Open Grid Forum: Data Format Description Language Working Group OGF DFDL Working Group Call, July 14-2010 Attendees Steve Hanson (IBM) Alan Powell (IBM) Stephanie Fetzer (IBM) Tim Kimber(IBM) Suman Kalia (IBM) Apologies Mike Beckerle (Oco) 1. Current Actions Updated below 2 Nils and Defaults. Discussed Alan updates (v9). Updates still needed. 3. using textStringPadCharacter with charRef '%#r' on multi-byte encoding textStringPadCharacter DFDL String literal The padding character or byte value that is used when justifying or trimming text elements. A pad character can be specified using a literal or DFDL entities. A pad byte value must be specified using the %#r entity. DFDL validation rules - if a pad character is specified when lengthUnits='bytes' then the pad character must be a single-byte character. - if a pad byte value is specified when lengthUnits='characters' then the encoding must be a fixed-width encoding. Padding and trimming is be applied using an array of N pad bytes values, where N is the width of a character in the fixed-width encoding. Annotation: dfdl:element, dfdl:simpleType Would adding fillByte to dfdl:textPadKind enumeration make it clearer. padChar is a pad character (%#r not allowed) fillByte is a fill byte Although there was some merit in having separate textPadKinds it was felt it was too late to make the change. The property descriptions (for numbers, calendars and booleans as well) will be improved as above. 4. nilIndicatorPath and nilIndicatorIndex properties These properties seem a bit of an anomaly. Tim has suggested they can be simplified. Tim had emaled Mike with a proposal and Steve had responded with 3 alternatives 1) Leave things as they are, perhaps renaming 'nilIndicator' to 'indicator' 2) Collapse nilIndicatorPath and nilIndicatorIndex to a single property. It means the path is repeated, but it makes nils consistent with length and occurs. For all, you can always use a string variable to hold the constant part of the path and concatenate with the index. And if we improve usability in a future DFDL release, we would improve it for nils, lengths and occurs at the same time. 3) Drop the properties for 1.0 altogether on the grounds that it is a rare case. I've never seen such a format, but we know Mike has. As part of nils default discussion the WG decided to drop dfdl:nilIndicator and dfdl:nilIndicatorIndex as it was felt the complications it introduced on unparsing were not worth the additional function. This may be revisited for DFDL v2. 5. Alignment enumerations when dfdl:lengthUnits = 'bits' When alignmentUnits='bits', the rule that alignment must be 1,2,4,8 etc or multiple of 2 looks rather hard to justify. We should either a) remove the rule entirely and allow any postiive integer > 1 or b) make the rule apply only when alignmentUnits='bytes' It was agreed to remove the restriction on aligmentUnits for both 'bytes' and 'bits'. Also suggested that a note should be added to the default alignment and implicit length tables that 'although the values are specified in bits that does not imply that the dfdl:lengthUnits can be specified as 'bits' for all the simple types 6. Semantics of 'fixed' Not discussed. Action Raised. 7. Clarify the specification of error reporting from a DFDL processor - section 2.3 needs to be updated Not discussed. Action Raised 8 Asserts and discriminators - specify the scope of forward references. Must be downward-only. The expression must be resolvable by the time the component on which it is positioned goes out of scope - otherwise it is a processing error. Not discussed. Action Raised 9 Expressions Discuss error behaviour when evaluating an expression in various contexts - All properties: wrong type returned : schema definition error exception when evaluating expression : schema definition error referenced variables/paths not available : schema definition error - Properties which allow a forward reference referenced variables/paths not available : no error. DFDL processor continues processing until the expression result is available, then acts on the result. Not discussed. Action Raised 10. Grammar for DFDL expression - currently allows multiple predicates on a single expression. Sounds as if this was unintended. Not discussed. Action Raised Meeting closed, 17:00 Next call Wednesday 21 July 2010 15:00 UK (10:00 ET) Next action: 106 Actions raised at this meeting No Action 101 Semantics of 'fixed' 102 Clarify the specification of error reporting from a DFDL processor - section 2.3 needs to be updated 103 Asserts and discriminators - specify the scope of forward references. Must be downward-only. The expression must be resolvable by the time the component on which it is positioned goes out of scope - otherwise it is a processing error. 104 Expressions Discuss error behaviour when evaluating an expression in various contexts - All properties: wrong type returned : schema definition error exception when evaluating expression : schema definition error referenced variables/paths not available : schema definition error - Properties which allow a forward reference referenced variables/paths not available : no error. DFDL processor continues processing until the expression result is available, then acts on the result. 105 Grammar for DFDL expression - currently allows multiple predicates on a single expression. Sounds as if this was unintended. Current Actions: No Action 066 Investigate format for defining test cases 25/11:IBM to see if it is possible to publish its test case format. 