Re: Problems with "tagged" and "non tagged" text

2010-12-11 Thread Celso Fontes
Dear Erick, Sorry i am using really "AND" operator, i wrote wrong in email (i am very tired)... But..Follow the 'main' part of code: Document document = new Document(); String path = file.getCanonicalPath(); document.add(new Field("title", path,

Re: Problems with "tagged" and "non tagged" text

2010-12-11 Thread Erick Erickson
Unless you provide details on how you are indexing these documents, it's pretty hard to help. It's also hard to reconcile your statement that OR is the default operator with the results you posted, the '+' all over the place really points to AND as the default. There's no magic in Lucene that wil

Problems with "tagged" and "non tagged" text

2010-12-11 Thread Celso Fontes
Hi, i have the same text in two files: TXT file: http://pastebin.com/u9Rd9VVA (X)HTM file: http://pastebin.com/ydHmTQZ8 And i running this Question: APC (adenomatous polyposis coli) actin assembly with OR operator and SNOWBALL Analyser results in: +content:apc +(+content:ad

Re: RE: maybe I hit a bug of Term ?

2010-12-11 Thread Ian Lea
1. Please post new questions in a new thread. 2. Try a less specific search, maybe "lucene TREC". -- Ian. On Sat, Dec 11, 2010 at 11:25 AM, Duy DINH wrote: > Hi Lucene community, > This is not a novel question but I tried to search on Google the query: > > "how to index TREC collection with

Re: RE: maybe I hit a bug of Term ?

2010-12-11 Thread Duy DINH
Hi Lucene community, This is not a novel question but I tried to search on Google the query: "how to index TREC collection with Lucene" I found nothing about that except some of your answers about the papers using Lucene for indexing TREC collection, and I think that I have to write code for inde