http://www.externaldomain.com/?q=mydomain.com

Your code treats it as an INTERNAL link since your logic searches for ‘your domain’ anywhere in the href value (since you use the ‘asterisk equals’ operator). To be more robust you would have to find links that start with it by using ‘carrot equals’ operator:

// needs to be repeated for ‘https://’
$(“a[href^='http://']:not([href^='http://"+location.hostname+"']));

The code above works almost 100% of the time, but the domain for many sites includes the ‘www.’ prefix and some don’t. So if a link in your body was: “http://www.mydomain.com/page.html” but the address bar of your browser read: “http://mydomain.com”, then it would not match my jQuery above. So we need to be even more precise:

// again, needs to be repeated for ‘https://’
$(“a[href^='http://']:not([href^='http://"+location.hostname+"']:not([href^='http://www."+location.hostname+"']));

Personally, this is getting a bit too long, so I preferred to use regular expressions:

// for every link
$(“a”).each(function () {
var href = $(this).attr(“href”);

// remove the ‘www.’ prefix from the href value (if it exists)
var domain = location.hostname.replace(/^(www\.)?/, “”);

// set up a regular expression that sees if a string starts with http[s]://[www.]mydomain.com
var re = new RegExp(“^https?:\\/\\/(www.)?” + domain);

// guaranteed to be an external link
if (!re.test(href)) {
$(this).attr(“target”, “_blank”).attr(“title”, “Opens a new window”);
}
});