04/12: no update ... 17/02: IBM is willing in principle to publish the test case format and some of the test cases. May need some time to build a 'compliance suite' 24/03: No progress 03/03: Discussions have been taking place on the subset of tests that will be provided. 10/03: work is progressing 17/03: work is progressing 31/03: work is progressing 14/04: And XML test case format has been defined and is being tested. 21/04. Schema for TDML defined. Need to define how this and the test cases will be made public 05/05: Work still progressing 12/05: Work still progressing 02/06: Work still progressing on technical and legal considerations ... 14/07: work continues 085 ALL: publicize Public comments phase to ensure a good review.. 14/04: see minutes 21/04: Press release, OMG and other standards bodies. 05/05: Alan and Steve H have contacted other standards bodies. Will ask them to add comments on spec 15/05: still no public comments 02/06: No public comments 16/06: Public comments period has ended with no external comments. Alan had posted changes made in draft 041. Steve suggested send a note to the WG highlighting these changes. Steve also suggested requesting an extension as other IBM groups may review. We discussed whether this was necessary as changes will need to be made during the implementation phase anyway. Alan to ask OGF what the process is for changes post public comment. 23/06: Still no comments. Alan will contact OGF to understand the rest of the process. 30/06: Alan has emailed Joel asking what the process is now public comment period is over andcan we update the published version with WG updates. No response yet. 07/07: No response. Alan will chase up 14/07: No response from Joel. Sent email to Greg Newby by no response. 086 AP: Nils and Defaults during unparsing - update table 31/03: TK to documetn use cases for parsing 14/04: Investigate new property to control empty string behaviour. 21/04: After investigation a new property is not required. New rules developed and tables updated. Need examples of complexTypes to confirm tables apply. Review Nils, defaulting spec section. 05/05: Discussed defaulting complex elements. Tables updated but need to add terminator. SH; to confirm WMD behaviour when infoset item has no value on unparsing Need to describe defaulting choices. 15/05: More discussion. Alan updating sections 26/05: Discussed draft updates. Stephanie to confirm asserts do not make an element required. Alan will update draft.. All: review rest of draft. 02/06: Alan updated description. Please review. Discussed Stephanie's example using discriminators. Decided no changes needed. 16/05: went through Steves comments. Steve to update draft. 23/06: Steve's updates to the rules discussed. See minutes. Rest of document needs updating. 30/06: Discussed Alans updates. Some corrections. Alan will send out updated copy for review before next call. 07/07: Discussed Alan updates and Tim and Steve's comments. Still some corrections and updates. 14/07: Discussed Alan updates (v9) Still some corrections and updates. 099 Splitting the specification in simpler sections. 07/07: Steve sent a proposal but not discussed. Alan will arrange a separate call. 14/07:Discussed Steve's proposal and Suman's and Alan's comments. Need to add choice, validation, facets. Also how does an implementation declare which subsets it supports. Suggested levels and/or profiles. Steve highlighted a problem when a DFDL schema from an implementation of just the core functions was moved to a full DFDL implementation what should happen about the missing properties. Does the full implementation need to be aware of subsets of functions? Should it raise a schema definition error for use of a function not in the subset. 101 Semantics of 'fixed' 102 Clarify the specification of error reporting from a DFDL processor - section 2.3 needs to be updated 103 Asserts and discriminators - specify the scope of forward references. Must be downward-only. The expression must be resolvable by the time the component on which it is positioned goes out of scope - otherwise it is a processing error. 104 Expressions Discuss error behaviour when evaluating an expression in various contexts - All properties: wrong type returned : schema definition error exception when evaluating expression : schema definition error referenced variables/paths not available : schema definition error - Properties which allow a forward reference referenced variables/paths not available : no error. DFDL processor continues processing until the expression result is available, then acts on the result. 105 Grammar for DFDL expression - currently allows multiple predicates on a single expression. Sounds as if this was unintended. Closed actions No Action 088 define semantics of choiceKind 'fixedLength' 31/03: TK to provide definition of calculable length. Investigate PL/I varchars and Cobol occurs dependingon. 14/04Tim had distributed a document starting the definition of calculable length for the longest choice member. Alan had done some investigation of COBOL occurs depending on and when used in the working section of a program then the maximum storage was reserved but when used in the linkage section the dependent number was uses. We need to understand how the WMB COBOL importer deals with ODO. 21/04: Need to define 'calculable length' and WMB importer ODO behaviour. 