I have been reading with interest the various requests by different people
for new SWISH options.
One nice option for the HTTP method would be to have a spider which allowed
one to index each site to a different depth, and to a different depth
depending on whether one were in the domain of that site or not. I have
written my own programs in Perl to give me this functionality, using the
SWISH spider program to get web pages and their links for me. Incorporated
into this is a handler for pages with frames (I have already uploaded a
copy of the spider with this frames support, but it doesn't sit that well
with the spidering process as it currently stands because of the
restrictions on out-of-the-domain links.) I would interested to know how
people felt generally about the use of SWISH for HTTP as opposed to
FILESYSTEM indexing. I have only ever used SWISH for HTTP indexing.
I do have a question as well:
Is there an option to get SWISH to output ALL of its indexed files? I
realise that for some this would be a really big list, but it could be very
useful in certain circumstances, e.g. if you wanted to filter the results
by some other method using your own software.
Received on Tue Aug 1 06:43:58 2000