Hi Marc-André,

CMIS defines three bindings and one of those is the Web Services binding. The
binding implementations are hidden from the client and server APIs. Ideally,
neither a client nor a server developer needs to know anything about the
binding. That's why you don't find any JAX-WS traces in the InMemory repository
code. The JAX-WS code is a layer below the repository code.

Have a look at the modules chemistry-opencmis-client-bindings (for the client
code), chemistry-opencmis-server-bindings (for the server code), and
chemistry-opencmis-commons-impl (for common binding code). Here you find the
JAX-WS implementations.

If you are interested in the SOAP traffic, install the InMemory repository and
the CMIS Workbench, run the TCK from the CMIS Workbench, and record the HTTP
traffic. The TCK calls almost all CMIS operations and should provide a good
sample for your analysis.


- Florian

> 
> Hello,
> 
> I guess I used the wrong word when I said 'the API'. I am seeing that
> you guys took  design best practices very seriously, with
> interface/implementation separation for many implementations.
> 
> Right now, I'm only able to handle Jax-WS web service calls. And I'm not
> detecting uses of the Jax-WS APIs in the in-memory test case. So either
> I got a bug, or this test target doesn't use the ws implementation of
> the API.
> 
> Anybody knows about the latter hypothesis?
> 
> Regards,
> 
> Marc-André Laverdière-Papineau
> Doctorant - PhD Candidate
> 
> On 07/23/2014 10:35 AM, Huebel, Jens wrote:
> > HI Marc-André,
> > 
> > You can use the in memory server. Just unzip the downloaded achieve and
> > copy the file chemistry-opencmis-server-inmemory-0.11.0.war to the web app
> > dir of Tomcat (I would recommend to rename it to inmemory.war)
> > 
> > You can use then a URL like http://localhost:8080/inmemory/browser in the
> > workbench or whatever client you use. With a few exceptions the whole CMIS
> > 1.1 API is covered. But there are some restrictions.
> > 
> > Jens
> > 
> > 
> > On 23/07/14 03:28, "Marc-André Laverdière"
> > <marc-andre.laverdiere-papin...@polymtl.ca> wrote:
> > 
> >> Thanks for the pointers.
> >>
> >> Two questions: what about the in-memory server? Is this something that
> >> implements the whole API?
> >>
> >> Also, I would need some entry points - preferably servlets. I am seeing
> >> some small code on the samples folder (and they use main), but that's
> >> about it.
> >>
> >> Do you know of any other sample app?
> >>
> >> Regards,
> >>
> >> Marc-André Laverdière-Papineau
> >> Doctorant - PhD Candidate
> >>
> >> On 07/22/2014 04:58 PM, Mark Streit wrote:
> >>> We use Chemistry's OpenCMIS JARs for a project that interacts with
> >>> Alfresco
> >>> Enterprise server and on the CLIENT side we use the following 5
> >>> artifacts
> >>> and their dependencies:
> >>>
> >>>
> >>> /org/apache/chemistry/opencmis/chemistry-opencmis-client-api/0.10.0/chemi
> >>> stry-opencmis-client-api-0.10.0.jar
> >>>
> >>> /org/apache/chemistry/opencmis/chemistry-opencmis-commons-api/0.10.0/chem
> >>> istry-opencmis-commons-api-0.10.0.jar
> >>>
> >>> /org/apache/chemistry/opencmis/chemistry-opencmis-client-bindings/0.10.0/
> >>> chemistry-opencmis-client-bindings-0.10.0.jar
> >>>
> >>> /org/apache/chemistry/opencmis/chemistry-opencmis-commons-impl/0.10.0/che
> >>> mistry-opencmis-commons-impl-0.10.0.jar
> >>>
> >>> /org/apache/chemistry/opencmis/chemistry-opencmis-client-impl/0.10.0/chem
> >>> istry-opencmis-client-impl-0.10.0.jar
> >>>
> >>>
> >>>
> >>> http://chemistry.apache.org/java/0.10.0/maven/dependency-convergence.html
> >>>
> >>> There are better experts than me out there, but hope this helps some.
> >>>
> >>> Mark
> >>>
> >>> On Tue, Jul 22, 2014 at 3:33 PM, Marc-André Laverdière <
> >>> marc-andre.laverdiere-papin...@polymtl.ca> wrote:
> >>>
> >>>> Hello,
> >>>>
> >>>> I'm a PhD student working on vulnerability analysis for JEE
> >>>> applications. I have selected Chemistry for analysis because it is a
> >>>> large-scale project with mature code and web services.
> >>>>
> >>>> Since the project is so big, I end up being unsure of what parts to
> >>>> include and which ones to leave out.
> >>>>
> >>>> I would like - for starters - to look at a full round-trip between a
> >>>> client and a server. Which would be the minimal set of artifacts to
> >>>> include?
> >>>>
> >>>> Regards,
> >>>>
> >>>> --
> >>>> Marc-André Laverdière-Papineau
> >>>> Doctorant - PhD Candidate
> >>>>
> >>>
> > 
> > 
>

Reply via email to