lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Hostetter <hossman_luc...@fucit.org>
Subject RE: BooleanQuery.TooManyClauses on MultiSearcher
Date Thu, 15 Jun 2006 17:43:29 GMT

: Incidentally, I'm getting BooleanQuery.TooManyClauses when I search on
: "james", but I don't when I search on "James". Surely  the number of clauses
: isn't dependent on the number of hits?!

not the numebr of hits -- just hte number of terms in your index that
start with the prefix.

: However, I know that "fred" is relatively uncommon in my index and "neil" is
: relatively common and yet "fred" is getting the BooleanQuery.TooManyClauses
: and "neil" isn't. Does that make sense?
:
: Should the actual term used in a PrefixQuery effect the number of clauses?

yes .. the Term used in the PrefixQUery is just a convinient holder for a
fieldname and a term value prefix -- what matters is how many terms in
that field start with that prefix.  if "james*" causes a problem, but
"James*" doesn't then it sounds like your indexing analyzer is case
sensative and you have a lot more lowercase values starting with james
then upercase values starting with James .. if "fred*" causes a problem
but "neil*" doesn't then you probably have a lot more terms that start
with "fred" then you do that start with "neil" -- it doesn't matter if
"neil@foo" is the value more documents then the total number of docs that
contain any value starting with "fred", what matters is how many unique
values there are starting with "fred"




-Hoss


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org


Mime
View raw message