chemistry-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Suhaib Fahad" <>
Subject RE: CMIS Swagger REST API - PogeyanOSS
Date Fri, 06 Oct 2017 11:45:51 GMT
Hi Florian,

Thanks for the clarity, I understand this can't be a CMIS specification in itself, its more
a standalone component. What I was actually meaning is to include in this link,

Very similar to the OpenCMIS Bridge.

For other things, I'll post the questions for sure.


-----Original Message-----
From: Florian Müller [] 
Sent: Friday, October 6, 2017 4:10 PM
To: Suhaib Fahad <>
Subject: Re: CMIS Swagger REST API - PogeyanOSS

Hi Fahad,

You are basically defining a new (fourth) CMIS binding. To be honest, I don't believe it will
be successful. It's not part of the CMIS specification and therefore no major ECM vendor will
adopt it.
Although, the browser binding doesn't follow all REST principals (for historical reasons),
it is as easy to use as a REST interface - even without a library. A Swagger definition for
the browser binding would be something useful...

The deployment question depends on the environment. The OpenCMIS bridge is similar to want
you described. It's an independent, standalone façade in front of a CMIS server. So, that
should work.

If you want to contribute something to Apache Chemistry, you have found the right place. When
you have a proposal, please post it on the mailing list and we will discuss it.

- Florian

> Hi Florian,
> Thanks for your point on licensing, I have updated that.
> From the project itself, we would like to see feedbacks on the 
> following,
> 1) Can we add more REST endpoints that will help expose CMIS APIs thru 
> simple REST based access. For example: currently we have a POST HTTP 
> action generated for every base type + custom type that is found thru 
> the Type Service, the REST side of the story exposes a simple API, but 
> underlying it will create either a FOLDER, DOCUMENT or ITEM. This is 
> really simple to consume in existing applications without any 
> dependency on CMIS libs or understanding of underlying CMIS APIs.
> Similarly, Can we think of abstract notions for other services, such 
> as versioning for example, or navigating folder hierarchy etc., And 
> also would be helpful if someone can use it and give feedbacks 😊.
> 2) The Swagger definitions are basically a wrapper, which means it has 
> run as another micro-service in existing landscape. We wanted to have 
> separate components so we can have deployments done easily, is this 
> fine?
> 3) And lastly, if such a project is useful, Can it be taken into the 
> existing OSS projects in Apache Chemistry? We would like to contribute 
> such components in future and this will give us an insight on how to 
> proceed with the OpenCMIS community.
> Thanks,
> Fahad
> -----Original Message-----
> From: Florian Müller []
> Sent: Wednesday, October 4, 2017 7:11 PM
> To:; Suhaib Fahad <>
> Subject: Re: CMIS Swagger REST API - PogeyanOSS
> Hi Fahad,
> That sounds like an interesting project.
> How exactly can we help you?
> There is one thing you should definitely do. All source files must 
> have a header the matches the license - in your case the Apache 
> license.
> You can find the a header template here (scroll down to the appendix):
> - Florian
>> Hi,
>> I would like to introduce myself to the Apache Chemistry devs. We 
>> have been using Apache Chemistry/OpenCMIS from last 1.5 years within 
>> our internal content projects/customer projects.
>> Me and my team are working on some core CMIS components that we would 
>> like to open source as well. As part of the OSS release, our first 
>> component is the Swagger REST API wrapper to CMIS (planning to 
>> release some more in the coming months). It can work with any CMIS 
>> compliant server.
>> We believe that this component will help many adapt CMIS in an easier 
>> way into their existing client applications.
>> This is our first OSS release in the open, and we would like to hear 
>> feedbacks and comments about the approach.
>> Thanks,
>> Fahad

View raw message