Hello Erick, I was trying to optimise the searching. Basically my data is like field1 has less no of docs matching compared to field2, which has larget sets.
So if search goes by order to order, then i can make field1 to be search first, (by making order of boolean query such )and from thr the docs result set could be searched for field2. So number of docs to be matched will be less. I am not using filters or any caching since the search is unique everytime acc to parameter, and its using boolean query or is the searching takes place like searches each doc for field1="abc" and field2="def" then get resultset I hope I am clear in explaining my query. thanks, Suman On Thu, 1 Apr 2010 08:42:16 -0700, Erick Erickson <erickerick...@gmail.com> wrote: > Why do you care? By that I mean "what problem are you trying to solve" > (See "The XY problem at http://people.apache.org/~hossman/). The reason > I'm asking here is that very often, when people ask this kind of question > without providing background, they're trying the wrong approach to solve > a problem. > > Unless you're asking for theoretical reasons, I almost guarantee that > trying > to solve any problem in Lucene by changing the order of the clauses isn't > going to do you much good in the long run, there'll *always* be variant > n+1... > > Best > Erick > > On Thu, Apr 1, 2010 at 2:21 AM, <suman.hol...@zapak.co.in> wrote: > >> >> Query I >> >> its written "to do a "search within search", so that the second search is >> constrained by the results of the first query" >> we can use boolean query. >> So doesn't it mean the order of query ll be preserved >> >> give me an simple example of how the docs get searched in lucene. >> 10 docs with 3 fields to b searched >> so ll it search for field1 for all 10 docs..and the result set docs ll b >> then searched for field2....and so far... >> or is it something else.. >> >> >> >> thanks, >> Suman >> >> On Thu, 1 Apr 2010 10:05:01 +0100, Ian Lea <ian....@gmail.com> wrote: >> >> Query I >> >> Does the order of query play role in searching >> >> example:doc has fields >> >> rollno(pk), name, marks >> >> >> >> Query : marks=90&rollno=2&name=abc >> >> >> >> Query :rollno=2&name=abc&marks=90 >> >> >> >> which query processing will be more efficient. >> >> is it work like search doc field by field , it will look for doc >> >> having >> >> rollno=2 then from thos will look for name abc and from the selected >> docs >> >> will again look for marks param . >> > >> > I don't think it makes any difference. >> > >> > >> >> Query II >> >> which is more faster . >> >> a) adding one more multivalued field into current index(large) >> >> b) or creating a separate index for the field and searching it as a >> >> subindex . >> >> My main apprehensions is in regard to large size of index >> > >> > Probably your first option. >> > >> > >> > -- >> > Ian. >> > >> > --------------------------------------------------------------------- >> > To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org >> > For additional commands, e-mail: java-user-h...@lucene.apache.org >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org >> For additional commands, e-mail: java-user-h...@lucene.apache.org >> >> --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
