I am g开发者_如何学运维rabbing meta descriptions from remote urls and when a url contains "£" it returns as a black diamond with a question mark in firefox. When I apply utf8_encode on the string, it returns "£" as it should, however then other UTF characters will not display correctly. What can I do?
Did you set the content type meta tag on the page?
http://www.w3schools.com/tags/att_meta_http_equiv.asp
Edit:
To solve what you put in your comment, I might do something like this (super quick and dirty):
<?php
$ch = curl_init('http://www.dailymail.co.uk/health/article-1374575/Under-18s-sunbed-ban-cut-skin-cancer-toll.html');
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
$data = curl_exec($ch);
preg_match('/(<meta[^\n]+)(name=\"description\"[^\n])(content=\")([^\n]+)(\")/', $data, $m);
echo urlencode($m[4]);
?>
Having said that, the correct way of doing this would be to parse the HTML returned by the curl_exec
, find the appropriate node (with name="description"
) and return the urlencode
d value of the content
attribute
精彩评论