Forwarding to WG. Agenda item on today's call.

Regards
 
Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848
----- Forwarded by Steve Hanson/UK/IBM on 02/12/2014 10:11 -----

From:        Steve Hanson/UK/IBM
To:        Mike Beckerle <mbeckerle.dfdl@gmail.com>
Cc:        Alex Wood1/UK/IBM@IBMGB
Date:        28/11/2014 16:45
Subject:        Re: [DFDL-WG] dfdl github - mil-std-2045 schema updated

---

Mike

Some more.  The hiddenGroupRefs revealed another bug in IBM DFDL (not surprising as we don't support them yet but have a lot of code in the editor to display hidden groups nonetheless). Fixing that caused some more validation to take place which revealed some more errors. I've added them below.

I also notice that your XPath comparisons are using '=' rather than 'eq', that is, a general comparison rather than a value comparison. My reading of section 23.4 of the DFDL spec is that DFDL expressions do not support general comparisons.

Regards
 
Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848



From:        Steve Hanson/UK/IBM
To:        Mike Beckerle <mbeckerle.dfdl@gmail.com>
Date:        28/11/2014 15:26
Subject:        Re: [DFDL-WG] dfdl github - mil-std-2045 schema updated

---

Mike

Taken a quick look at these, the following properties need adding to defineFormat "thesePropertiesShouldNotMatter" to stop errors:

        truncateSpecifiedLengthString="no"
        textPadKind="none"
        binaryNumberCheckPolicy="lax"
        fillByte="%#r00;"
        textBidi="no"
        floating="no"
        choiceLengthKind="implicit"
        useNilForDefault="no"

Seeing several of this error: It is because you are using lengthUnits 'bits' with xs:nonNegativeInteger (your common type tBigIntField). I don't believe the spec allows this.
CTDV1532E : DFDL property 'lengthUnits' can only be 'bits' if the representation is binary and the type is boolean, byte, unsignedByte, short, unsignedShort, int, unsignedInt, long or unsignedLong. Element: #xmlns(p="urn:milstd2045DFDL")xscd(/type::p:future_use_group_type/model::sequence/schemaElement::future_use_group_data).        

There are several things not yet supported by IBM DFDL, which we will get to eventually, but fyi they are:

        hiddenGroupRef
        encodingErrorPolicy 'replace'
        bitOrder
        encoding 'US-ASCII-7-bit-packed'

Plus a bug in IBM DFDL was revealed where if a1.xsd in tns 'a' includes a2.xsd in tns 'a' to pull in defineFormat name='xxx' then dfdl:format ref="tns:xxx" is not resolving. It works if a2.xsd does not have a tns (chameleon include), and it works if a2.xsd is tns 'b' (import).

Regards
 
Steve Hanson
Architect, IBM DFDL
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848




From:        Mike Beckerle <mbeckerle.dfdl@gmail.com>
To:        "dfdl-wg@ogf.org" <dfdl-wg@ogf.org>
Date:        19/11/2014 23:48
Subject:        [DFDL-WG] dfdl github - mil-std-2045 schema updated
Sent by:        dfdl-wg-bounces@ogf.org

---

I finally got around to enhancing the mil-std-2045 schema using the ideas that came out of review of the original work.

It is vastly improved in terms of complexity of the schema, and other schemas that have to generate this sort of thing are greatly simplified by the techniques illustrated here which avoid the need to generate any top-level group definitions.

I did have to put in the workaround of using "WSP*" in a terminator instead of "ES" because Daffodil doesn't support ES in terminators yet (there's a bug).

I also left out the defaults, which are needed for unparsing, due to a bug in Daffodil.

https://github.com/DFDLSchemas/mil-std-2045


Mike Beckerle | OGF DFDL Workgroup Co-Chair | Tresys Technology | www.tresys.com
Please note: Contributions to the DFDL Workgroup's email discussions are subject to the OGF Intellectual Property Policy
--
 dfdl-wg mailing list
 dfdl-wg@ogf.org
 https://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

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

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