More than 4 years ago I developed the first library to parse WURFL. Initially it was only a PHP file that parsed the XML and created an associative array.
Later I added a class to also search a device given a user-agent and get capability values for the user-agent. With the growth of WURFL, caching was needed to try to speed up the search and retrival of data.
My WURFL PHP library has a very minimal set of requirements, PHP and XML support compiled in. This is something that I never wanted to change during these years to make sure that ANYONE could download and use it.
I am happy to announce that more development has been done on top of my original library and that Steve Kamerman, the author, has achieved great performance results with his new library.
The new library uses MySQL, so a new requirement is added. For this reason I will keep my library as-is, but still, if you have a database available and want some good performances, then I suggest you take a look at Tera-WURFL.
It’s worth visiting the links presented below to see it in action.
There is a test page that lets you see how user-agents are matched and some speed test.
Also, since Tera-WURFL uses a database you will need to update it when a new XML comes. Steve provides a nice admin panel to manage the script and data (don’t mess up with it too much!).
So, once again, congratulations to Steve for his huge work and great new library.
Andrea Trasatti's tech notes and more 