php function to extract link from string - php

Question

i want to extract href link from text or string. i write a little function to do that but this is slow when a string to transform is large. My code is
function spy_linkIntoString_Format($text) {
global $inc_lang; $lang = $inc_lang['tlang_media'];
$it = explode(' ' ,$text);
$result = '';
foreach($it as $jt) {
$a = trim($jt);
if(preg_match('/((?:[\w\d]+\:\/\/)?(?:[\w\-\d]+\.)+[\w\-\d]+(?:\/[\w\-\d]+)*(?:\/|\.[\w\-\d]+)?(?:\?[\w\-\d]+\=[\w\-\d]+\&?)?(?:\#[\w\-\d]*)?)/', $jt)) {
$pros_lis = str_replace('www.','',$jt);
$pros_lis = (strpos($pros_lis, 'http://') === false ? 'http://'. $pros_lis : $pros_lis);
$urlregx = parse_url($pros_lis);
$host_name = (!empty($urlregx['host']) ? $urlregx['host'] : '.com');
if($host_name == 'youtube.com') {
$string_v = $urlregx['query']; parse_str($string_v, $outs); $stID = $outs['v'];
$result .= '<a title="Youtube video" coplay="'.$stID.'" cotype="1" class="media_spy_vr5" href="#"><span class="link_media"></span>'.$lang['vtype_youtube'].'</a> ';
} elseif($host_name == 'vimeo.com') {
$path_s = $urlregx['path']; $patplode = explode("/", $path_s); $stID = $patplode[1];
$result .= '<a title="Vimeo video" coplay="'.$stID.'" cotype="2" class="media_spy_vr5" href="#"><span class="link_media"></span>'.$lang['vtype_vimeo'].'</a> ';
} elseif($host_name == 'travspy.com') {
$result .= '</span>'.$pros_lis.' ';
} else {
$result .= '<span class="jkt_445 c8_big_corner"></span>'.$pros_lis.' ';
}
} else {
$result .= $jt.' ';
}
}
return trim($result);/**/
}
Can i do this run fast?

Answer 1

You should rewrite this to use preg_match_allinstead of splitting the text into words (i.e. drop the explode).
$regex = '/\b((?:[\w\d]+\:\/\/)?(?:[\w\-\d]+\.)+[\w\-\d]+(?:\/[\w\-\d]+)*(?:\/|\.[\w\-\d]+)?(?:\?[\w\-\d]+\=[\w\-\d]+\&?)?(?:\#[\w\-\d]*)?)\b/';
preg_match_all($regex, $text, $matches, PREG_SET_ORDER);
foreach ($matches as $match) {
$url = $match[0];
// your link generator
}

Answer 2

You seem to be breaking the text into words separated by spaces, and then matching each word against a regular expression. This is very time consuming indeed.
A faster way to do this is to perform the regular expression search on the entire text and then iterate over it's results.
preg_match_all($regex, $text, $result, PREG_PATTERN_ORDER);
foreach($result[0] as $jt){
//do what you normally do with $jt
}