> Date: Tue, 14 Dec 2004 14:59:37 +0100
> From: Hedzer Westra <[log in to unmask]>
> > recall that the "word" structure is the default for the relation "="
> > "=" means "=/cql.word"
> Hmm, this is (for me) not exactly the same as what the CQL context set
> "= is used:
> For word adjacency, when the term is a list of words. That is to
> say that the words appear in that order with no others
> Otherwise, for exact equality of value"
> from the term 'exact' I don't get the impression that cql.word is
> implied for = on single terms..
The context-set prose is insufficiently explicit. What we're
describing here is the _former_ case, a list of words -- it's just
that you're interested in the special case where the list has only a
single element. In other words, the dichotomy is not between single
and multiple words, it's between words (any number of them) and other
things (such as integers).
> [lengthy URI & word separation discussion by Rob & Mike]
> Well I guess I stirred up something that hasn't been totally agreed
> upon yet.. I have to finalize my SRU code so I don't have time to
> wait for the finalization of this thread. You've already seen my
> Adlib Base Profile document: I just documented how I implemented it]
Yup, that's the way to go for now.
>>> You mean SRW being Z39.50 all over - something like a bulky,
>>> difficult to implement protocol?
>> Well, I reject that description of Z39.50.
> Apologies if I offended you ;-)
No offence, you have merely been misled :-)
> The XPath sorting approaches sorting from the 'output' (i.e. record
> schemas) side of things whereas I would approach sorting from the
> 'input' (i.e. context set & index) side.
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ Since Oscar Wilde's quote "There is no sin except stupidity"
is clearly stupid, we should probably consider it sinful in
deference to him.
Listen to free demos of soundtrack music for film, TV and radio