Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Subject: Highlighter IOOBE with modified HyphenationCompoundWordTokenFilter
From: =?utf-8?Q?Markus_Jelsma?= <markus.jelsma@openindex.io>
To: =?utf-8?Q?java-user=40lucene=2Eapache=2Eorg?=
  <java-user@lucene.apache.org>
Date: Thu, 4 Oct 2012 13:41:16 +0000
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Message-Id: <zarafa.506d91fc.37b8.5c673de60d932e43@mail.openindex.io>

Hi,

I've modified the HyphenationCompoundWordTokenFilter to emit less subtokens because the original filter can emit all kinds of subtokens that have a very different meaning on their own. I've modified it so no overlapping subtokens are emitted and no subtokens are emitted that can be found within another subtoken. I've also modified it to force that the generated subtokens comprise the original token and if they don't forget the subtokens. It also doesn't return the original token anymore, the original filter produces a duplicate of the original input token. For example: verzekeringmaatschappij now becomes verzekering and maatschappij and not verzekeringmaatschappij, ver, zeker, verzeker, zekering, ringmaat, maat and more.

But it seem that i have done something wrong because my modified version sometimes causes the Highlighter to throw the following IOOBE:

java.lang.StringIndexOutOfBoundsException: String index out of range: -14
        at java.lang.String.substring(String.java:1937)
        at org.apache.lucene.search.vectorhighlight.BaseFragmentsBuilder.makeFragment(BaseFragmentsBuilder.java:172)
        at org.apache.lucene.search.vectorhighlight.BaseFragmentsBuilder.createFragments(BaseFragmentsBuilder.java:138)
        at org.apache.lucene.search.vectorhighlight.FastVectorHighlighter.getBestFragments(FastVectorHighlighter.java:186)
        at org.apache.solr.highlight.DefaultSolrHighlighter.doHighlightingByFastVectorHighlighter(DefaultSolrHighlighter.java:571)
        at org.apache.solr.highlight.DefaultSolrHighlighter.doHighlighting(DefaultSolrHighlighter.java:401)
        at org.apache.solr.handler.component.HighlightComponent.process(HighlightComponent.java:136)
        at org.apache.solr.handler.component.SearchHandler.handleRequestBody(SearchHandler.java:214)
        at org.apache.solr.handler.RequestHandlerBase.handleRequest(RequestHandlerBase.java:129)
        at org.apache.solr.core.SolrCore.execute(SolrCore.java:1750)
        at org.apache.solr.servlet.SolrDispatchFilter.execute(SolrDispatchFilter.java:455)
        at org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:276)
        at org.eclipse.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1337)
        .....

Anyone to point me in the right direction=3F I've checked the LIA book on how to manipulate the tokenstream and thought it should be alright. My analysis tests also yield good results, nothing strange to be found. Or could it be an error in the highlighter that only now shows up=3F

Thanks,
Markus

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