[ 
https://issues.apache.org/jira/browse/JUDDI-881?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14255935#comment-14255935
 ] 

Alex O'Ree commented on JUDDI-881:
----------------------------------

Right, we're looking at the same thing. It's more than possible that the way 
the test case used to be, was incorrect.

Find business, with the tModel Bag specified, should return all businesses that 
contain services, that contain binding templates that reference all keys 
specified in the bag (unless there's a find qualifier that changes it from AND 
to OR). Because we're returning a businessEntity set, it is inclusive of all 
services owned by those businesses. This is consistent with the test and the 
results of the test, given the input files.

Previously, to JUDDI-573, findBusiness was returning only all businesses that 
contain services, that contain binding templates that reference all keys 
specified in the bag (unless there's a find qualifier that changes it from AND 
to OR). Previously, it was returning the business and only services owned by 
those businesses that use that tModel key. 

I don't see any verbiage in the spec that supports the previous behavior. If 
you see something please quote or link to it.

> find_business does not filter services when providing a tModelBag
> -----------------------------------------------------------------
>
>                 Key: JUDDI-881
>                 URL: https://issues.apache.org/jira/browse/JUDDI-881
>             Project: jUDDI
>          Issue Type: Bug
>          Components: core
>    Affects Versions: 3.1.5, 3.2, 3.2.1
>            Reporter: Jelle Van den Driessche
>            Assignee: Kurt T Stam
>             Fix For: 3.3
>
>
> When providing a tModelBag in the find_business call, the resulting services 
> should be filtered (UDDI spec 5.1.10.3: If a tModelBag or find_tModel was 
> used in the search, the resulting serviceInfos structure reflects data only 
> for the businessServices that actually contained a matching bindingTemplate).
> This has been fixed in JUDDI-398 and a testclass was written. For JUDDI-573, 
> the behavior of the testclass has been changed so that it no longer is 
> conform the spec.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to