Understanding CMIS

Question

I have been investigating CMIS and frankly I am so super confused, as I am new to it. I have gone over the information in CMIS v1.1 released 23 May 2013 and find that there are too many areas they do not cover, like, how would one create your own repository, for example.

Now my question is, if I wanted to create a CMS following the CMIS standard, how would I get started? Because somehow I feel as though I am not understanding the concepts?

Just an example, you get the DotCMIS, which as I understand is a client library, for what? To create a CMIS or to query a CMIS?

Answer 1

CMIS aims at becoming the standard API for B2B content integration. As such, it describes an interface that a CMS must expose, as a REST or SOAP set of endpoints, to be compliant to the spec. This also means that implementing a repository is completely out of scope for CMIS itself, since it's meant to hide all implementation details behind a standard interface.

The easiest way to get started with a CMIS compliant server would be to use the Java SPIs provided by the Chemistry project, which basically implement the protocol itself and ask you to provide concrete implementations for the persistency and business logic layers. I'm not into .NET and can't advise you there.

Bottom line is, you can choose every single detail you want when implementing your CMS, and find the proper mapping to CMIS if you want to expose that API in terms of data model and operations.

Answer 2

I am new in this area too and the way how I got started in this was using java Eclipse EE IDE and the Chemistry project. I also use Alfresco as the container repo. So you need to install Alfresco, start the server and then you can start programing in java. Don't forget to configure the java IDE for this purpose. In the oficial web of Apache Chemistry you got some example of code. Alfresco installed CMIS as default, but if you use other repo as Magnolia you need to install CMIS as a module. Hope this will help you to initiate.