Re: [sca-assembly] NEW ISSUE: URI attribute should not be mandatory for bindings

[Scott Vorthmann <scottv@tibco.com>]

NEW ISSUE logged as:

http://www.osoa.org/jira/browse/ASSEMBLY-10


On Oct 11, 2007, at 12:05 AM, Peshev, Peter wrote:

>
> TARGET: SCA Assembly Specification, section 1.7 Binding
>
> DESCRIPTION:
> The current description of the uri attribute for a binding says  
> that the
> uri attribute is mandatory in case the binding is attached to a
> reference in the domain.
>
>
> The current wording on line 2300 is :
>
> binding element has the following attributes:
>  * uri (optional) - has the following semantic.
>   ...
>  It is optional for references defined in composites used as component
> implementations, but required for references defined in composites
> contributed to SCA domains.
>
>
> As clarified by Mike E.
> (http://www.oasis-open.org/apps/org/workgroup/sca-assembly/email/ 
> archive
> s/200710/msg00059.html) that may not be true.
>
>
>
> PROPOSAL:
> Remove the requirement for the uri and add the following clarification
> paragraph :
>
>
> Some bindings MAY identify a target endpoint without using the uri
> attribute  via usage of some binding specific tags and attributes. In
> that case the uri for references in composites contributed to the SCA
> domain is OPTIONAL. However such binding definition without uri
> attribute MUST uniquely specify a target endpoint, since references
> contributed to the SCA domain can't be reconfigured by anything and  
> the
> SCA runtime will need enough information to ivoke correctly such
> reference. In case the binding relies only on the uri to configure the
> target endpoint, the uri attribute MUST be present for bindings for
> references contributed to the SCA domain.
>
>