[Zope] cataloging SQL data

[George Runyan]

can anyone give me a quickie HOWTO on cataloging SQL data?  any gotchas
would be appreciated.  I havent started on the proof of concept (on NT ...
ick!) but we are interested in cataloging a large set of data and using the
Zope catalog to do the searching.  
the main reason we want the zope catalog is to give hand back a result set
that is ranked by relevance (any have any experience w/ relevance ranking of
search results?)

what would the pro's or con's be for using Zope as the main method to access
data inside a SQLServer?  i.e. I'm thinking of Siteserver's catalog that has
some form of linguistical summation (prob. not the right tech
term)/abstraction of free-text fields.  
and siteserver kinda does some relevance ranking... 

in summary: I have a proof of concept to show how relevance would be
incorporated by search results coming from Zope's catalog mechanism which
would index sqlserver data.  digicool has said they have done this w/ one of
their clients.  I have not seen much information on this in the public
sector.  I will most likely need this done by beginning of next week.  

questions FOR YOU: have you used zope's catalog for this?  why wouldn't you?
how does relevance play apart in the zCatalog?  is there max. size a catalog
should be?  I'm not python literate enough to read the src and understand-
nor do i have the time (imagine that!)

many thanks,
george runyan		grunyan@emerging.com