You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by Felipe Lobo <fe...@jusbrasil.com.br> on 2009/09/30 23:25:51 UTC
Implement SpanScorer on 2.9 lucene lib!
Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
spanscorer but the constructor is protected.
I looked in the javadoc of lucene and saw 2 subclasses of it
(PayloadNearQuery.PayloadNearSpanScorer,
PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
Using this classes is the best way of implementing spanscorer??
I was using queryscorer but after update to the new lib the highlither
stopped hightlithing my wildcard queries, and i'm rewriting.
Thanks,
--
Felipe Lobo
www.jusbrasil.com.br
Re: Implement SpanScorer on 2.9 lucene lib!
Posted by Mark Miller <ma...@gmail.com>.
Felipe Lobo wrote:
> Here's the code:
> ------------------------------------------------------
> Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new
> QueryScorer(query));
>
> highlighter.setTextFragmenter(new SimpleFragmenter(999999999));
>
> String fieldName = "Title";
>
> String text = document.getField(fieldName).stringValue();
>
> TokenStream tokenStream = analyzer.tokenStream(fieldName, new
> StringReader(text));
>
> String markedResult = highlighter.getBestFragments(tokenStream, text, 3,
> "...");
> ----------------------------------------------------
> and when i sysout the markedResult shows this:
>
> "Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
>
> when i put a shorter number, like 4 characther, it makes the highlight, but
> this "100930027253" it don't.
>
Thanks a lot - I'll check it out and get back to you.
> the name is realy TermQueryScorer or is QueryTermScorer(i found that in the
> package)??
>
Sorry! Thats what happens when I trust my memory ;)
Its QueryTermScorer.
> Thanks.
>
>
> On Thu, Oct 1, 2009 at 10:44 AM, Mark Miller <ma...@gmail.com> wrote:
>
>
>> Felipe Lobo wrote:
>>
>>> Hi, thanks for the answer but it didn't work.
>>> I stopped rewriting the query and used the queryscorer but it don't
>>> highlight.
>>> The part of the query i'm doing wildcard is the number part, like this:
>>> "HC 100930027253"
>>> The HC is hightlighted but the numbers aren't:
>>> "Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
>>>
>>>
>> Hmm ... I don't know. Should work. Can you post some code?
>>
>>> And of what lib is TermQueryScorer? I check out on lucene javadoc and
>>>
>> didn't
>>
>>> see.
>>>
>>>
>> Its in the same package as the QueryScorer, in the Highlighter contrib.
>>
>>> Thanks!
>>>
>>> On Wed, Sep 30, 2009 at 6:38 PM, Mark Miller <ma...@gmail.com>
>>>
>> wrote:
>>
>>>
>>>> Felipe Lobo wrote:
>>>>
>>>>
>>>>> Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
>>>>> spanscorer but the constructor is protected.
>>>>> I looked in the javadoc of lucene and saw 2 subclasses of it
>>>>> (PayloadNearQuery.PayloadNearSpanScorer,
>>>>> PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
>>>>> Using this classes is the best way of implementing spanscorer??
>>>>> I was using queryscorer but after update to the new lib the highlither
>>>>> stopped hightlithing my wildcard queries, and i'm rewriting.
>>>>>
>>>>> Thanks,
>>>>>
>>>>>
>>>>>
>>>> Checkout contrib CHANGES.
>>>>
>>>> The SpanScorer is now the QueryScorer - if you want to highlight
>>>> wildcard queries, you should no longer rewrite on your own.
>>>>
>>>> If you want the old behavior, use TermQueryScorer instead.
>>>>
>>>> --
>>>> - Mark
>>>>
>>>> http://www.lucidimagination.com
>>>>
>>>>
>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>>>> For additional commands, e-mail: java-user-help@lucene.apache.org
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>> --
>> - Mark
>>
>> http://www.lucidimagination.com
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: java-user-help@lucene.apache.org
>>
>>
>>
>
>
>
--
- Mark
http://www.lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: Implement SpanScorer on 2.9 lucene lib!
Posted by Felipe Lobo <fe...@goshme.com>.
Here's the code:
------------------------------------------------------
Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new
QueryScorer(query));
highlighter.setTextFragmenter(new SimpleFragmenter(999999999));
String fieldName = "Title";
String text = document.getField(fieldName).stringValue();
TokenStream tokenStream = analyzer.tokenStream(fieldName, new
StringReader(text));
String markedResult = highlighter.getBestFragments(tokenStream, text, 3,
"...");
----------------------------------------------------
and when i sysout the markedResult shows this:
"Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
when i put a shorter number, like 4 characther, it makes the highlight, but
this "100930027253" it don't.
the name is realy TermQueryScorer or is QueryTermScorer(i found that in the
package)??
Thanks.
On Thu, Oct 1, 2009 at 10:44 AM, Mark Miller <ma...@gmail.com> wrote:
> Felipe Lobo wrote:
> > Hi, thanks for the answer but it didn't work.
> > I stopped rewriting the query and used the queryscorer but it don't
> > highlight.
> > The part of the query i'm doing wildcard is the number part, like this:
> > "HC 100930027253"
> > The HC is hightlighted but the numbers aren't:
> > "Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
> >
> Hmm ... I don't know. Should work. Can you post some code?
> > And of what lib is TermQueryScorer? I check out on lucene javadoc and
> didn't
> > see.
> >
> Its in the same package as the QueryScorer, in the Highlighter contrib.
> > Thanks!
