Skip to main content.
home | support | download

Back to List Archive

Re: Index only meta names

From: Bill Moseley <moseley(at)not-real.hank.org>
Date: Wed Aug 09 2000 - 17:31:14 GMT
At 06:08 PM 08/09/00 +0200, jmruiz@boe.es wrote:

>> It might be nice to be able to say "Search in all the meta names" like
>> 
>>   swish-e -w *=(swish or "best search engines") and have it look in all
the 
>> meta names.
>> 
>> I'd also like to be able to do this:
>>   swish-e -w meta1,meta2,meta4=(just words in those meta tabs)
>> 
>
>You can do it if your files are in this way...
><metaT>
><meta1>bla bla</meta1>
><meta2>bla bla</meta2>
></metaT>

That's a cool feature, indeed.  But it doesn't allow you to pick any random
combination of meta names.

My example of
  swish-e -w meta1,meta2,meta4=(word)

Can be done currently like this:

  swish-e -w meta1=(word) or meta2=(word) or meta4=(word)

So I don't see the need for real the change.  Makes programming the search
query a bit easier the first way, but not that much easier.

But, it might be faster for swish, though, to be able to use the first
method.  In the second example, my guess is that swish does three searches
and then merges the results(?)  Using the first example above maybe swish
would only have to make one pass and not merge?

Another idea to throw out would be to have a reserved word meta name
"default", so you could say:

   swish -w default=(word) 

and say only search in the areas not defined by the METANAME configuration.

Then you could do thing like

   swish -w default,title,subject=(word)

To search in the default area, and in the two meta named areas.

BTW -- I'd be interested in seeing the indexing speed gains by Swish 2.0
from people with large data sources.  I've only got about 6000 records on
my machine and indexing time dropped from about 7 minutes to something like
20 or 30 seconds.  Is it as dramatic on larger data sets?


Bill Moseley
mailto:moseley@hank.org
Received on Wed Aug 9 13:34:59 2000