You will learn how to write PHP code to remotely consume services like Google Maps, Flickr, Amazon, YouTube, MSN Search, Yahoo!, Last.fm, and the Internet UPC Database, not to mention the California Highway Patrol Traffic data! You will also learn about the technologies, data formats, and protocols needed to use these web services and APIs, and some of the freely-available PHP tools for working with them.
You will understand how these technologies work with each other and see how to use this information, in combination with your imagination, to build your own cutting-edge websites.
Chapter 1 provides an overview of "mashups": what a mashup is, and why you would want one.
In Chapter 2 we create a basic mashup, and go shopping. We will simply look up products on Amazon.com based on the Universal Product Code (UPC). To do this, we cover two basic web services to get our feet wet — XML-RPC and REST. The Internet UPC Database is an XML-RPC-based service. while Amazon uses REST. We will create code to call XML-RPC and REST services. Using PHP’s SAX function, we create an extensible object-oriented parser for XML. The mashup covered in this chapter integrates information taken from Amazon's e-commerce service (ECS) with the Internet UPC database.
In Chapter 3, we create a custom search engine using the technology of MSN, and Yahoo! The chapter starts with an introduction to SOAP, the most complex of the web service protocols. SOAP relies heavily on other standards like WSDL and XSD, which are also covered in readable detail. We take a look at a WSDL document and learn how to figure out what web services are available from it, and what types of data are passed. Using PHP5’s SoapClient extension, we then interact with SOAP servers to grab data. We then finally create our mashup, which gathers web search results sourced from MSN and Yahoo!
For the mashup in Chapter 4, we use the API from the video repository site YouTube, and the XML feeds from social music site Last.fm. We will take a look at three different XML-based file formats from those two sites: XSPF for song playlists, RSS for publishing frequently updated information, and YouTube’s custom XML format. We will create a mashup that takes the songs in two Last.fm RSS feeds and queries YouTube to retrieve videos for those songs. Rather than creating our own XML-based parsers to parse the three formats, we have used parsers from PEAR, one for each of the three formats. Using these PEAR packages, we create an object-oriented abstraction of these formats, which can be consumed by our mashup application.
In Chapter 5, we screen-scrape from the California Highway Patrol. The CHP maintains a website of traffic incidents. This site auto-refreshes every minute, ensuring the user gets live data about accidents throughout the state of California. This is very valuable if you are in front of a computer. If you are out and about running errands, it would be fairly useless. However, our mashup will use the web service from 411Sync.com to accept SMS messages from mobile users to deliver these traffic incidents to users.
We've thrown almost everything into Chapter 6! In this chapter, we use RDF documents SPARQL, RAP, Google Maps, Flickr, AJAX, and JSON. We create a geographically-centric way to present pictures from Flickr on Google Maps. We see how to read RDF documents and how to extract data from them using SPARQL and RAP for RDF. This gets us the latitude and longitude of London Tube Stations. We display them on a Google Map, and retrieve pictures of a selected station from Flickr. Our application needs to communicate with the API servers for which we use AJAX and JSON, which is emerging as a major data format. The biggest pitfall in this AJAX application is race conditions, and we will learn various techniques to overcome these.