 years ago
function htmlspecialchars_array($arr = array()) {
   $rs =  array();
   while(list($key,$val) = each($arr)) {
       if(is_array($val)) {
           $rs[$key] = htmlspecialchars_array($val);
       }
       else {
           $rs[$key] = htmlspecialchars($val, ENT_QUOTES);
       }   
   }
   return $rs;
}
up
1
qshing1437 at hotmail dot com
3 months ago
If you use htmlspecialchars() to escape any HTML attribute, make sure use double quote instead of single quote for the attribute.

For Example,

> Wrap with Single Quote
<?php
echo "<p title='"  . htmlspecialchars("Hello\"s\'world") . "'">

// title will end up Hello"s\ and rest of the text after single quote will be cut off.
?>

> Wrap with Double quote :
<?php
echo '<p title="'  . htmlspecialchars("Hello\"s\'world") . '"'>

// title will show up correctly as Hello"s'world
?>
up
1
_____ at luukku dot com
17 years ago
People, don't use ereg_replace for the most simple string replacing operations (replacing constant string with another).
Use str_replace.
up
0
support at playnext dot ru
6 years ago
For those having problems after the change of default value of $encoding argument to UTF-8 since PHP 5.4.

If your old non-UTF8 projects ruined - pls consider:
1. http://php.net/manual/en/function.override-function.php
2. http://php.net/manual/ru/function.runkit-function-redefine.php

The idea - you override the built-in htmlspecialchars() function with your customized variant which is able to respect non UTF-8 default encoding. This small piece of code can be then easily inserted somewhere at the start of yout project. No need to rewrite all htmlspecialchars() entries globally.

I've spent several hours with both approaches. Variant 1 looks good especaially in combination with http://www.php.net/manual/en/function.rename-function.php as it allows to call original htmlspecialchars() with just altered default args. The code could be as follows:

<?php
rename_function
('htmlspecialchars', 'renamed_htmlspecialchars');
function
overriden_htmlspecialchars($string, $flags=NULL, $encoding='cp1251', $double_encode=true) {
   
$flags = $flags ? $flags : (ENT_COMPAT|ENT_HTML401);
    return
renamed_htmlspecialchars($string, $flags, $encoding, $double_encode);
}
override_function('htmlspecialchars', '$string, $flags, $encoding, $double_encode', 'return overriden_htmlspecialchars($string, $flags, $encoding, $double_encode);');
?>

Unfortunatelly this didn't work for me properly - my site managed to call overriden function but not every time I reloaded the pages. Moreover other PHP sites crashed under my Apache server as they suddenly started blaming htmlspecialchars() was not defined. I suppose I had to spend more time to make it work thread/request/site/whatever-safe.

So I switched to runkit (variant 2). It worked for me, although even after trying runkit_function_rename()+runkit_function_add() I didn't managed to recall original htmlspecialchars() function. So as a quick solution I decided to call htmlentities() instead:

<?php
function overriden_htmlspecialchars($string, $flags=NULL, $encoding='UTF-8', $double_encode=true) {
   
$flags = $flags ? $flags : (ENT_COMPAT|ENT_HTML401);
   
$encoding = $encoding ? $encoding : 'cp1251';
    return
htmlentities($string, $flags, $encoding, $double_encode);
}
runkit_function_redefine('htmlspecialchars', '$string, $flags, $encoding, $double_encode', 'return overriden_htmlspecialchars($string, $flags, $encoding, $double_encode);');
?>

You may be able to implement your more powerfull overriden function.
Good luck!
up
0
nachitox2000 [at] hotmail [dot] com
9 years ago
I had problems with spanish special characters. So i think in using htmlspecialchars but my strings also contain HTML.
So I used this :) Hope it help

<?php
function htmlspanishchars($str)
{
    return
str_replace(array("&lt;", "&gt;"), array("<", ">"), htmlspecialchars($str, ENT_NOQUOTES, "UTF-8"));
}
?>
To Top