Gilles, list members, I upgraded but am still getting the strange results. Do you (or anyone else) have any ideas?
Thanks! >From: Gilles Detillieux <[EMAIL PROTECTED]> >To: [EMAIL PROTECTED] (From Laura) >CC: [EMAIL PROTECTED] >Subject: Re: [htdig] Quoting Search String Behavior >Date: Thu, 19 Sep 2002 11:17:29 -0500 (CDT) > >According to From Laura: > > I am having a hard time understanding how HTDIG handles quoted search > > strings. My specific question is whether quotes indicate a next-to or > > ordered relationship. For instance, for my specific boolean query: > > > > #A (usval86 or usval34 or usval150) and "tablespace usage" > > Result: Returns pages as expected with the phrase "tablespace usage". > > > > So to further test quote behavior I issued the boolean query: > > > > #B (usval86 or usval34 or usval150) and "usage name tablespace" > > Result: No matches. (Note that each page in #A had the term "name" in >it). > > > > So, I decide that the quotes indicate order. But the following tests > > surprised me. This ALL query: > > > > #C "Unable Extend Objects to" > > Result: Returned a page with these terms ordered this way: > > "Objects Unable to Extend Notes" > > > > However, this query: > > #D "Extend to Unable Objects" > > Result: No matches. > > > > Am I correct to assume that there are some complicated algorithms behind >the > > scenes and there is not a black and white behavior for quotes? I was >unable > > to find any recent documentation. > > > > Note I am using version 3.2.0b3. > > > > Also a big thank you to the developers who created this product. I am >sure > > it requires a lot of your time. We all love it! > >Glad you like it. There are LOTS of bugs in 3.2.0b3, including some >with the query parser and phrase support! I'd recommend upgrading to the >latest snapshot of 3.2.0b4, from http://www.htdig.org/files/snapshots/, >and then see if you still get odd behaviour. One thing I do know is >that words smaller than minimum_word_length or words in bad_words are >not considered significant in your phrase, and therefore not needed for >a match. > >-- >Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]> >Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/ >Dept. Physiology, U. of Manitoba Winnipeg, MB R3E 3J7 (Canada) > > >------------------------------------------------------- >This sf.net email is sponsored by:ThinkGeek >Welcome to geek heaven. >http://thinkgeek.com/sf >_______________________________________________ >htdig-general mailing list <[EMAIL PROTECTED]> >To unsubscribe, send a message to ><[EMAIL PROTECTED]> with a subject of unsubscribe >FAQ: http://htdig.sourceforge.net/FAQ.html _________________________________________________________________ Join the world�s largest e-mail service with MSN Hotmail. http://www.hotmail.com ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ htdig-general mailing list <[EMAIL PROTECTED]> To unsubscribe, send a message to <[EMAIL PROTECTED]> with a subject of unsubscribe FAQ: http://htdig.sourceforge.net/FAQ.html
