org.apache.lucene.collation

Class ICUCollationKeyAnalyzer

  • All Implemented Interfaces:
    Closeable, AutoCloseable


    public final class ICUCollationKeyAnalyzer
    extends Analyzer

    Configures KeywordTokenizer with ICUCollationAttributeFactory.

    Converts the token into its com.ibm.icu.text.CollationKey, and then encodes the CollationKey directly to allow it to be stored as an index term.

    WARNING: Make sure you use exactly the same Collator at index and query time -- CollationKeys are only comparable when produced by the same Collator. com.ibm.icu.text.RuleBasedCollators are independently versioned, so it is safe to search against stored CollationKeys if the following are exactly the same (best practice is to store this information with the index and check that they remain the same at query time):

    1. Collator version - see Collator#getVersion()
    2. The collation strength used - see Collator#setStrength(int)

    CollationKeys generated by ICU Collators are not compatible with those generated by java.text.Collators. Specifically, if you use ICUCollationKeyAnalyzer to generate index terms, do not use CollationKeyAnalyzer on the query side, or vice versa.

    ICUCollationKeyAnalyzer is significantly faster and generates significantly shorter keys than CollationKeyAnalyzer. See http://site.icu-project.org/charts/collation-icu4j-sun for key generation timing and key length comparisons between ICU4J and java.text.Collator over several languages.