Re: search question
Ming-Li <[email protected]> Thu, 23 Dec 2004 05:14:42 +0800
| Newsgroups | gmane.mail.zoe.general |
|---|---|
| Message-ID | <[email protected]> |
Hi, On Tue, 21 Dec 2004 14:03:54 +0100 Zoe <[email protected]> wrote: > Yep... the default Lucene analyzer tries to recognize "common" text > entities (e.g. a host name) and then index them as one token. Thanks for the explanation. > > If this is a feature, it seems counter-intuitive to me. > > Oh, well... What I means is, people usually refer to the database service as "lexis". Its "official documents" (like the one granting your extension) might call themselves lexis.com, www.lexis.com, lexisnexis, lexis-nexis, etc. I thought searching for "lexis" would catch them all. Well, now at least I can use lexis* to catch them all. Wait a minute, will "lexis*" catch "www.lexis.com"? Or do I have to use "*lexis*"? Tried it just now, and it seems "lexis*" will also catch "www.lexis.com"? But that's kind of inconsistent, isn't it? Or is Lecene smart enough to get rid of the "www." prefix? In that case, why not the ".com" ending? And how about other kind of prefix, such as "web2" as westlaw like to use? Guess I'll have to read up on Lucene more. The problem is, whatever trick I learn by doing that, I have to "train" my wife, too. ;-) Ask her to put an asterisk after a word is easy enough, but tricks more complicated than that would be difficult. I guess that's the price of a powerful search tool. -- Best regards, Ming-Li ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/