> >
> > On Wed, Sep 30, 2009 at 6:38 PM, Mark Miller <ma...@gmail.com>
> wrote:
> >
> >
> >> Felipe Lobo wrote:
> >>
> >>> Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
> >>> spanscorer but the constructor is protected.
> >>> I looked in the javadoc of lucene and saw 2 subclasses of it
> >>> (PayloadNearQuery.PayloadNearSpanScorer,
> >>> PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
> >>> Using this classes is the best way of implementing spanscorer??
> >>> I was using queryscorer but after update to the new lib the highlither
> >>> stopped hightlithing my wildcard queries, and i'm rewriting.
> >>>
> >>> Thanks,
> >>>
> >>>
> >> Checkout contrib CHANGES.
> >>
> >> The SpanScorer is now the QueryScorer - if you want to highlight
> >> wildcard queries, you should no longer rewrite on your own.
> >>
> >> If you want the old behavior, use TermQueryScorer instead.
> >>
> >> --
> >> - Mark
> >>
> >> http://www.lucidimagination.com
> >>
> >>
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> >> For additional commands, e-mail: java-user-help@lucene.apache.org
> >>
> >>
> >>
> >
> >
> >
>
>
> --
> - Mark
>
> http://www.lucidimagination.com
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
>
--
Felipe Lobo
www.jusbrasil.com.br
Re: Implement SpanScorer on 2.9 lucene lib!
Posted by Mark Miller <ma...@gmail.com>.
Felipe Lobo wrote:
> Hi, thanks for the answer but it didn't work.
> I stopped rewriting the query and used the queryscorer but it don't
> highlight.
> The part of the query i'm doing wildcard is the number part, like this:
> "HC 100930027253"
> The HC is hightlighted but the numbers aren't:
> "Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
>
Hmm ... I don't know. Should work. Can you post some code?
> And of what lib is TermQueryScorer? I check out on lucene javadoc and didn't
> see.
>
Its in the same package as the QueryScorer, in the Highlighter contrib.
> Thanks!
>
> On Wed, Sep 30, 2009 at 6:38 PM, Mark Miller <ma...@gmail.com> wrote:
>
>
>> Felipe Lobo wrote:
>>
>>> Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
>>> spanscorer but the constructor is protected.
>>> I looked in the javadoc of lucene and saw 2 subclasses of it
>>> (PayloadNearQuery.PayloadNearSpanScorer,
>>> PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
>>> Using this classes is the best way of implementing spanscorer??
>>> I was using queryscorer but after update to the new lib the highlither
>>> stopped hightlithing my wildcard queries, and i'm rewriting.
>>>
>>> Thanks,
>>>
>>>
>> Checkout contrib CHANGES.
>>
>> The SpanScorer is now the QueryScorer - if you want to highlight
>> wildcard queries, you should no longer rewrite on your own.
>>
>> If you want the old behavior, use TermQueryScorer instead.
>>
>> --
>> - Mark
>>
>> http://www.lucidimagination.com
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: java-user-help@lucene.apache.org
>>
>>
>>
>
>
>
--
- Mark
http://www.lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
Re: Implement SpanScorer on 2.9 lucene lib!
Posted by Felipe Lobo <fe...@goshme.com>.
Hi, thanks for the answer but it didn't work.
I stopped rewriting the query and used the queryscorer but it don't
highlight.
The part of the query i'm doing wildcard is the number part, like this:
"HC 100930027253"
The HC is hightlighted but the numbers aren't:
"Habeas Corpus <B>HC</B> 100930027253 ES 100930027253"
And of what lib is TermQueryScorer? I check out on lucene javadoc and didn't
see.
Thanks!
On Wed, Sep 30, 2009 at 6:38 PM, Mark Miller <ma...@gmail.com> wrote:
> Felipe Lobo wrote:
> > Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
> > spanscorer but the constructor is protected.
> > I looked in the javadoc of lucene and saw 2 subclasses of it
> > (PayloadNearQuery.PayloadNearSpanScorer,
> > PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
> > Using this classes is the best way of implementing spanscorer??
> > I was using queryscorer but after update to the new lib the highlither
> > stopped hightlithing my wildcard queries, and i'm rewriting.
> >
> > Thanks,
> >
> Checkout contrib CHANGES.
>
> The SpanScorer is now the QueryScorer - if you want to highlight
> wildcard queries, you should no longer rewrite on your own.
>
> If you want the old behavior, use TermQueryScorer instead.
>
> --
> - Mark
>
> http://www.lucidimagination.com
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
>
--
Felipe Lobo
www.jusbrasil.com.br
Re: Implement SpanScorer on 2.9 lucene lib!
Posted by Mark Miller <ma...@gmail.com>.
Felipe Lobo wrote:
> Hi, i updated my lucene lib to 2.9.0 and i'm trying to instanciate the
> spanscorer but the constructor is protected.
> I looked in the javadoc of lucene and saw 2 subclasses of it
> (PayloadNearQuery.PayloadNearSpanScorer,
> PayloadTermQuery.PayloadTermWeight.PayloadTermSpanScorer).
> Using this classes is the best way of implementing spanscorer??
> I was using queryscorer but after update to the new lib the highlither
> stopped hightlithing my wildcard queries, and i'm rewriting.
>
> Thanks,
>
Checkout contrib CHANGES.
The SpanScorer is now the QueryScorer - if you want to highlight
wildcard queries, you should no longer rewrite on your own.
If you want the old behavior, use TermQueryScorer instead.
--
- Mark
http://www.lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org