andy rosbrook scribbled on 8/24/05 8:23 AM:

> and all this data would be contained within a single index. I was just 
> wondering weather it would be possible to search this index and return 
> results from only http://www.blah.com/sales.html? Or would a new index for 
> each page need to be created?

check the docs, esp SWISH-CONFIG. You just need to set swishdocpath as a 
MetaName, and then search like:

  swish-e -w swishdocpath=http://www.blah.com/sales.html and someword

make sure you include . and / as WordCharacters so that the entire URL gets 
indexed as a single word.

> 
> I suppose it depends on the structure of the indexes? I've tried to figure 
> out the structure but when i try and open the index.swish-e its not 
> readable..how are the indexes structured, something id like to know!
> 

the swish-e indexes are binary for speed. a text editor won't help. use the -T 
option to dump info from the index; that's the best way.
-- 
Peter Karman  .  http://peknet.com/  .  peter(at)not-real.peknet.com