FW: [Bug 5241] define how the SML URI scheme is resolved

Team,
I have added a proposal for this bug. Please reply to this email if you are not comfortable with this proposal.

Can the EPR gurus add a similar proposal to bug# 5242 (define how the EPR scheme is resolved)?



------- Comment #1 from kumarp@microsoft.com  2007-11-13 05:40 -------

Proposal:
Replace the current text of bullet 2 in section 4.2.1 SML URI scheme.

2. The SML URI scheme is resolved using the following steps:
a.      A document is obtained by dereferencing the URI, sans fragment
identifier, using the appropriate operation defined for the scheme used in that
URI. If there is no document retrieved, the scheme is unresolved.
b.      If a fragment identifier is not present in the URI, the root element of
the retrieved document is returned.
c.      If a fragment identifier is present in the URI, the fragment identifier
is applied to the root element of the retrieved document and the resulting
node-set is returned.

Reasons:
A URI is a broadly defined term. Although most URIs fit into the 'resource
access' paradigm, not all do (for example, the mailto scheme). The specific
operations supported for a given URI scheme depend on the scheme definition.
There is no standard set of common operations meaningful across all URI
schemes. This means that it is not possible to define a set of operations
required for dereferencing a URI because it depends on the scheme used. Since
the sml:uri definition does not place any restriction on the schemes that can
be used, it is not possible to define the set of operations to be performed in
order to dereference sml:uri. Consequently, we define this simply as
"dereference the URI using the appropriate operation defined for the scheme
used".

Received on Tuesday, 13 November 2007 05:57:12 UTC