Special Characters Problem (weird!)

[kurbsdude]

Ok here's my problem,

I need to read a text directly from a website ( suppose a certain div value)... the problem is that this value can sometimes be unicode characters (different languages) and sometimes umlauts...

 

now the problem is how to correctly display these characters... if i use htmlentities() unicode displays fine but umlauts are now turned into ����.. and if i use utf8 encode, umlauts display fine while unicode characters are turned into ����...

 

any suggestions?

cheers!!

 

[Mark Baker]

If the website that you're reading isn't utf-8, then you'll need to convert it from whatever charset it does use to utf-8

[kurbsdude]

If the website that you're reading isn't utf-8, then you'll need to convert it from whatever charset it does use to utf-8

 

that sounds logical, but is there a way to convert to utf-8 from php itself without changing the website?

[Mark Baker]

function getCharSetFromMetaTags($strBody) {
$returns = array ( "/\n/", "/\r/", "/\t/", "/\s+/" );
$nullReturns = array( ' ', ' ', ' ', ' ' );

    preg_match("/<meta\s?[^>]*content\s?=\s?\".*charset\s?=\s?(.*)\"\s?\/?>/Ui", preg_replace($returns,$nullReturns,$strBody), $strCharSet);

    return (isset($strCharSet[1])) ? strtoupper($strCharSet[1]) : false;
}	//	function getCharSetFromMetaTags()

 

Read $content from remote website

$charset = getCharSetFromMetaTags($content);
if ($charset === false) {
   // no idea what charset the remote site is using
} elseif ($charset != 'utf-8') {
   iconv($charset,'UTF-8',$content);
}

$content is now utf-8 (or unknown)

 

Do with it as thou wilt

[kurbsdude]

thanks i'll check that out