mlibrary / solr_analyzed_string

Create a string field using another fieldType's analysis chain

solr_analyzed_string -- fill a solr string based on a textField definition

There's a real advantage to doing data munging within the solr process, since you're guaranteed that index- and query-time analysis will be identical.

Unfortunately, solr TextField types aren't run through the analyzer when doing range queries, and getting an exact phrase match (as opposed to an exact subphrase match) can be difficult.

This code allows you to create a fieldType based on a solr.TextField, and then create a solr string field with the results of running the input through that set of filters.

Usage

First, put the latest release .jar somewhere your schema will find it.

Then add something like this to your schema.

<!-- A type defined with keywordTokenizer that does the work -->
  <fieldType name="browse_key_analysis" class="solr.TextField" positionIncrementGap="10000">
    <analyzer>
      <tokenizer class="solr.KeywordTokenizerFactory"/>
      <filter class="solr.ICUFoldingFilterFactory"/>
      <filter class="solr.PatternReplaceFilterFactory" pattern="^the\s+" replacement="" replace="all"/>
      <filter class="solr.PatternReplaceFilterFactory" pattern="[:\-]+" replacement=" " replace="all"/>
      <filter class="solr.PatternReplaceFilterFactory" pattern="[\p{P}\p{Sm}\p{So}]" replacement="" replace="all"/>
      <filter class="solr.TrimFilterFactory"/>
      <filter class="solr.PatternReplaceFilterFactory" pattern="\p{Z}" replacement=" " replace="all"/>
    </analyzer>
  </fieldType>

  <!-- A string type that will run the analyzer associated with the given 
  fieldType -->
  <fieldType name="browse_key" class="com.billdueber.solr.schema.AnalyzedString" fieldType="browse_key_analysis"/>

  <!-- A field that uses that type -->
  <field name="author_browse" type="browse_key" indexed="true" stored="true"/>