[RUS-WG] First Draft of the RUS Core IDL

[Gilbert Netzer]

Dear RUS members & chairs,

I want to submit the first draft of the interface definition language (IDL)
specification of the RUS Core.

The document can be found as an attachment to the following tracker:

https://forge.gridforum.org/sf/go/artf6015


This specification is the outcome of the discussions at OGF19 were it was
suggested to follow the idea of the ByteIO-WG and split the specification
into an abstract part and concrete profile renderings.

It is basically based on the current OGSA RUS draft 17, with the addition
of the suggestions from OGF20 and group discussions. Beyond that I also
used the opportunity to add extension points to the current draft, which I
would like to suggest for discussion.

The following changes have been incorporated:
- listMandatoryElements returns list of QNames instead of empty elements.
- extractUsageRecords (former extractRUSUsageRecords) returns list of plain
UsageRecords, the RUSUsageRecord element is gone.
- extractRecordHistory added to expose the record history that was exposed
via the RUSUsageRecord
- extractUsageRecords can now also return a WS-Enumeration context to allow
extraction of larger result sets
- incrementUsageRecordPart and replaceUsageRecords have been removed
because they can be expressed with modifyUsageRecords
- modifyUsageRecordPart was extended to become modifyUsageRecords to allow
general modification of many usage records with one request
- deleteSpecificUsageRecords was removed because can be expressed with
deleteUsageRecords
- deleteRecords was renamed to deleteUsageRecords

New features:
- All search terms/filters/update expressions were changed to xs:any and
now have an URI attribute to allow different query dialects to be used.
Currently the specification mandates XPath 1.0 for delete and extract and
XQuery Update Facilities for modify.
- xs:any extension points were added to all operation request and response
messages as well as to the record history.
- The type of operation in the record history is now specified using a URI
allowing future and implementor extensiability.


This document currently is very much a work in progress, some sections are
still missing (authors, abstract, contributors, appendix amongst others)
and the text can need some more work. However I feel that it is time to
present this to the working group for discussion so that we can come to a
consent about the general direction and move on from there.

Besides wanting to encourage people to critically examine the presented
document, I want also to especially encourage expressions of support!
I also want to suggest that this document be used as basis for discussion
at the upcoming OGF21 were the RUS-WG will have two sessions.

Thank you and best regards!
Gilbert Netzer
KTH/PDC