5

Is it possible to highlight whole query terms? f.e. when I ask for "United States" I want to get:

<em>United States</em>

and not:

<em>United</em> <em>States</em>

I've searched the whole Internet for an answer, used all combinations of hl.mergeContiguous, hl.usePhrasesHighlighter and hl.highlightMultiTerm parameters and still cannot make it work.

my query is:

http://localhost:8983/solandra/idxPosts.proj350_139/select?q=post_text:"Janusz Palikot"&hl=true&hl.fl=post_text&hl.mergeContiguous=true&hl.usePhrasesHighlighter=true&hl.highlightMultiTerm=true

the answer is:

...
<arr name="post_text"><str>Tag: <em>janusz</em> <em>palikot</em> - Sowiniec: "Sowiniec"</str></arr>
...

my "post_text" field is:

<field name="post_text" type="text" indexed="true" stored="true" termVectors="true" termPositions="true" termOffsets="true" required="true" />

my "text" type is:

<fieldType name="text" class="solr.TextField">
    <analyzer type="index">
        <tokenizer class="solr.StandardTokenizerFactory" />
        <filter class="solr.TrimFilterFactory" />
        <filter class="solr.LowerCaseFilterFactory" />
        <filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_pl.txt" />
        <filter class="solr.ReversedWildcardFilterFactory" />
    </analyzer>
    <analyzer type="query">
        <tokenizer class="solr.StandardTokenizerFactory" />
        <filter class="solr.TrimFilterFactory" />
        <filter class="solr.LowerCaseFilterFactory" />
        <filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_pl.txt" />
    </analyzer>
</fieldType>

I also tried to use FastVectorHighlighter with hl.useFastVectorHighlighter=true but encountered an error:

Problem accessing /solandra/idxPosts.proj350_139/select. Reason:

    -6

java.lang.ArrayIndexOutOfBoundsException: -6
    at lucandra.TermFreqVector.getOffsets(TermFreqVector.java:224)
    at org.apache.lucene.search.vectorhighlight.FieldTermStack.<init>(FieldTermStack.java:100)
    at org.apache.lucene.search.vectorhighlight.FastVectorHighlighter.getFieldFragList(FastVectorHighlighter.java:175)
    at org.apache.lucene.search.vectorhighlight.FastVectorHighlighter.getBestFragments(FastVectorHighlighter.java:166)
    at org.apache.solr.highlight.DefaultSolrHighlighter.doHighlightingByFastVectorHighlighter(DefaultSolrHighlighter.java:509)
    at org.apache.solr.highlight.DefaultSolrHighlighter.doHighlighting(DefaultSolrHighlighter.java:376)
    at org.apache.solr.handler.component.HighlightComponent.process(HighlightComponent.java:116)
    at org.apache.solr.handler.component.SearchHandler.handleRequestBody(SearchHandler.java:194)
    ...

Can you help me, please?

Tom_LK
  • 113
  • 1
  • 2
  • 7
  • did you ever find a way to mark the whole phrase (and only that) in the Solr results? ... I know this is years later, but I still cannot find a working answer – mibm Dec 06 '17 at 13:54

2 Answers2

7

For the phrase highlight, there is a Jira stilling waiting to get through to the Solr code.

Jayendra
  • 52,349
  • 4
  • 80
  • 90
-1

Chech the solr doc for this, there is parameter hl, set this true.