org.apache.lucene.analysis.phonetic
Class PhoneticFilterFactory

java.lang.Object
  org.apache.lucene.analysis.util.AbstractAnalysisFactory
      org.apache.lucene.analysis.util.TokenFilterFactory
          org.apache.lucene.analysis.phonetic.PhoneticFilterFactory

All Implemented Interfaces:

ResourceLoaderAware

public class PhoneticFilterFactory
extends TokenFilterFactory
implements ResourceLoaderAware

Factory for PhoneticFilter. Create tokens based on phonetic encoders from Apache Commons Codec.

This takes one required argument, "encoder", and the rest are optional:

encoder

required, one of "DoubleMetaphone", "Metaphone", "Soundex", "RefinedSoundex", "Caverphone" (v2.0), or "ColognePhonetic" (case insensitive). If encoder isn't one of these, it'll be resolved as a class name either by itself if it already contains a '.' or otherwise as in the same package as these others.

inject

(default=true) add tokens to the stream with the offset=0

maxCodeLength

The maximum length of the phonetic codes, as defined by the encoder. If an encoder doesn't support this then specifying this is an error.

 <fieldType name="text_phonetic" class="solr.TextField" positionIncrementGap="100">
   <analyzer>
     <tokenizer class="solr.WhitespaceTokenizerFactory"/>
     <filter class="solr.PhoneticFilterFactory" encoder="DoubleMetaphone" inject="true"/>
   </analyzer>
 </fieldType>

See Also:

PhoneticFilter

Field Summary

static String	ENCODER parameter name: either a short name or a full class name
static String	INJECT parameter name: true if encoded tokens should be added as synonyms
static String	MAX_CODE_LENGTH parameter name: restricts the length of the phonetic code

Fields inherited from class org.apache.lucene.analysis.util.AbstractAnalysisFactory

args, luceneMatchVersion

Constructor Summary

PhoneticFilterFactory()
Sole constructor.

Method Summary

PhoneticFilter	create(TokenStream input)
protected org.apache.commons.codec.Encoder	getEncoder() Must be thread-safe.
void	inform(ResourceLoader loader)

Methods inherited from class org.apache.lucene.analysis.util.TokenFilterFactory

availableTokenFilters, forName, lookupClass, reloadTokenFilters

Methods inherited from class org.apache.lucene.analysis.util.AbstractAnalysisFactory

assureMatchVersion, getArgs, getBoolean, getBoolean, getInt, getInt, getInt, getLines, getLuceneMatchVersion, getOriginalArgs, getPattern, getSnowballWordSet, getWordSet, init, setLuceneMatchVersion, splitFileNames

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

ENCODER

public static final String ENCODER

parameter name: either a short name or a full class name

See Also:

Constant Field Values

INJECT

public static final String INJECT

parameter name: true if encoded tokens should be added as synonyms

See Also:

Constant Field Values

MAX_CODE_LENGTH

public static final String MAX_CODE_LENGTH

parameter name: restricts the length of the phonetic code

See Also:

Constant Field Values

Constructor Detail

PhoneticFilterFactory

public PhoneticFilterFactory()

Sole constructor. See AbstractAnalysisFactory for initialization lifecycle.

Method Detail

inform

public void inform(ResourceLoader loader)
            throws IOException

Specified by:

inform in interface ResourceLoaderAware

Throws:

IOException

getEncoder

protected org.apache.commons.codec.Encoder getEncoder()

Must be thread-safe.

create

public PhoneticFilter create(TokenStream input)

Specified by:

create in class TokenFilterFactory