Re: TMQL - Proposal for a %base directive

Robert Barta <[email protected]> Sat, 7 May 2011 09:07:18 +0200
Newsgroups gmane.text.xml.xtm.general
Message-ID <[email protected]>
On Tue, May 03, 2011 at 10:20:40AM +0200, Lars Heuer wrote:

> IMO TMQL has define somewhere how to resolve local identifiers. It has
> to it regardless of the %base proposal. Section 5.1 defines:
> 
>        """
>        [Context map]
> 
>        [...}
>        All item references and navigation steps are interpreted
>        relative to this map.
>        """
> 
> I think that this definition is rather vague, maybe intentionally.

Obviously it does not say, whether this is at "compilation time" or at
"evaluation time". Mainly because these are rather artificial phases
anyway. All that implies that item references only make sense in the
presence of a map being queried.

If you ask for a %base directive, then you simply allow/force queries
to be compiled. I - for myself - would be happy to have it open,
i.e. using some late binding (when the map is clear).

On a side note: Before you ask for %directives, %pragmas, %controls,
%regimes, %contexts or whatever, it might make sense to think whether
Topic Maps as mechanism can actually cover that somehow. It is amazing
how you get, when you are willing to give up old concepts from the
programming language world.

\rho