Tracking The Packages - Page 26

Tracking Federal Express Packages

The FedEx tracking application works great for tracking packages on the Web. It provides detailed information on where your package is and where it's been. Federal Express also provides an API for interacting with the FedEx tracking system. A Perl module is available for this API, but the module included many other features that we don't need and it must run on a win32 system. Also, FedEx requires you to register a FedEx number and fill out a form, accepting a number of licensing terms including the restriction that the API could not be used in an application that provided information about another carrier.

Instead of using the API, I decided that it would be simpler to write a program using the LWP libraries that submits the tracking number to the FedEx Web site and parses the resulting HTML page for the results. It did take some time to construct the regular expression, but I was able to tune it so that the code is minimal. It is possible that the script will break in the future if and when FedEx changes the interface to the tracking pages. For our purposes, the solution is short and sweet. Perhaps in the future, FedEx will relax its terms and the author of the FedEx module will include support for Unix in addition to win32 systems.

I created the &parse_fedex_html subroutine on lines 60 through 89 for this purpose. Line 62 builds the URL to get the tracking information from the FedEx Web site. Line 63 calls the get() method of the LWP::Simple module, which sends the URL, retrieves the data, and assigns the results to the $page variable. If the tracking number was invalid for some reason, the subroutine will return a 0 on line 65. Otherwise, we search the $page variable on line 69 for the string CLASS="resultstableheader" followed by a </tr> tag and some text followed by a </table> tag. We actually want to grab the text between the </tr> and </table> tags because it contains the tracking information we're looking for. This text is assigned to the $results variable on line 70.

Now that we have the block of text that contains the tracking information, we have to break it up a bit more to get the details. The regular expression on line 75 grabs the text between a <tr> and </tr> tag and assigns it to the $record variable on line 76. Line 79 contains a more complex regular expression that's actually made up of other regular expressions on lines 72 and 73. Line 72 is matching text within a set of <td> and </td> tags. Line 73 builds a more complex regular expression based on the one we build on line 72. It contains three instances of the previous regular expression meaning that we're looking for three separate <td> and </td> tag pairs. These matches will contain the tracking location, date and time, and any notes related to the package. As a result of the match on line 79, we set tracking variables on line 80-82. This information is then passed back as an array on line 88, where we'll print it to the user's browser.

Tracking UPS Packages

Tracking a package with UPS is a bit simpler because we are using the Business::UPS module. The &parse_ups_data subroutine on lines 90 through 100 grabs the tracking information for a given UPS tracking number. The UPStrack function on line 93 retrieves the tracking information for a package and returns the results in the form of a hash. The value that we're looking for is the last location that the package was tracked. This information is contained in the Last Scanned at hash key value. This value is assigned on line 96 and returned to the caller on line 99. Like I said, tracking packages with the UPS module is a bit easier. Don't rely on the documentation that comes with the module however. The module itself needs to be updated to handle the latest return values from the UPS server. The module would have been able to return more data if all of the features were working.

Displaying the Form and Results

Now that we've covered the major sections of the script, let's talk about the main body on lines 7 to 39. Lines 7 to 13 initialize all of the variables that we use in the application. Line 15 prints the HTTP header; line 16 prints the top portion of the HTML results. On line 18, we check for a value in the $order_num variable, which was set on line 11. If it contains a value, it means that the user entered an order number and pressed the submit button. On line 19, we look up the corresponding tracking number with the &get_tracking_number subroutine. If the routine returns a tracking number, we grab the tracking information on line 24 or 26 with the &parse_fedex_html or &parse_usp_data depending on whether the $track_type variable was set to FedEx or UPS by &get_tracking_number. If we didn't get any errors when retrieving the tracking information, the results are printed to the browser on lines 32 and 33. Lastly, we print the HTML form by calling the &print_form subroutine on line 38. This routine simply contains the HTML form into which the user will enter their order number. The script ends on line 39 where we print </body></html> to end the HTML output.

The resulting script will print an order number field. When an order number is entered and the user presses the submit button, the script attempts to lookup the tracking number and retrieve the tracking information, which is printed back to the browser. The same form is also printed again with the order number pre- filled to make it easy for users to continue to track the package as it moves from one delivery point to another.

Conclusion

You can use this script out of the box to enable your customers to track their packages without going off your Web site or having to keep track of a separate order and tracking number. You'll probably want to modify the script to integrate your site's design elements since the output of this script is minimal. As I hope you've learned with this script, Perl is a powerful text processing language. I could go on to integrate other carrier tracking systems so that the script had better coverage. Another feature that might be useful is the ability to get the price of a package that will be shipped. Some e-commerce packages already include this functionality, but others don't. This feature can be added using the same techniques for the FedEx Web site and with the Business::UPS module. If you'd like to see this feature added, write to me at eisen@ferrumgroup.com.

Tracking FedEx and UPS Packages Online - Page 25
Weaving Magic With Regular Expressions