Re: [nm-wg] immediate future of current schemas

[Paul D Mealor <pdm@hep.ucl.ac.uk>]

Hi all,

I have added a new draft version of the request schema and associated work
to my website, which includes some CHANGES, a COMBINED SCHEMA, and a
demonstration WSDL that supports BATCH HANDLING. See below.

http://www.hep.ucl.ac.uk/~pdm/nmwg/

Clearly you don't have to move onto this, but it contains some tidying up 
and it now follows all the requirements (that I'm going to support).

CHANGES

The changes to the main schema are are, I hope, minor. They are:

Ensure that the schema follows all the requirements in the requirements 
document except for:
  * time period can only be represented with a time and two tolerances, 
    not a start time and an end time

This means that a Host (for source and destination) can now contain just a 
name, and no hostname/ip, for if you want to just identify your monitoring 
nodes with a URI or somesuch.

Extracted the request definition into nm_requestbody.rnc, leaving
nm_request.rnc with just one "import" and a "start =". This is because
Trang does not understand the "external" keyword, and you cannot "import"  
a file with "start =" in, and this is required for the Fault schemas
below, and probably for almost any other schema that uses the request or
report schemas.


COMBINED SCHEMA

I have also added an equivalent combined schema:
http://www.hep.ucl.ac.uk/~pdm/nmwg/2004-10-28-NM-WG-Request-Schema-combined.rnc.html
and
http://www.hep.ucl.ac.uk/~pdm/nmwg/2004-10-28-NM-WG-Request-Schema-rnc/nm_combined.rnc

This is an exactly (I hope) equivalent schema in a single file, 
automatically generated from the individual schema components.

Generated XSD files of the major schema parts are available too.


BATCH HANDLING

I have, just for demonstration's sake, created a WSDL file
http://www.hep.ucl.ac.uk/~pdm/nmwg/2004-10-28-NM-WG-Request-Schema-rnc/services.wsdl

and some associated RNCs for faults and batch handling. These are
converted to XSD for use in the WSDL.

This contains a number of different service definitions (well, Port 
Types):

* to demonstrate that batch handling and transaction IDs can be done
  outside the request schema (but in an arena we may have control over)
* to show some use of SOAP faults, which probably should be discussed at 
  some point.
* to show how this might be put into an RPC or a DocLit service.

Cheers,

Paul

-- 
Paul Mealor
  --+->    pdm@hep.ucl.ac.uk    <-+--
  <-+-- www.hep.ucl.ac.uk/~pdm/ --+->
    |     +44 (0)20 7679 3044   <-+--
  --+->         x33044            |