Hi Jason Sorry if I wasn't clear I mentioned "Resource definitions" and "Backend endpoints" not "backend resource definitions". Please refer http://raml.org/docs.html it has a summary of RAML Spec.
And following is a mapping we can have with RAML and API in AM In ROOT Section 1 ) title = name 2 ) version = version 3 ) baseUri = This can be interpreted as the back-end endpoint once the API is published it will get change to the proxy. 4 ) "Protocols" in ROOT sections can be mapped to "Transports" in API. Details provided in RESOURCES and METHODS sections can be mapped to APIM Resource definitions. Cheers Jo PS : Following was the use case I shared with Jason. Please see the following use case on how RAML can be used. * User will define an RESTful API with RAML either in a RAML editor integrated with APIM or in a 3rd party RAML editor and upload it to APIM. * Then user can tell APIM to create an API based on the written RAML document. * APIM will process the RAML and extract API Details , Resource Definitions and Back-end endpoint details and create the API. * APIM will also use the RAML document to generate the API Console and the documentation. * If a user wants to change the API he/she will modify the RAML and ask APIM to apply the changes to the API and documentation. In the above use case we are using RAML as an alternative to form driven UI which is more efficient for a developer to use. On Mon, Apr 7, 2014 at 3:00 PM, Jason Catlin <[email protected]> wrote: > Hey Guys, > > I'm looking into RAML/API Manger integration. > > So the first thing that I'm just a little confused about is that I'm > not excatly sure what deep processing of RAML will be done by the API > Manager. > > when the API Manger presents a WSDL, for instance, it doesn't actually > really look inside it; it just uploads the WSDL given by the user and > presents it to the API consumer. > > So, it's not really clear to me how RAML should be any different. > Won't the APIM just take the user's RAML and give it to the API > consumer? People keep seeming to imply that there will be some sort > of deeper processing involved, but I can't figure out what that is. > > In an earlier email, Joe Fonseka mentioned backend resource > definitions. But, I've read through this http://raml.org/spec.html > pretty carefully, and RAML doesn't specify backing resources. It's an > interface definition language. So we can't get the backend resources > from RAML. > > I'm sure we could extract a few things like names and descriptions, > but remember that this wouldn't really involve any deep processing of > the RAML. I mean, if that's what we want, fine, but just saying. > > jc > -- -- *Joseph Fonseka* WSO2 Inc.; http://wso2.com lean.enterprise.middleware mobile: +94 772 512 430 skype: jpfonseka * <http://lk.linkedin.com/in/rumeshbandara>*
_______________________________________________ Architecture mailing list [email protected] https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
