On Tue, 29 Jul 2003, LeVan,Ralph wrote:
> But, the discussion (at the link) points out the need to separate the term
> and the index, rather than confuse things with the full CQL term, including
> relation. But, the relation implies the structure. So, we'll need that
> somehow.
Yes. Trawling through the previous discussion, it came down to:
Matthew originally proposing use of searchClause.
Rob starting with index and term, then switching to searchClause, but
wondering what the other relations were to be interpreted as.
Ralph suggesting that the only meaningful distinction was keyword vs
string.
Rob suggesting that stem was a valid relation modifier but not being very
sure about it.
What happens about the indexes of dimensioned terms (range/area/volume)
and scan? For example the geo searches? I think we'll need Mike's term
describing relation qualifiers so that the server knows how to interpret
the term at all?
> Now, the real question is; what is the "clause" in your ScanRequest, Rob?
A search clause string in cql. xClause being XCQL.
I'm not very sure about the TermReference structure... I think the idea
was that the database would be an SRW database that respected Mike's ZThes
indexSet... but are people happy with that restriction, otherwise we need
to give a full search clause (which in my mind is too heavy to attach to
every single term)?
The other option suggested was that the server URL etc should be in the
Explain record in a named <link> ... that would cut out having to repeat
it ad nauseum on every response.
Rob
> > -----Original Message-----
> > From: Robert Sanderson [mailto:[log in to unmask]]
> > Sent: Tuesday, July 29, 2003 12:09 PM
> > To: [log in to unmask]
> > Subject: 1.1: Scan
> >
> >
> > To refresh, what we were thinking on Scan:
> >
> > http://listserv.loc.gov/cgi-bin/wa?A2=ind0301&L=zng&F=&S=&X=49
> > 9A0A17B318735D41&[log in to unmask]&P=1930
> >
> > and subsequent.
> >
> >
> > Issues I can find that are outstanding:
> >
> > > Things which would be useful which Theo (IIRC) brought up:
> > Some sort of
> > > marker for where the term would be if it's not present, or
> > that the term
> > > is the one used from the request.
> >
> > Which would also fill:
> >
> > > If we were to contemplate something equivalent to step size, I
> > > would propose a collapsed element and leave the server to
> > interpret its
> > > meaning.
> >
> > and:
> >
> > > 241 Scan: Beginning or end of term list
> > > Is 241 a surrogate diagnostic? If so, should we have some
> > facility for
> > > including a marker for the beginning/end of the list as well as the
> > > supplied term?
> >
> > Either, we can have a list of mixed object types (Please No, this will
> > play hell with implementations, I fear) or we need to allow
> > some sort of
> > diagnostic to be attached to the term structure as a surrogate term.
> >
> > eg:
> >
> > <term>
> > <diagnostics>
> > <diagnostic>
> > <code>150</code>
> > <details>Start of TermList</details>
> > </diagnostic>
> > </diagnostics>
> > </term>
> >
> > Then the list of surrogate diagnostics for scan would be:
> > Start of TermList, End of TermList, Requested Term Position.
> >
> > This means that all of the parameters for <term> need to be
> > optional, but
> > our prose says 'you must return at least one of value and diagnostics'
> >
> >
> > My current structure (without the above per term diagnostics):
> >
> > ScanRequest:
> > clause String
> > xClause xcql:SearchClause type
> > numberOfEntries Integer
> > responsePosition Integer
> >
> >
> > ScanResponse:
> > terms List of ScanTerm type
> > diagnostics Diagnostic type
> > echoedRequest ScanRequest type (for SRU)
> >
> > ScanTerm:
> > value String (value of the term)
> > frequency Integer (number of occurences)
> > displayTerm String (term to display instead of value)
> > references List of TermReference type (pointers to thesauri)
> >
> > TermReference:
> > thesaurus String (name.. eg 'lcsh' 'unesco')
> > server String (URI to SRW/U server)
> > useTerm String (term to use in search if different)
--
,'/:. Dr Robert Sanderson ([log in to unmask])
,'-/::::. http://www.o-r-g.org/~azaroth/
,'--/::(@)::. Special Collections and Archives, extension 3142
,'---/::::::::::. Nebmedes: telnet: nebmedes.o-r-g.org 7777
____/:::::::::::::. WWW: http://nebmedes.o-r-g.org:8000/
I L L U M I N A T I
|