I haven't tried multi field query parser. Though about the usage of adding
multiple values while indexing, here's a code snippet that might help.
--snip starts--
IndexWriter iw = new IndexWriter(...);
Document d = new Document();
doc.add(new Field("field", new FileReader(f11));
doc.add(new Field("field", new FileReader(f12));
iw.addDocument(doc);
--snip ends--
--
Anshum Gupta
Naukri Labs!
http://ai-cafe.blogspot.comThe facts expressed here belong to everybody, the opinions to me. The distinction is yours to draw............ On Tue, Nov 10, 2009 at 2:50 PM, Wenhao Xu <xuwenhao2...@gmail.com> wrote: > Thanks. Yes, I mean multi valued fileds, but I am still confused how to use > it. > > BTW, I also found another class MultiFieldQueryParser. If I don't use > multi > valued fields, it seems I can use this class to compose a query over > multiple fields. But is there difference of the result returned with these > two methods? Or the performance difference? > > cheers, > W. > > On Tue, Nov 10, 2009 at 12:55 AM, Anshum <ansh...@gmail.com> wrote: > > > If you are talking about multi valued fields, the answer is yes. > > You may create multiple field objects with the same name and add to the > > same > > document. > > It would lead to adding the values to the same field name for the > document. > > > > -- > > Anshum Gupta > > Naukri Labs! > > http://ai-cafe.blogspot.com > > > > The facts expressed here belong to everybody, the opinions to me. The > > distinction is yours to draw............ > > > > > > On Tue, Nov 10, 2009 at 2:14 PM, Wenhao Xu <xuwenhao2...@gmail.com> > wrote: > > > > > Hi, guys, > > > I have such a problem: > > > I have lots of files. In these files, two of them are related to > each > > > other and should be deemed as a whole (There is a map file to map them > > > together). So they are somewhat like a set of pairs of files: <f11, > f12>, > > > <f21, f22> , <f31, f32> ... . For a keyword search, the result should > > also > > > be like <fi1, fi2>,.... Their score should be computed together. > > > > > > As far as I know, I could use "doc.add(new Field("field1", new > > > FileReader(f11)));", and "doc.add(new Field("field2", new > > > FileReader(f12)))" > > > for each file respectively. But in this way, when searching, these two > > > fields (field1 and field2) can not be searched together in a single > > search. > > > As I said, these two files should be deemed as a whole and returned > > > together. So I am wondering, could I map one filed to more than one > > value, > > > somewhat like doc.add(new Field("field", new FileReader(f11), new > > > FileReader(f12)); > > > > > > Is there some way to do this? Or could you give me some suggestions > how > > > to solve this problem? > > > > > > Thanks, > > > W. > > > > > > -- > > > ~_~ > > > > > > > > > -- > ~_~ >
