Re: [cmis] Nikunj's blog postings on AtomPub vs CMIS

[Al Brown <albertcbrown@us.ibm.com>]

Yes.  One way to remove it is not provide an atom or app xsd and say what should be included where.  Each implementor will have to parse atom on their own.

Sent from BlackBerry.


----- Original Message -----
From: David Caruana [david.caruana@alfresco.com]
Sent: 05/13/2009 06:42 PM CET
To: Julian Reschke <julian.reschke@greenbytes.de>
Cc: CMIS TC List <cmis@lists.oasis-open.org>
Subject: Re: [cmis] Nikunj's blog postings on AtomPub vs CMIS



Ah yes, I was thinking of cmis:object which is optional (according to
the schema), but there is indeed cmis:terminator which is required,
but Alfresco doesn't enforce that one on inputs.

So, yes, if Alfresco enforced that one, it would fail AppClientTest
create tests.

It would be good to remove cmis:terminator element completely, as I
believe it's only there to work-around an XSD limitation, and is of no
value in the request/response documents at all.

Dave

On 13 May 2009, at 18:15, Julian Reschke wrote:

> David Caruana wrote:
>> I regularly execute the AppClientTest (by Joe Gregorio) against the
>> Alfresco server.
>> http://code.google.com/p/feedvalidator/wiki/AppClientTest
>> I haven't noticed any problems of the severity described. Managed
>> to create, read, update (sort of) and delete items.
>> Having said that, I don't discount the comment. There seems to be a
>> great deal of interesting CMIS analysis on that blog.
>> Dave
>
> Hi David,
>
> from my understanding, this could be because the Alfresco server
> doesn't  require that CMIS extension element to be present, right?
>
> BR, Julian


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php