[DFDL-WG] DFDL: Minutes from OGF WG call, 14 Nov 2007

[Ian W Parkinson]

Open Grid Forum: Data Format Description Language Working Group

Weekly Working Group Conference Call
17:00 GMT, 14 Nov 2007

Attendees
Geoff Judd (IBM)
Steve Hanson (IBM)
Suman Kalia (IBM)
Simon Parker (PolarLake)
Ian Parkinson (IBM)

Apologies
Mike Beckerle (IBM)


1. Information Set Proposal
Following last week's review of the Information Set proposal, Simon 
updated the proposal and distributed a new draft, version 7. Alan queried 
the reference to empty lists under "No Value", asking if this was 
necessary as DFDL does not use a concept of lists - this is carried over 
from the XML Information Set specification, and is only theoretical. The 
proposal will be distributed wider within IBM to solicit final comments.

2. Review of Revised DFDL Data Grammar
Mike has updated the Data Grammar portion of the DFDL specification, to 
take into account the InfoSet proposal, and distributed this via the 
DFDL-WG mailing list.

The meeting reviewed the revised grammar, touching upon the following 
points:
Both the ComplexContent and ComplexElement productions include a "Prefix" 
element. Alan asked whether this duplication was necessary. Steve and 
Simon noted that this had not changed during the revision and that there 
are cases where this is useful. The meeting agreed not to change this.
Alan asked why we need both Element and ComplexContent in the SequenceItem 
production. Element effectively introduces a name.
Simon noted that most of the changes were in the productions, and wondered 
what influence the InfoSet proposal actually had on the specification. 
Steve said that there had been real changes, for example with 
SimpleElement and ComplexElement.
Steve observed that Prefix and Postfix Separators no longer appear in the 
productions table. The meeting agreed that these were probably no longer 
necessary, and that this was a useful simplification, but to check this 
with Mike.

3. Object Constraint Language
The UML diagram in the revised InfoSet proposal uses Object Constraint 
Language1 (OCL) to specify an invariant on the Element class. Simon asked 
the meeting to consider whether this might be a useful approach in other 
sections of the DFDL specification, in particular when describing parser 
behaviour. Suman was concerned that it might prove useful only for 
specifying parser internals which should not be covered in the language 
specification.

4. Other Business
Suman asked if we need to define a parser API. Steve suggested that this 
should not form part of the language specification, but that there may 
need to be other specification documents.
Steve and Alan will be compiling a list of remaining work items to move 
the specification towards completion.
Work has started within IBM, to attempt to describe a number of industry 
standard data formats, including SWIFT and ISO 8583, using DFDL. Simon has 
a long standing work item to attempt to describe a number of custom 
application formats. These efforts should help validate the utility of 
DFDL and suggest specification revisions.
Thanksgiving is next week - will this affect attendence of next weeks 
meeting?

Meeting closed 17:35 GMT


Actions
Mike to comment on the omission of initial prefix Separators and final 
postfix Separators from the productions table, in the revised DFDL Data 
Grammar. Are these intended to be covered by other elements in the 
productions, or do they need to be added?

References
Object Constraint Language; specification at 
http://www.omg.org/technology/documents/formal/ocl.htm



Ian Parkinson
WebSphere ESB Development
Mail Point 211, Hursley Park, Hursley, Winchester, SO21 2JN, UK





Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU





-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ogf.org/pipermail/dfdl-wg/attachments/20071115/1132396e/attachment.html