Re: Search for documents where field does not exist?

> On Friday 17 June 2005 22:05, [EMAIL PROTECTED] wrote: > >> I checked the docs missing >> from the result set and found that those documents actually contain the >> word 'information' but lucene is not returning them anyway. Is it a >> known issue? Any way around? > > See > http://lucene.apache.

Re: Search for documents where field does not exist?

On Friday 17 June 2005 22:05, [EMAIL PROTECTED] wrote: > I checked the docs missing > from the result set and found that those documents actually contain the > word 'information' but lucene is not returning them anyway. Is it a > known issue? Any way around? See http://lucene.apache.org/java/doc

About the field of PhraseQuery

hi, I spent an hour today to make my field-name feed correctly into my phrase-query. A ridiculous bug of mine. Debugging experience seemed to indicate that the field was the field of the first term sent. But it struck me that the word field simply does not appear in the javadoc of PhraseQuer

Re: Search for documents where field does not exist?

Are the missing documents relatively long? If so, check this: http://lucene.apache.org/java/docs/api/org/apache/lucene/index/IndexWriter.html#maxFieldLength Otis --- [EMAIL PROTECTED] wrote: > I just installed lucene for full text search,since mysql was too slow > at > that. But I see, for same

Re: Search for documents where field does not exist?

I just installed lucene for full text search,since mysql was too slow at that. But I see, for same query (e.g. 'information') lucene is returning a slightly fewer number of documents than mySQL. I checked the docs missing from the result set and found that those documents actually contain the word

discrepancy in the number of hits

I just installed lucene for full text search,since mysql was too slow at that. But I see, for same query (e.g. 'information') lucene is returning a slightly fewer number of documents than mySQL. I checked the docs missing from the result set and found that those documents actually contain the word

Re: Search for documents where field does not exist?

I dont believe you can. What you can do is index a NULL term. That is, a term that will not occur naturally in your index. Then you can do such a search: field:NULL This requires you to index NULL when you know the field is going to be empty. C --- Dan Armbrust <[EMAIL PROTECTED]> wrote: >

Search for documents where field does not exist?

I'm pretty sure the answer is no.. but I'll check with the guru's anyway... In my collection of documents, I have a non-tokenized field that only occurs 0 or 1 time per document. It is possible to create a query so that a documents would be returned if (field == "some value" OR field does not

RE: IndexHTML.java location

Right Now, I'm just using the compiled version that I downloaded. By Default URL, I mean the location of the indexed files. I have the sample web project (index.jsp) on server A, and my indexed files are on server B. Everything works till I click on the link in the results.jsp page. It looks one d

RE: IndexHTML.java location

IndexHTML.java is located in the demo jarfile which is named differently depending on whether you built it yourself with Ant or just downloaded it ready-jarred. What do you mean by "default URL"? The location of the document-set to index, the location of the document-set to search (presumably thro

IndexHTML.java location

Not sure if this is the right address, to request this kind of help, so if it isn't please point me else where. Basically I think I have an understanding of how Lucene works, in general. I believe I'm at a point where I need to change the "default" url, so I was planning to make the change in the

Re: Wordperfect file parser

On Jun 17, 2005, at 3:15 AM, Diego Manilla Suárez wrote: Hi, does anyone know if a Wordperfect file (*.wpd) parser exist for Java? The OpenOffice SDK provides this capability (from what I've read). It's not a pure Java solution, but has a Java API. Erik ---

Re: Performance with multi index

Each time you close indexWriter ram cached documents are flushed to disk. In case you open and close it per document many one-document segments are created on disk. minMergeDocs sets number of documents that are cached in ram. JM Tinghir wrote: do you keep your indexWriter open all the time

Re: Performance with multi index

> do you keep your indexWriter open all the time during process? I think that might be the real cause. And as it reopen it all the time, the mergeFactor isn't used at all I guess... I'll try to modify that. Thanks. Jean-Marie Tinghir -

Wordperfect file parser

Hi, does anyone know if a Wordperfect file (*.wpd) parser exist for Java? Thanks - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]