Re: [dita] More attribute discrepancies - conref attributes on map

[Robert D Anderson <robander@us.ibm.com>]

I actually noticed the exact same issue as described below on the <anchor> element, so I did a more complete check, and found the issue for all of these elements:
map
bookmap
anchor
resourceid

Oddly, our other map specializations are all correct in the spec, listing the full conref attribute group. I think this indicates that the best fix is to update the spec topics for the 4 topics above, so that they match both the shipped DTD/XSD and the existing specializations. These four were overlooked when updating the spec for 1.2 due to the unusual way that they reused the definition of conref -- nearly every other element previously referenced both @id and @conref as a set, but these four all had unique definitions of @id.

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit (http://dita-ot.sourceforge.net/)

Robert D Anderson---12/17/2013 14:55:36---I've actually gone through quite a few of these tables, and have found fewer issues than I expected,

From: Robert D Anderson/Rochester/IBM@IBMUS
To: dita <dita@lists.oasis-open.org>,
Date: 12/17/2013 14:55
Subject: [dita] More attribute discrepancies - conref attributes on map
Sent by: <dita@lists.oasis-open.org>

---

I've actually gone through quite a few of these tables, and have found fewer issues than I expected, but I've run into a small group today that require input.

According to the specification, we define @conref on the map element. It's been there from 1.0, even if it's not used very often:
http://docs.oasis-open.org/dita/v1.2/os/spec/langref/map.html

In 1.2 nearly every element that uses conref was updated to use the new conref group of attributes: conref, conrefend, conaction, conkeyref. This was done for <map> as well, in both the DTD and the XSD. @conrefend doesn't make sense on map, but the full group is used for consistency.

Getting repetitive here, but the specification is officially canonical. In this case, given that the attributes were added to both DTD and XSD, and that having them simplifies both the definition and the spec, I think I'd suggest "fixing" the spec to reference the full group.

Thanks,

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit (http://dita-ot.sourceforge.net/)