Question

According to a function here, http://www.unicodetools.com/unicode/convert-to-html.php, the function is used to convert string to HTML encoded text.

The JavaScript is:

function a(b) {
    var c= '';

    for(i=0; i<b.length; i++) {
        if(b.charCodeAt(i)>127) {
            c += '&#' + b.charCodeAt(i) + ';'; 
        } else { 
            c += b.charAt(i); 
        }
  }

  document.forms.conversionForm.outputText.value = c;
}

And my try is:

function str_to_html_entity($str) {
    $output = NULL;

    for($i = 0; $i < strlen($str); $i++) {
        if(ord($str) > 127) {
            $output .= '&#' + ord($str) + ';'; 
        } else { 
            $output .= substr($str, $i); 
        }
  }

  return $output;
}

echo str_to_html_entity("Thére Àre sôme spëcial charâcters ïn thìs têxt");

My PHP function run correctly, but the result is not what I expected:

my result:

Thére Àre sôme spëcial charâcters ïn thìs têxthére Àre sôme spëcial charâcters ïn thìs têxtére Àre sôme spëcial charâcters ïn thìs têxt�re Àre sôme spëcial charâcters ïn thìs têxtre Àre sôme spëcial charâcters ïn thìs têxte Àre sôme spëcial charâcters ïn thìs têxt Àre sôme spëcial charâcters ïn thìs têxtÀre sôme spëcial charâcters ïn thìs têxt�re sôme spëcial charâcters ïn thìs têxtre sôme spëcial charâcters ïn thìs têxte sôme spëcial charâcters ïn thìs têxt sôme spëcial charâcters ïn thìs têxtsôme spëcial charâcters ïn thìs têxtôme spëcial charâcters ïn thìs têxt�me spëcial charâcters ïn thìs têxtme spëcial charâcters ïn thìs têxte spëcial charâcters ïn thìs têxt spëcial charâcters ïn thìs têxtspëcial charâcters ïn thìs têxtpëcial charâcters ïn thìs têxtëcial charâcters ïn thìs têxt�cial charâcters ïn thìs têxtcial charâcters ïn thìs têxtial charâcters ïn thìs têxtal charâcters ïn thìs têxtl charâcters ïn thìs têxt charâcters ïn thìs têxtcharâcters ïn thìs têxtharâcters ïn thìs têxtarâcters ïn thìs têxtrâcters ïn thìs têxtâcters ïn thìs têxt�cters ïn thìs têxtcters ïn thìs têxtters ïn thìs têxters ïn thìs têxtrs ïn thìs têxts ïn thìs têxt ïn thìs têxtïn thìs têxt�n thìs têxtn thìs têxt thìs têxtthìs têxthìs têxtìs têxt�s têxts têxt têxttêxtêxt�xtxtt

expected result:

Thére Àre sôme spëcial charâcters ïn thìs têxt

Could someone please advise what wrong with my PHP function?

Thanks

UPDATE

function str_to_html_entity($str) {
    $result = null;
    for ($i = 0, $length = mb_strlen($str, 'UTF-8'); $i < $length; $i++) {
        $character = mb_substr($str, $i, 1, 'UTF-8');
        if (strlen($character) > 1) {  // the character consists of more than 1 byte
            $character = htmlentities($character, ENT_COMPAT, 'UTF-8');
        }
        $result .= $character;
    }

  return $result;
}

echo str_to_html_entity("Thére Àre"); // Th&eacute;re &Agrave;re
echo str_to_html_entity("中"); // 中

Answer 1

Generally:

• Javascript strings are Unicode aware, which means str[0] will return one character, however long this character is. charCodeAt will correctly return character codes for any character.
• PHP strings are dumb binary arrays, in which a character may take up more than one byte. $str[0] and ord only work on single bytes and will therefore mangle any multi-byte characters. See What Every Programmer Absolutely, Positively Needs To Know About Encodings And Character Sets To Work With Text for an in-depth explanation of this.

Because of this, you can’t replicate the exact same algorithm in PHP. Also, in your loop, you’re using the whole $str instead of a string offset, which is your other primary problem. To make it Unicode aware, this is probably the nicest way:

$result = null;
foreach (preg_split('/./u', $str) as $character) {
    if (strlen($character) > 1) {  // the character consists of more than 1 byte
        $character = mb_convert_encoding($character, 'HTML-ENTITIES', 'UTF-8');
    }
    $result .= $character;
}

This expects the string to be UTF-8 encoded. As you can see though, there’s a nice function called mb_convert_encoding, which can escape a whole block of text in one go, which you’re essentially reinventing. Use it instead.

Alternative version for Unicode-impaired PCREs:

$result = null;
for ($i = 0, $length = mb_strlen($str, 'UTF-8'); $i < $length; $i++) {
    $character = mb_substr($str, $i, 1, 'UTF-8');
    if (strlen($character) > 1) {  // the character consists of more than 1 byte
        $character = mb_convert_encoding($character, 'HTML-ENTITIES', 'UTF-8');
    }
    $result .= $character;
}

But seriously, just use $str = mb_convert_encoding($str, 'HTML-ENTITIES', 'UTF-8') and be done with it. No looping required.