[This screencast is] part 2 in a series on building a JSON end-point. We're using simple RESTful verbage to grab a list of names from a session store.

You'll need to have the base scripts created in part one to follow along effectively. He picks up right where the previous part ends, creating a new view that uses some simple javascript (jQuery) code to pull the "names" values from his simple JSON endpoint. Also included is an example of a form for adding a new name to the list. The full source for the example can be found over on github.

You'll need some of the base that he created in this previous screencast to follow along (the basic structure, really). His takes his basic "hello world" application and builds on it to add a "model" to pull name data from and has the "/names" action respond with a JSON-formatted message (and an "application/json" content-type). His "model" pulls the data out with findAll() and find() methods.

You can grab the source to follow along.

This article uses the pecl_oauth extension and builds on Rasmus' OAuth Provider post. [...] OAuth has a little more baggage with it than just passing a username and password to an API.

She lists the five things you'll need for your service and talks a bit about the registration process and how the consumer key/consumer secret keys are generated. There's no strict definition on them, so her example uses a combination of sha1, mt_rand and substr to get the job done. She also includes a sample "consumers" table for your reference.

However, unit testing models that talk to web services presents us with a different set of problem. Usually our application runs requests against something that is not in our control. Unless the service provider gives us some kind of test mode, we're operating with dynamic live data which shouldn't be used for testing, since it's harder to test in the first place, but also could be harmful to be manipulated.

They present two method you can use to test the models: creating mock objects to handle and return correctly formatted data and intercepting the HTTP request to either send it to a Zend_HTTP adapter or a different endpoint (service location).

A common reason to use Ajax in a website is to make it feel faster, so you usually want Ajax requests be processed as quickly as possible. While there are many ways to speed up Zend Framework based applications, there are still some things like routing and dispatching which still add up to the total. There is, however, another way to make Ajax work even faster

Since Ajax requests don't usually require all of the complex routing and dispatching that a normal Zend Framework request might, he recommends creating a separate Ajax handler. His includes an example of a simple searching endpoint with a class handling the backend logic. A simple handler script loads up the bootstrap and makes the request for the search, returning the search results in a JSON format.