05/05: TK: Still need definition of calculable length. SKK: WMB COBOL imported behaviour with ODO 15/05: Suman sent an expmle of an imported Cobol ODo which suggested that the maximum space was reserved. He will extend the example. 02/06: no progress 16/06: no progress 23/06: no progress 30/06: Alan looked at Tim's description of calculable length and suggested that that real use case may be much simpler. If real use case is COBOL and C importers then it would be cleaner to require the 'fixed length' to be specified on the enclosing complex element and remove choiceKind 'fixed length'. Ask Suman is COBOL and C importers can be enhance to provide length on cpmplex element. 07/07: Steve had proposed 5 possible approaches to supporting main use case. i) Change the importer to add a parent element for each group. Con: This changes the logical model by inserting an extra level. ii) Allow dfdl:length to be carried on embedded xs:choice (along with dfdl:lengthUnits & dfdl:lengthKind). iii) New dfdl:choiceLength property to be carried on embedded xs:choice (always bytes). iv) dfdl:choiceKind as today. Con: length from TD model is lost and must be recalculated. v) dfdl:choiceKind but with limitations to make the calculation easier. Con: length from TD model is lost and must be recalculated. Agreed a modified version of iii) dfdl:choiceKind becomes dfdl:choiceLengthKind with emuns 'implicit' (length of selected branch) and 'explicit' ( length specified by dfd;choiceLength) New property dfdl:choiceLength 14/07: Closed 096 AP: using textStringPadCharacter with charRef '%#r' on multi-byte encoding 14/07: Although there was some merit in having separate textPadKinds it was felt it was too late to make the change. The property descriptions (for numbers, calendars and booleans as well) will be improved. Closed 097 nilIndicatorPath and nilIndicatorIndex properties 07/07: Tim had emaled Mike with a proposal and Steve had responded with 3 alternatives 1) Leave things as they are, perhaps renaming 'nilIndicator' to 'indicator' 2) Collapse nilIndicatorPath and nilIndicatorIndex to a single property. It means the path is repeated, but it makes nils consistent with length and occurs. For all, you can always use a string variable to hold the constant part of the path and concatenate with the index. And if we improve usability in a future DFDL release, we would improve it for nils, lengths and occurs at the same time. 3) Drop the properties for 1.0 altogether on the grounds that it is a rare case. I've never seen such a format, but we know Mike has. Not discussed. 14/07: As part of nils default discussion the WG decided to drop dfdl:nilIndicator and dfdl:nilIndicatorIndex as it was felt the complications it introduced on unparsing were not worth the additional function. This may be revisited for DFDL v2.. Closed 100 Alignment enumerations when dfdl:lengthUnits = 'bits' When alignmentUnits='bits', the rule that alignment must be 1,2,4,8 etc or multiple of 2 looks rather hard to justify. It was agreed to remove the restriction on aligmentUnits for both 'bytes' and 'bits'. Closed Work items: No Item target version status 005 Improvements on property descriptions not started 012 Reordering the properties discussion: move representation earlier, improve flow of topics not started 036 Update dfdl schema with change properties ongoing 042 Mapping of the DFDL infoset to XDM none not required for V1 specification 070 Write DFDL primer 071 Write test cases. 083 Implement RFC2116 105 AP: Describe trailingSkipBytes for delimited formats. Alan suggested 'dfdl:terminator must be specified and not empty if dfdl:lengthKind is delimited or endOfParent.' 106 AP: Skip Bytes should allow bits Ageed that it should be possible to specify bits. - LSB and TSB renames to dfdl:leadingSkip, dfdl:trailingSkip - units are specified by dfdl:alignmentUnits. 107 Remove timing from dfdl:assert 108 AP: Confirm behaviour of defaulting with various occursCountKinds and separator policies. 30/06: Decided that defaulting for variable occurrence arrays should always be to minOccurs. If separatorPolicy is 'required' then just the separators will be output up to maxOccurs and unbounded is an error. 109 define semantics of choiceKind 'fixedLength' dfdl:choiceKind becomes dfdl:choiceLengthKind with emuns 'implicit' (length of selected branch) and 'explicit' ( length specified by dfd;choiceLength) New property dfdl:choiceLength 110 nilIndicatorPath and nilIndicatorIndex properties 14/07: As part of nils default discussion the WG decided to drop dfdl:nilIndicator and dfdl:nilIndicatorIndex 111 AP: using textStringPadCharacter with charRef '%#r' on multi-byte encoding 14/07: Although there was some merit in having separate textPadKinds it was felt it was too late to make the change. The property descriptions (for numbers, calendars and booleans as well) will be improved. Closed 112 Alignment enumerations when dfdl:lengthUnits = 'bits' When alignmentUnits='bits', the rule that alignment must be 1,2,4,8 etc or multiple of 2 looks rather hard to justify. It was agreed to remove the restriction on aligmentUnits for both 'bytes' and 'bits'. Closed Regards Alan Powell Development - MQSeries, Message Broker, ESB IBM Software Group, Application and Integration Middleware Software ------------------------------------------------------------------------------------------------------------------------------------------- IBM MP211, Hursley Park Hursley, SO21 2JN United Kingdom Phone: +44-1962-815073 e-mail: alan_powell@uk.ibm.com 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