A while ago I remember someone mentioning that it was possible to
use webinator as a catalog search engine. I don't remember hearing
a result to that problem, but it is similar to a problem I am trying
to solve.
I am designing a searchable book catalog for a book store. Their
current inventory includes only 80K titles but they are hoping to
increase their catalog to contain 500K - 1million records.
Each record contains 5 fields (isbn, author, title, price, subject)
and is no longer than 70 characters. The data is presented to me as
a flat file:
ISBN AUTHOR TITLE PRICE SUBJECT CODE
ISBN AUTHOR TITLE PRICE SUBJECT CODE
etc.
etc.
Users should be able to search by author or words in title.
That we allow them to search for words in the title means that
a regular SQL implementation doesn't work very well, as we
end up with a query that has a leading and trailng wildcard.
Is it possible to use webinator to index the flat file
and then when a user queries webinator about that database,
have it spit out the appropriate line(s) that contain the
query term?
My investigation of the tables that gw creates when it indexes
suggests that it isn't possible, however, since I remember
someone suggesting that it can be done, I guess I'm missing something.
Additionally, one of the webinator demo sites, (taxonomy.com) produces
output very different than the standard webinator output, so I must
be missing something.
Thanks very much for your help,
David