Re: [dita] Question about -dita-use-conref-target in unexpected places

[Eliot Kimber <ekimber@contrext.com>]

I agree with Chris: it should be universal.

Cheers,

E.
—————
Eliot Kimber, Owner
Contrext, LLC
http://contrext.com




On 1/16/15, 12:22 PM, "Chris Nitchie" <chris.nitchie@oberontech.com> wrote:

>IMHO, yes, -dita-use-conref-target should be allowed wherever its
>presence would not result in DTD validation errors, regardless of the
>semantic purpose defined in the spec.
>
>
>Chris
>Chris Nitchie
>(734) 330-2978
>chris.nitchie@oberontech.com
>www.oberontech.com
> <http://www.oberontech.com/>
>Follow us:
> <https://www.facebook.com/oberontech>
> <https://twitter.com/oberontech>
> <http://www.linkedin.com/company/oberon-technologies>
> 
> 
>
>
>
>
>
>
>
>From: Robert D Anderson <robander@us.ibm.com>
>Date: Thursday, January 15, 2015 at 4:28 PM
>To: "dita@lists.oasis-open.org" <dita@lists.oasis-open.org>
>Subject: [dita] Question about -dita-use-conref-target in unexpected
>places
>
>
>
>In a DITA-OT forum we got a question about the extent to which
>"-dita-use-conref-target" is allowed as an attribute token.
>
>
>Reminder / background: this admittedly quirky token is the only way to
>override conref rules when you have a required attribute. Local
>attributes on a referencing element override those on the referenced
>element. If you have a required attribute, it used to
> be impossible to have conref set an attribute like @name on <param>,
>because you MUST set @name, and then @name overrides anything on the
>target:
><param name="must-set-this" conref="...."/>
>
>The token lets you use conref, pulling in @name from the referenced
>element, which is probably what you wanted:
><param name="-dita-use-conref-target" conref="...."/>
>
>The DITA 1.2 spec says "The value -dita-use-conref-target is available on
>enumerated attributes and may also be specified on other attributes."
>The intent was that you can use this anywhere, and the attribute does not
>have to be required (also knowing that a constraint or specialization can
>make any attribute required).
>
>So the question is - what about for attributes where we declare that the
>attribute must be something else? Specifically, what about @cols on
><tgroup>, which is defined as a number - specifically, the number of
>columns in the group? Or what about the date attributes
> on <created>, <copyryear>, and so on, which are defined as using a date
>format? 
>
>Some attributes are defined in a way that "-dita-use-conref-target" is
>simply not valid, so I'm not asking about those - this is only a question
>about attributes where the grammar files technically allow it, but the
>prose in the spec explicitly sets a range
> of values that does not include -dita-use-conref-target. In those cases,
>is this token implicitly allowed?
>
>Thanks,
>
>Robert D Anderson
>IBM Authoring Tools Development
>Chief Architect, DITA Open Toolkit (http://www.dita-ot.org/)
>
>