Hi Achim, Re 1) The error message "Operation not supported by the repository for this object!" is generated by the client library (AtomPub only). It means that the server has not provided an AtomPub link for the operation you want to execute. It does not depend on the allowable actions.
In your case, the test creates a document with content but your server doesn't return a "Content" tag which makes it impossible for the client to retrieve the content. (Make sure that the ObjectInfo.hasContent flags on the server is set to true if the document has content.) Re 2) cmis:isVersionSeriesCheckedOut is a boolean and has to be either true or false. The spec says "Repository MUST return this property with non-empty value...". If the type is not versionable the repository can pick a value. False would be logical... Mh, actually that should be a FAILURE, not a WARNING. Florian On 26/08/2011 14:52, Weigel, Achim wrote: > Hi, > > I have two questions about the TCK in the CMIS Workbench: > > 1) When I let run the test "Create and Delete Document Test" for a type with > versioning I get the following error: > > UNEXPECTED_EXCEPTION: Exception: > org.apache.chemistry.opencmis.commons.exceptions.CmisNotSupportedException: > Operation not supported by the repository for this object! > (AbstractSessionTest.java:144) > > Unfortunately it is not clear what is going wrong here. But after several > trials I found out the following: > When I change my coding such that I add the allowed action > "canSetContentStream" to documents which are the latest version but not > checked out the problem disappears. But on such documents this operation is > not allowed in our repository. > > 2) When I let run the tests on the cmis:document base type I get many > warnings as the following: > > WARNING: New document object spec compliance (AbstractSessionTest.java:361) > WARNING: Property cmis:isVersionSeriesCheckedOut is not set! > (AbstractSessionTest.java:678) > > I do not understand why this is a warning for a type without versioning. > > > > Thank you very much in advance and best regards, > Achim > > >
