In the absence of a compelling reason to allow relative paths here, I think the conservative choice is to retain this rule from 6.3.2.

However, I did think about this and my evaluation on this topic is that allowing relative paths makes things just a little more complicated analytically to insure one gets good performance, but as DFDL does not allow recursive types, it is a relatively simple analysis to convert any relative path use in DFDL into absolute paths. So I am not worried about there being performance impact.

...mikeb

On Mon, Apr 2, 2012 at 7:30 AM, Steve Hanson <smh@uk.ibm.com> wrote:
We took action 168 on the last call to discuss the issue of relative paths in expressions for properties in global formats.

168
Relative path expressions in global annotations (Mike)
28/03: Mike to check that allowing relative paths does not cause performance problems for an implementation.
05/04:


Just seen this in section 6.3.2. of the spec:  "DFDL expressions that are used to provide the value of DFDL properties in the dfd:format annotation on the xs:schema MAY NOT contain relative paths".

Regards

Steve Hanson
Architect, Data Format Description Language (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 02/04/2012 12:28 -----

From:        Steve Hanson/UK/IBM
To:        dfdl-wg@ogf.org
Date:        14/03/2012 16:09
Subject:        Escape scheme properties that can be expressions



The spec allows the escapeCharacter and escapeEscapeCharacter of DFDL escapeScheme to be expressions, in case the characters are supplied dynamically in the data stream.
However given that escape schemes are defined globally, should these properties be permitted to take relative paths in the expression?  
Today the spec implicitly allows this, meaning an implementation can not entirely validate an escapeScheme annotation up front, but must validate escapeCharacter and escapeEscapeCharacter at each point of use.
I'm not sure it makes much sense for the path to be anything than absolute.

Regards

Steve Hanson
Architect, Data Format Description Language (DFDL)
Co-Chair, OGF DFDL Working Group
IBM SWG, Hursley, UK
smh@uk.ibm.com
tel:+44-1962-815848



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











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







--
 dfdl-wg mailing list
 dfdl-wg@ogf.org
 https://www.ogf.org/mailman/listinfo/dfdl-wg



--
Mike Beckerle | OGF DFDL WG Co-Chair 
Tel:  781-330-0412