I think that Larry is right. I think that our problem could be resolved if
we examine what CQL really is. The current definition indicates "SRW
specifies string queries. The query language, called CQL ("Common Query
Language") is a human-readable-string query-representation based loosely on
CCL ....." If we change that to say:
"SRW queries are expressed in CQL. The query language, called CQL ("Common
Query Language") is a human-readable-string query-representation based
loosely on CCL ....."
CQL is a string because it includes the whole query along with its
attributes. I don't think that this should mandate that the search argument
is also a string. The search argument could contain just one single word.
The location of the truncation symbol defines the sort of truncation.
Examples:
Tasmania? - right truncation
Tasmania? tiger? - right truncation
Tasmania? tiger - truncation 104
Tasmanian tiger? - truncation 104
Janifer Gatenby
Consultant OCLC PICA ITC
Schipholweg 99, 2300 AW Leiden, The Netherlands
+ 31 71 524 65 00
+ 31 71 522 31 19 (fax)
[log in to unmask]
-----Original Message-----
From: Ray Denenberg [mailto:[log in to unmask]]
Sent: Thursday, 16 May 2002 22:34
To: [log in to unmask]
Subject: Re: cql index definitions
"Larry E. Dixson" wrote:
> Ray,
> 1) keyword
> 2) keyword with right truncation
> ....
> Can't they be represented as (where "x" is a particular index like
> "title"):
> 1) bath.xWord (no truncation character required in search term)
> 2) bath.xWord (include "?" at end or term -- the "*" is a bonus)
But Bath specifies right truncation -- truncation value 1. We would have to
use a
truncation value of 104 if we put the truncation character in the string.
If we specify a truncation value different than Bath, can we call it a Bath
search?
Then (1) and (2) above would be the same index, representing two different
bath
searches. (We lose the correspondence of srw indexes to Bath searches.) Is
this a
good thing to do?
--Ray
|