Author: Ben
Email: [EMAIL PROTECTED]
Message:
Sure... in the docs there is an example query like:
SELECT url.url, count(*) as c
FROM url,dict
WHERE dict.url_id=url.rec_id
AND dict.word IN ('some','word')
GROUP BY url.url
ORDER BY c DESC
which will return all urls with ANY of the search terms in the IN() condition, the
equivalent of an "OR search term" search. Is it possible to do the equivalent of an
"AND search terms" search, i.e., ALL the words are found in every document? Or is that
the same as adding something like WHERE c >= (num_of_search_terms)?
By the way, I did get the indexer not to crash. The keys seemed to be setting
MaxDocSize very large (8M), since I have 50,000 primary key values. Also, I realized
in tweaking my HTDBDoc, I managed to make the query use an unindexed field on a join,
so it was very slow. That's fixed now, and so far seems pretty good! I'm working on
building the frontend now...
Ben
Reply: <http://www.mnogosearch.org/board/message.php?id=4533>
___________________________________________
If you want to unsubscribe send "unsubscribe general"
to [EMAIL PROTECTED]
