RE: [dita] inherited attribute values in a DITA relationship table

["Grosso, Paul" <pgrosso@ptc.com>]

Data point from the CALS spec, ref:

http://www.oasis-open.org/specs/tr9503.html#c38ab3b8

then scroll down to 7.2.6 rowsep where it says:

 

Default: IMPLIED, from the closest element in the inheritance path:

 

    * <row> then
    * <colspec> then
    * <tgroup> then
    * <table>.

 

So Michael's suggested relative order of rows/columns/tables agrees with that in CALS.

 

paul

---

From: Michael Priestley [mailto:mpriestl@ca.ibm.com]
Sent: Tuesday, 2006 October 24 09:54
To: Grosso, Paul
Cc: dita@lists.oasis-open.org
Subject: RE: [dita] inherited attribute values in a DITA relationship table

I would propose that more specific metadata overrides more general metadata, and that the order of general-to-specific would go:

table
column
row
cell
topicref

Michael Priestley
IBM DITA Architect and Classification Schema PDT Lead
mpriestl@ca.ibm.com
http://dita.xml.org/blog/25

"Grosso, Paul" <pgrosso@ptc.com>

10/23/2006 12:40 PM

To	<dita@lists.oasis-open.org>
cc
Subject	RE: [dita] inherited attribute values in a DITA relationship table

This issue/question remains open.
 
paul

---

From: Grosso, Paul [mailto:pgrosso@ptc.com]
Sent: Wednesday, 2006 October 11 09:05
To: dita@lists.oasis-open.org
Subject: [dita] inherited attribute values in a DITA relationship table

In the DITA 1.1 Architecture Spec, on page 21 toward the end of the section on “Inheritance of attributes and metadata” it says:
 
            Tables can be particularly useful for attribute and metadata management, since they
can be applied to entire columns or rows as well as individual cells.
 
Elements within a cell inherit default attribute values much as the same elements outside of a table.  And within a table the order of inheritance goes from elements in a relcell to the relcell to the relrow to the reltable and then on up to the map element.  What isn’t clear is where inheritance by column (relcolspec) fits into the above list and how conflicts between relrow and relcell values are resolved.
 
We need to figure out the answer and make it clear in the spec.