Re: [dita] Resolution on unresolvable key maps

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

Hi,

I just want to clarify one item from #2:
> 2. A key reference for which no definition is available MUST always issue
> an error.

I think I heard on the call that this was limited to key references where there is no @href for fallback linking - is that the case? Otherwise it seems that using that @href fallback is literally an error condition, where processors must use that @href to handle the error.

I thought from the distant past that this case was legal, and not defining the key simply removed the link. But it sounds like maybe it's an error condition, resolved by removing the link?
<link keyref="unresolved"/>

Similarly, in the more common case, I did not think this was a situation where we MUST generate an error, though many want some note of it:
<link keyref="undefined" href="" uri..."/>

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

<dita@lists.oasis-open.org> wrote on 12/09/2014 11:51:02:

> From: Chris Nitchie <chris.nitchie@oberontech.com>
> To: "dita@lists.oasis-open.org" <dita@lists.oasis-open.org>
> Date: 12/09/2014 11:51
> Subject: [dita] Resolution on unresolvable key maps
> Sent by: <dita@lists.oasis-open.org>
>
> 1. If a map contains a reference to another DITA map with scope="peer" and
> keyscope set to some value, and that peer map cannot be loaded by the
> processor for whatever reason, the processor SHOULD issue a warning.
>
> 2. A key reference for which no definition is available MUST always issue
> an error. If that key reference specifies a scope prefix matching the
> scope name on a peer mapref referring to unavailable content, processors
> MAY issue additional verbiage stating that the key might be defined in a
> currently-unavailable peer map.
>
> 3. If a key reference specifies a scope prefix matching the scope name on
> a peer mapref referring to unavailable content, but the local map's key
> space also includes a definition for that key, the key resolves to the
> local map definition, and no message should be issued by the processor.
> That is, it is treated as a normal, resolvable key reference. This is true
> regardless of the relative precedence of the local definition and the peer
> mapref.
>
> 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>
>  
>  
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>