making a case for maps in xquery

[Michael Kay]

On 22/07/2011 11:35, James Fuller wrote:
> I have been using Saxon and MarkLogic implementation of map:map for
> the past few months and am convinced of their usefulness.
>

I agree. I'm sure this feature will become commonplace even if we don't 
manage to get it into the spec. But it would be really nice to have it 
standard.

I don't think there's much serious opposition to the idea in principle. 
Just some people want to do it "better" than the current proposed spec, 
some people want time to chew over it, some people want to ship XQuery 
3.0 now with no new features - all the usual difficulties.

For XQuery one of the use cases is as an equivalent to xsl:key in XSLT - 
it provides some "hand optimization" possibilities. How useful that is 
depends of course how smart the automatic optimization is in your 
preferred product.

Michael Kay
Saxonica