rest - Is the Twitter API *really* RESTful? -
With more than half of the web developer community, I'm really fighting really and actually leaning towards the REST style I am More specifically, I am trying to make some opinions about how practically pure pure architecture really is between a web browser and an episode server.
As part of my learning endeavor, I see some online examples of rest in this case specifically on Twitter. In them, they discuss their various "REST API methods".
I am struggling with rationalizing that in fact, most of these are really lush, for example, for a simple Consider the GET request. In the net implementation of REST, I expect the same requests at that URL, even if the initial client, in this special case to return the same responses, although we clearly have different responses based on our current certified users Which means that before our responses are generated, our requests are being linked to any kind of client status on the server Is.
Hopefully provides enough context for my question - can I really say "REST"? I get the impression that 90% of the so-called lush implementation between web browser and application server displays this same inconsistency, where restrictions on customer status stored on the server are ignored.
Twitter breaks every rest of the constraints too much to return different results based on authenticated user It is not really related to idempotency, it is about being able to do the same request several times about the idepotency and it has only one effect. Even Twitter has respected Edempotency. If I receive my favorite multiple times, I still get my favorite back, how often I do not affect the result with GET. There are several other ways in which Twitter has broken the other barriers. Many of these issues have been included here before. Update After using a little bit of Twitter API dos, there is actually an alternate URI format that correctly identifies the preferred resources. They show how to create a URL: There is also a long way from being calm, but at least this is a step in the right direction. is an example of Twitter violating the "resource identification" barrier. Every interesting resource should have a unique identifier. My twitter favorites and your twitter favorites are two different resources and therefore there should be two different URIs.
http://api.twitter.com/1/favorites/bob.json
Comments
Post a Comment