I\'m using cURL to pull the contents of a remote site. I need to check all \"href=\" attributes and determine if they\'re relative or absolute path, then get the value of the li
A combination of a regex* and HTML's parse_url() should help:
// find all links in a page used within href="" or href='' syntax
$links = array();
preg_match_all('/href=(?:(?:"([^"]+)")|(?:\'([^\']+)\'))/i', $page_contents, $links);
// iterate through each array and check if it's "absolute"
$urls = array();
foreach ($links as $link) {
$path = $link;
if ((substr($link, 0, 7) == 'http://') || (substr($link, 0, 8) == 'https://')) {
// the current link is an "absolute" URL - parse it to get just the path
$parsed = parse_url($link);
$path = $parsed['path'];
}
$urls[] = 'http://www.website.com/index.php?url=' . $path;
}
To determine if the URL is absolute or not, I simply have it check if the beginning of the URL is http://
or https://
; if your URLs contain other mediums such as ftp://
or tel:
, you might need to handle those as well.
This solution does use regex to parse HTML, which is often frowned upon. To circumvent, you could switch to using [DOMDocument][2]
, but there's no need for extra code if there aren't any issues.