Date:         Tue, 18 Jun 2002 14:38:38 +0200
Reply-To:     "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Sender:       "Z39.50 Next-Generation Initiative" <[log in to unmask]>
From:         Rob Koopman <[log in to unmask]>
Subject:      Re: Betr.: Re: result set model for srw
Comments: To: [log in to unmask]
Content-Type: text/plain; charset="iso-8859-1"

Robert Sanderson wrote:

>bla.cgi?set1=bib1&index1=title&field1=war+and+peace&bool1=and&...

As soon as queries get complicated you need postfix in your example.
Apart from that the number of name/value pairs will tend to get out of hand.
In practice it is much simpler to work with a string. I would stick with
CQL.

for example we currently do:
.../DB=2.41/SET=3/CMD?ACT=SRCH&SRT=RLV&TRM=aut%3D%22veen%2C+theo%22+and+all%
3D%22library%22

(the "path" part of the URL is more or less the "current page" context.
(makes links shorter) we probably do not want them for SRU)

This looks horrid but URL escaping is not our problem it is "routine" for
web
users and taken care of by browsers or other tools. (chinese url encoded
queries are special fun).

Rob Koopman