Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (nike.apache.org: domain of vfunstein@gmail.com designates
 209.85.215.48 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAG0hk_8=VK-mKuH=qTKZeyrxxtKNi8+=2fJUMekZtdq5do4rAQ@mail.gmail.com>
References: 
 <CAG0hk_8y0MDN4un6UuZGNuBkZo=ziWmreCWZo39xAU=WZcz2pQ@mail.gmail.com>
	<c67e84a3-96e7-4e23-81e2-acff98be8db1@email.android.com>
	<CAG0hk_8=VK-mKuH=qTKZeyrxxtKNi8+=2fJUMekZtdq5do4rAQ@mail.gmail.com>
Date: Tue, 9 Oct 2012 17:16:56 -0700
Message-ID: 
 <CALr4Hzrni94j9vOs0SwasY9LjsAezcJfLnixuAH+V6Vyt_Fu8w@mail.gmail.com>
Subject: Re: Lucene 4.0-BETA : MultiReader isCurrent openIfChanged
From: Vitaly Funstein <vfunstein@gmail.com>
To: java-user@lucene.apache.org
Content-Type: multipart/alternative; boundary=f46d042dfe995a745104cba95f88

--f46d042dfe995a745104cba95f88
Content-Type: text/plain; charset=ISO-8859-1

You have probably figured it out by now, but my suggestion would be to use
SearcherManager the way it is documented for maintaining a searcher backed
by an NRT reader.

On Sun, Aug 26, 2012 at 2:03 PM, Mossaab Bagdouri <bagdouri_mossaab@yahoo.fr
> wrote:

> Thanks for the quick reply.
>
> I've changed my code to the following. The problem now is that the
> MultiReader doesn't seem to get closed. In fact, the number of open files
> (returned by "lsof | grep index/ | wc -l") keeps increasing whenever the
> IndexWriter adds new documents until the webapp crashes. Am I missing
> something?
>
> private IndexSearcher getIndexSearcher() {
>         try {
>             boolean refresh = false;
>             if (is == null) {
>                 refresh = true;
>             } else {
>                 MultiReader ir = ((MultiReader) is.getIndexReader());
>                 if (ir.getRefCount() == 0) {
>                     refresh = true;
>                 } else {
>                     for (DirectoryReader dr : (List<DirectoryReader>)
> ir.getSequentialSubReaders()) {
>                         if (!dr.isCurrent()) {
>                             refresh = true;
>                             ir.close();
>                             break;
>                         }
>                     }
>                 }
>             }
>
>             if (refresh) {
>                 DirectoryReader readers[] = new DirectoryReader[2];
>                 for (int i = 0; i < 2; i++) {
>                     readers[i] =
> DirectoryReader.open(MyFSDirectories.get(i));
>                 }
>                 is = new IndexSearcher(new MultiReader(readers, true));
>             }
>         } catch (Exception e) {
>             e.printStackTrace();
>         }
>         return is;
>     }
>
> Regards,
> Mossaab
>
> 2012/8/26 Uwe Schindler <uwe@thetaphi.de>
>
> > 1. getRefCount()
> > 2. No
> >
> >
> >
> > Mossaab Bagdouri <bagdouri_mossaab@yahoo.fr> schrieb:
> >
> > >Hi,
> > >
> > >I've just migrated my webapp from Lucene 3.6 to 4.0-BETA. My 2 indexes
> > >are
> > >updated every couple of minutes by a batch. The webapp searcher needs
> > >to
> > >get refreshed whenever this happens. In 3.6, I was doing it this way:
> > >
> > >private IndexSearcher getIndexSearcher() {
> > >        try {
> > >            if (is == null || !is.getIndexReader().isCurrent()) {
> > >                if (is != null) {
> > >                    is.close();
> > >                }
> > >                IndexReader readers[] = new IndexReader[2];
> > >                for (int i = 0; i < 2; i++) {
> > >                 readers[i] = IndexReader.open(MyFSDirectories.get(i));
> > >                }
> > >                is = new IndexSearcher(new MultiReader(readers));
> > >            }
> > >        } catch (Exception e) {
> > >            e.printStackTrace();
> > >        }
> > >        return is;
> > >    }
> > >
> > >Now that the API changed in 4.0-BETA. I'm doing it this odd way:
> > >
> > >private IndexSearcher getIndexSearcher() {
> > >        try {
> > >            boolean refresh = false;
> > >            if (is == null) {
> > >                refresh = true;
> > >            } else {
> > >                MultiReader ir = ((MultiReader) is.getIndexReader());
> > >                List<DirectoryReader> list = (List<DirectoryReader>)
> > >ir.getSequentialSubReaders();
> > >                try {
> > >                    if (!list.get(0).isCurrent() ||
> > >!list.get(1).isCurrent()) {
> > >                        refresh = true;
> > >                    }
> > >                } catch (Exception e) {
> > >                    e.printStackTrace();
> > >                    refresh = true;
> > >                }
> > >
> > >            }
> > >            if (refresh) {
> > >                if (is != null) {
> > >                    ((MultiReader) is.getIndexReader()).close();
> > >                }
> > >                DirectoryReader readers[] = new DirectoryReader[2];
> > >                for (int i = 0; i < 2; i++) {
> > >                    readers[i] =
> > >DirectoryReader.open(MyFSDirectories.get(i));
> > >                }
> > >                is = new IndexSearcher(new MultiReader(readers));
> > >            }
> > >        } catch (Exception e) {
> > >            e.printStackTrace();
> > >        }
> > >        return is;
> > >    }
> > >
> > >The problem is that sometimes isCurrent() returns an exception
> > >indicating
> > >that the underlying IndexReader is closed.
> > >
> > >My questions are:
> > >1. How can I verify of the IndexReader is open before calling
> > >isCurrent();
> > >2. Is there a better way to deal with MultiReaders?
> > >
> > >Regards,
> > >Mossaab
> >
> > --
> > Uwe Schindler
> > H.-H.-Meier-Allee 63, 28213 Bremen
> > http://www.thetaphi.de
>

--f46d042dfe995a745104cba95f88--