[ZOOM] Re: Required Resultsets Redux

Mike Taylor mike at miketaylor.org.uk
Sat Sep 7 18:40:13 CEST 2002

> Date: Mon, 2 Sep 2002 14:27:35 +0100 (BST)
> From: Robert Sanderson <azaroth at liverpool.ac.uk>
> Back in January, we agreed that I had a legit need to have access to
> the underlying resultset name of a resultset object.

Yes -- because of your need to specify the result-set name in a
relevance-feedback search.

> I don't see any mention of an interface to this in the API though.

Jeez, lady, give the guy a chance!  :-)

> I suspect it should be:
> resultSet.resultSetName()

Or we could make it masquerade as a "magic option":

	String rsName = rs.getOption("name");

I sort of prefer that because it feels more lightweight, and doesn't
dignify what is, truthfully, rather a fringey kind of operation with a
whole section of the AAPI to itself, which it doesn't really merit.

I am open to be persuaded otherwise, though.

> There also needs to be (as Sebastian has pointed out) a way to turn
> on or off result set naming. I suggest:
> connection.set("namedResultSets", 1)

That would turn it _on_, right?  But if we are to present any
meaningfully OO model to the world, then "on" must be the default
(i.e. your old RS objects don't just vanish into the ether every time
you do some unrelated option like a new search).  So pedagogically,
this would be better expresses by the example:

	connection.set("namedResultSets", 0); # don't use names

Seb, please could supply some terse yet crystal-clear prose that I can
paste into the AAPI to explain why we need to make our nice, clean
specification support this brain-damage?

> A third interface needed is to explicitly set resultset on a query
> object to be submitted, such that whenever the query is executed it
> creates a resultset of that name rather than allowing the library to
> randomly generate one for it:
> query.set("resultSet", "NAMEOFRESULTSET")

Remind me again why this one is desirable?

> References:
> http://www.o-r-g.org/~cheshire/cgi-bin/zoomarchive.cgi?format=long&fieldidx1=DOCID&fieldcont1=166
> http://www.o-r-g.org/~cheshire/cgi-bin/zoomarchive.cgi?format=long&fieldidx1=DOCID&fieldcont1=187
> http://www.o-r-g.org/~cheshire/cgi-bin/zoomarchive.cgi?format=long&fieldidx1=DOCID&fieldcont1=190
> http://www.o-r-g.org/~cheshire/cgi-bin/zoomarchive.cgi?format=long&fieldidx1=DOCID&fieldcont1=191

I appreciate your provision of these URLs, but I am working at home on
a laptop with extremely intermittent and slow internet connection, so
I can't really use then.  In any case, since I _personally_ detest all
three of these "enhancements", I am quite happy to rely on other
people to craft the accompanying prose rather than mangling their old
words myself :-)

 _/|_	 _______________________________________________________________
/o ) \/  Mike Taylor   <mike at miketaylor.org.uk>   www.miketaylor.org.uk
)_v__/\  "Network: Any thing reticulated or decussated, at equal
	 distances, with interstices between the intersections." --
	 Samuel Johnson's "Dictionary of the English Language"

More information about the ZOOM mailing list