lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Otis Gospodnetic <otis_gospodne...@yahoo.com>
Subject RE: lucene and ejb applications
Date Fri, 20 Aug 2004 13:40:10 GMT
Option b) sounds simpler and sufficient to me.  I don't see why you
would need to involve RMI for something as simple as this.  I use
something similar to your b) option for some indices behind
http://www.simpy.com/ .  I don't store IndexSearcher in the servlet
context, though - I just have some logic like this:


    /**
     * Returns an instance of {@link IndexDescriptor} for the given
     * <code>indexID</code>, which must represent an absolute file
     * path to the index directory.
     * <p/>
     * This method caches {@link IndexDescriptor}s in a LRU Map and
     * first tries to retrieve them from there.
     * <p/>
     * If the specified index has been changed since the the last time
     * it was used, its {@link Searcher} is reloaded.
     *
     * @param indexID the full path to the index directory
     * @return an instance of {@link IndexDescriptor}
     * @throws SearcherException if the given index cannot be accessed
     */
    IndexDescriptor getUserSearcherIndexDescriptor(String indexID)
        throws SearcherException
    {
        File indexDir = validateIndex(indexID);
        IndexDescriptor indexDescriptor =
getIndexDescriptorFromCache(indexDir);

        try
        {
            // if this is a known index
            if (indexDescriptor != null)
            {
                // if the index has changed since this Searcher was
created, make a new Searcher
                long currentVersion =
IndexReader.getCurrentVersion(indexDir);
                if (currentVersion > indexDescriptor.lastKnownVersion)
                {
                    indexDescriptor.lastKnownVersion = currentVersion;
                    indexDescriptor.searcher = new
LuceneUserSearcher(indexDir);
                }
            }
            // if this is a new index
            else
            {
                indexDescriptor = new IndexDescriptor();
                indexDescriptor.indexDir = indexDir;
                indexDescriptor.lastKnownVersion =
IndexReader.getCurrentVersion(indexDir);
                indexDescriptor.searcher = new
LuceneUserSearcher(indexDir);
            }
            return cacheIndexDescriptor(indexDescriptor);
        }
        catch (IOException e)
        {
            throw new SearcherException("Cannot open index: " +
indexDir, e);
        }
    }

IndexDescriptor is a simple struct-like class.


Otis


--- Rupinder Singh Mazara <rsmazara@ebi.ac.uk> wrote:

> hi erik
> 
>  thanks for the warning and the code.
>  Let me re-phrase the question,
> 
>  i have a index generated by lucene, i need to have the search
> capabilty
>  to have a high availabilty. What solutions would be the most optimal
> 
>  Currentlly i have two senarions in mind
>   a) setup a RMI based app. that on start-up initializes a
> IndexSearcher
> object
>      and waits for invocation of a method like Vector
> executeQuery(Query )
> 
>   b) create a web based app(jsp/servlet or struts)  that initialises
> the
> IndexSearcher object, and stores in the servletContext on
> intialization, and
> all request invoke the Hits search(Query q)
> 
>   with senario a)  i can have more control over updates, insert, and
> deletes
>   where as with  senario b) has higher availabilty
> 
>  I want to create and store the IndexSearcher object, during
> initailization
> to save on
>  mutlitple open and reads. once updates are ready signal can be sent
> to
> block further searches while the updates are integrated into the
> existing
> index.
> 
> 
> 
> >-----Original Message-----
> >From: Erik Hatcher [mailto:erik@ehatchersolutions.com]
> >Sent: 20 August 2004 11:13
> >To: Lucene Users List
> >Subject: Re: lucene and ejb applications
> >
> >
> >What would be the best way?  Use Lucene outside of EJB.  It's quite
> >silly to make such a decision "purely due to a policy decision" when
> >the technicalities of it show that it is an unwise decision.
> >
> >You're going to navigate Hits through a session bean?  And as you
> said,
> >the EJB spec says not to use file I/O from EJB's.  That is a good
> >recommendation if you are distributing your system across servers
> and
> >replication is occurring - if another call to a session bean occurs
> and
> >ends up on a different server, then the file handle is lost.
> >
> >I violate the spec in my JavaDevWithAnt project and have one mode
> where
> >I have a stateless session bean returning search results:
> >http://www.ehatchersolutions.com/JavaDevWithAnt - but I definitely
> do
> >not recommend it.  It works when you are in a single-server
> >environment.
> >
> >In summary - EJB and Lucene are not a good mix - don't force it just
> to
> >be buzzword compliant.
> >
> >	Erik
> >
> >
> >On Aug 20, 2004, at 4:32 AM, Rupinder Singh Mazara wrote:
> >
> >> hi all
> >>
> >>    purely due to a policy decision, we would like to host our
> lucene
> >> search
> >> application , in a j2ee container, preferable by means of a ejb.
> >> Since access to java.io is restricted by the ejb specification,
> what
> >> would
> >> be the best way to create desgin the application ?
> >>   i have taken a look at ejindex@sf.net but it my relies on mbeans
> and
> >> not a
> >> session bean
> >>   does any one have pointers or samples that can be looked at



---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-user-help@jakarta.apache.org


Mime
View raw message