[DFDL-WG] Fw: DFDL: Action 027 - second rework of property precedence

Wed May 20 07:05:30 CDT 2009

Alan, Mike

Thanks for the feedback. Here is latest version that addresses all 
comments except where noted below. The biggest changes are:

- Trimming & padding are now part of the 'extraction' & 'insertion' 
phases, and not conversion.
- Complex element separated from simple element
- Simple type / local element / global element distinctions called out
- Repeat properties pulled forward into own phase

Notes:

1) Mike - I agree that dfdl:inputValueCalc and dfdl:outputValueCalc should 
not  be allowed in dfdl:format annotation (ie, scoping). Spec & xsds 
should be updated.

2) Mike - you say that escape schemes are needed when looking for 
initiators - we should discuss - I'm not sure this is a real requirement 
in practice - I've never come across this.

3) Mike - the DFDL parser/unparser is driven by model position - not by 
infoset position - you only get elements in the infoset. 

4) Mike - I don't think that dfdl:initiatedContent is needed when 
unparsing.

5) Alan - encoding and byteOrder do apply for sequence, choice and any, 
because there could be markup involved. 

6) Alan - lengthKind = 'pattern' - is this something that we should 
support on complex elements - assumed so for now. 

Regards

Steve Hanson
Programming Model Architect
WebSphere Message Brokers
Hursley, UK
Internet: smh at uk.ibm.com
Phone (+44)/(0) 1962-815848
----- Forwarded by Steve Hanson/UK/IBM on 13/05/2009 10:25 -----

My markup on this attached - mostly using comment balloons on the side.

Alan Powell/UK/IBM
11/05/2009 16:57

To
Steve Hanson/UK/IBM at IBMGB
cc
dfdl-wg at ogf.org, dfdl-wg-bounces at ogf.org
Subject
Re: [DFDL-WG] DFDL: Action 027 - rework of property precedence - plus 
issues arising

Steve

Comments

I don't think the tables answer the precedence of padding, escaping and 
encoding that we had when discussing escape schemes. On parsing it should 
be remove padding, remove escape characters, apply encoding. On unparsing 
it is the reverse but the tables look the same.
I think the difficulty is that escape scheme is used for identification 
and extraction but also during conversion.
I don't think the core properties (inputvaluecalc, etc) apply to sequence, 
choice or any.
xxxPadKind  is checked before xxxPadCharacter, xxpadxxx
Calendar-binary.  binaryCalendarFormatRef is only used when binCalRep = 
packed or bcd

Alan Powell

 MP 211, IBM UK Labs, Hursley,  Winchester, SO21 2JN, England
 Notes Id: Alan Powell/UK/IBM     email: alan_powell at uk.ibm.com 
 Tel: +44 (0)1962 815073                  Fax: +44 (0)1962 816898

From:
Steve Hanson/UK/IBM at IBMGB
To:
dfdl-wg at ogf.org
Date:
08/05/2009 17:24
Subject:
[DFDL-WG] DFDL: Action 027 - rework of property precedence - plus issues 
arising

I've created a separate property precedence for each schema object that 
can carry non-scoping DFDL properties (attached for review). 

The following issues were noted: 

1) Missing property dfdl:textBooleanJustification - similar properties 
exist for string, number and calendar types. 

2) What is the rule when the same DFDL properties occur on a xs:simpleType 
and a xs:element that uses that type?  Does this work a) like 
element/group references (ie, properties combined with element winning) or 
b) like complex element and its sequence (ie, element and simpleType are 
considered separate objects)? I don't think section 10 covers this case. 

3) Should we allow the DFDL nil & default control properties on a simple 
type?  xs:nillable and xs:default are element only attributes in xsd. Spec 
currently allows this. 

4) Should we allow DFDL occurs properties on global elements?  Whether 
something repeats is a particle thing. Spec currently allows this. (IBM's 
WTX and MRM don't allow this). 

5) Missing work item to get BiDi properties into shape and incorporated 
into spec. Should these be grouped, like escape scheme, calendar scheme, 
etc? Do they apply to calendar and number types? 

6) Should dfdl:integerBooleanXXXRep be renamed dfdl:binaryBooleanXXXRep ? 

7) We might want to reconsider the name of the new flag dfdl:initiated - 
it could be read that the xs:sequence itself is initiated rather than its 
children.   

8) Should dfdl:initiated also apply to xs:choice? 

9) Draft 33 property precedence had dfdl:outputLengthCalc - but that is 
not in the spec anywhere else? 

Regards

Steve Hanson
Programming Model Architect
WebSphere Message Brokers
Hursley, UK
Internet: smh at uk.ibm.com
Phone (+44)/(0) 1962-815848

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 

[attachment "Property Precedence 034.doc" deleted by Alan Powell/UK/IBM] 
--
  dfdl-wg mailing list
  dfdl-wg at ogf.org
  http://www.ogf.org/mailman/listinfo/dfdl-wg

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/20090520/7c7fd639/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Property Precedence 034-mikeb.doc
Type: application/octet-stream
Size: 179200 bytes
Desc: not available
Url : http://www.ogf.org/pipermail/dfdl-wg/attachments/20090520/7c7fd639/attachment-0002.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Property Precedence 034.doc
Type: application/octet-stream
Size: 93696 bytes
Desc: not available
Url : http://www.ogf.org/pipermail/dfdl-wg/attachments/20090520/7c7fd639/attachment-0003.obj 