org.nuxeo.common.utils

Class FullTextUtils

java.lang.Object

org.nuxeo.common.utils.FullTextUtils

public class FullTextUtils
extends Object

Functions related to simple fulltext parsing. They don't try to be exhaustive but they work for simple cases.

Field Summary

Fields

Modifier and Type	Field and Description
static int	MIN_SIZE
static String	STOP_WORDS
static Set<String>	stopWords
static String	UNACCENTED
static Pattern	wordPattern

Method Summary

Modifier and Type	Method and Description
static Set<String>	parseFullText(String string, boolean removeDiacritics) Extracts the words from a string for simple fulltext indexing.
static String	parseWord(String string, boolean removeDiacritics) Parses a word and returns a simplified lowercase form.

Methods inherited from class java.lang.Object

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

Field Detail

wordPattern

public static final Pattern wordPattern

MIN_SIZE

public static final int MIN_SIZE

See Also:

Constant Field Values

STOP_WORDS

public static final String STOP_WORDS

See Also:

Constant Field Values

stopWords

public static final Set<String> stopWords

UNACCENTED

public static final String UNACCENTED

See Also:

Constant Field Values

Method Detail

parseFullText

public static Set<String> parseFullText(String string,
                                        boolean removeDiacritics)

Extracts the words from a string for simple fulltext indexing.

Initial order is kept, but duplicate words are removed.

It omits short or stop words, removes accents and does pseudo-stemming.

Parameters:

string - the string

removeDiacritics - if the diacritics must be removed

Returns:

an ordered set of resulting words

parseWord

public static String parseWord(String string,
                               boolean removeDiacritics)

Parses a word and returns a simplified lowercase form.

Parameters:

string - the word

removeDiacritics - if the diacritics must be removed

Returns:

the simplified word, or null if it was removed as a stop word or a short word