Filtering out the un-needed

Frederick

I am using cURL to display the tracking information of FedEx shipments within a project manager page. I need to know how I can remove all of the unwanted content from the results and just display the tracking. The unwanted are items such as images, their nav menus, useless crap, etc.

Thanks for the help in advance.

zabmilenko

Can you post an example of the code you use to get the html to begin with?

Frederick

<?php
$tracking_number = "1188889542";
$ch = curl_init("http://www.fedexfreight.fedex.com/trackmain.do?as_type=PRO&as_pro=".$tracking_number);
curl_setopt($ch, CURLOPT_HEADER, 0);
$results = curl_exec($ch);
curl_close($ch);

echo $results;
?>

zabmilenko

Ahhh. Looks like you have a bunch of html source...

Most examples I have use preg_match to pull blocks of html out of a larger html document. For an online game I used to code for we pulled our ranking out of the popular game sites and gave bonuses to the players from that info.

What you need to know is a good chunk of html before and after each data you want to snatch out of the page, and then build a perl-compatible regular expression from that.

For example:

<h3>Expected Arrival Date</h3>
<span>22-Nov-2006</span>

Your preg would be similar to:

"/Expected Arrival Date<\/h3>\n<span>(date stuff here)>\/span>/iU"

But the above expression won't do it. Most sites are hip to this and will catch on. I highly recommend setting a fake user agent string in cURL to send so they won't notice what you are doing. Note also that you need to change your PCRE if they change their page too much.

Frederick

I figured as much, was hoping on an easier way of filtering because of the mass amount of HTML to remove just so I can get 8 lines out of it.

Thank for your help.

Weedpacket

I'd've thought Federal Express would provide a machine-friendly document somewhere for exactly this purpose.

Frederick

They have a client pro area for tracking your orders that comes with your accounts. Extra charge item.