Hey all. I want to identify what URL the user came from in my code. I can do this via $_SERVER[‘HTTP_REFERER’]; but in this instance it’s blank for when a user clicked on our ad link. I appreciate how it isn’t the most reliable way (to rely on this given that browsers can switch off or spoof it) but I need it!
The story is a user is on an a site with a picture of an item, they click our ad. I want to know where on the site they were so I can grab a particular picture (by pulling a particular line of code that shows the image path). All of it works except the referrer value! I know the referrer code is correct by testing a direct link from other sources so it seems to be something to do with the ad (or my code).
The ad links to http://pagead2.googlesyndication.com/ which must then direct the user to our site. I presume therefore the referrer’s getting lost in translation but I wonder if there’s something else I can do with my code. Any help would be greatly appreciated, I’m stumped on this one!
$urlreferrer=$_SERVER['HTTP_REFERER'];
//print $_SERVER["HTTP_REFERER"];
$lines = file($urlreferrer);
// Loop through our array, show HTML source as HTML source; and line numbers too.
foreach ($lines as $line_num => $line)
{
//echo "Line #<b>{$line_num}</b> : " . htmlspecialchars($line) . "<br />\
";
if($line_num==62){$imageline=htmlspecialchars($line);
}
$imagepath=substr($imageline, 53,-13);
return $imageline;
Note - the referrer and all the code does work when I create a link to the page and click it (ignore that I’ve got no validation, I want to get it working first!). It doesn’t work when the user clicks on our ad and comes to the page. So I wonder whether there’s another way to grab the URL they came from?
I’m not sure referrer would work (even if you were getting it), as you have a middle man.
Your workflow is:
User Site -> Google Syndication -> Your Server
All of your requests are coming from the Google Syndication (if I understood your original post correctly). Your best bet is to have the User Site append some variable to the URL that indicates where it originated from (their user_id or a token that is unique to them).
Completely agree! In this instance we’re simulating/validating integration. Ideally the link to us is hardcoded and passes some relevant info across. For now though we’re running an ad instead (to validate is there enough of a demand to justify integration) and so am trying to see if we can still figure out a way to pull across some info. By getting the referrer I can crawl the content and pull an image across.