Re: [dita] Question from keys review: addressing non-topicref elements in a map

[Chris Nitchie <chris.nitchie@oberontech.com>]

I'd settle for removing the word 'non-topicref' so that it reads, "and elements within maps." If you bind a map to a key

<mapref href="" keys="mapKey"/>

Then to reference any element in that map via key, keyref="mapKey/elementId". For example:

<topicref conkeyref="mapKey/topicrefId"/>

-or-

<navtitle conkeyref="mapKey/someTitle"/>

You'd use conkeyref instead of keyref on the former to pull in the whole branch, as opposed to referencing whatever content is referenced by that topicref. I don't think you could use conkeyref="someRandomKey" to effectively conref the key-defining topicref.

Chris

P.S. I'm on vacation this week and only checking e-mail occasionally, but I wanted to contribute my $0.02 while my thoughts were fresh.

---

From: dita@lists.oasis-open.org <dita@lists.oasis-open.org> on behalf of Robert D Anderson <robander@us.ibm.com>
Sent: Thursday, May 7, 2015 11:58 AM
To: DITA TC
Subject: [dita] Question from keys review: addressing non-topicref elements in a map

 

DITAWeb URL:
http://ditaweb.com/oasis-dita/#/Collaborate/00160583-FA_1/DB00074180/Using%20keys%20to%20address%20DITA%20elements

Topic (title as it appears in the review): "Using keys to address DITA elements"

Text in question, about the syntax for using keyref:
For references to non-topic elements within topics and non-topicref elements within maps, the value of the @keyref attribute is a key name, a slash ("/"), and the ID of the target element (for example, keyref="topic-key/some-element-id".)

Original comment thread:

• [Chris Nitchie] Non-topicrefs in maps? Whenever I need to conkeyref something into I do it by putting it in the of a topic bound to a key and conkeyref it from the topic. Does the OT permit referencing elements in a map via key?
  And what about topicref elements? Why wouldn't you need an element ID to reference those? A reference to a key is a reference to the content referenced by the effective key definition, not the topic ref itself.
  All of this is a long-winded way of saying I think we need to remove the phrase, "and non-topicref elements within maps".
• [Robert Anderson] This is word-for-word what appeared in DITA 1.2, apart from changing "solidus" to "slash", and adding "for example" to the example: http://docs.oasis-open.org/dita/v1.2/os/spec/archSpec/using_keys_to_address_dita_elements.html
  But - I think I agree with Chris's comment. I don't know how it would be used for non-topicref element within a map. This needs to go to the TC.
• [Eliot Kimber] It is possible (if not particularly useful or recommend) to address any element in any map by ID (because all the elements take IDs). So the spec has to account for the case. That's why it was in 1.2 and needs to remain.

My own additional commentary:
For elements within topics, we always have the #topic/element syntax for linking with @href. So it's clear to me that when a key name resolves to a topic, "key/element" means the XML element with id="element" inside that topic. It maps easily to the @href syntax of #[variable-topic]/element.

When referencing any element in a map (a topicref, a navtitle, a phrase inside link text), the syntax is file.ditamap#element or just #element. If we do in fact allow a key syntax of keyref="key/mapelement", I don't know how it resolves.

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