internal CharArraySetIterator(CharArraySet enclosingInstance) { InitBlock(enclosingInstance); GoNext(); }
/// <summary> Returns an unmodifiable {@link CharArraySet}. This allows to provide /// unmodifiable views of internal sets for "read-only" use. /// /// </summary> /// <param name="set">a set for which the unmodifiable set is returned. /// </param> /// <returns> an new unmodifiable {@link CharArraySet}. /// </returns> /// <throws> NullPointerException </throws> /// <summary> if the given set is <code>null</code>. /// </summary> public static CharArraySet UnmodifiableSet(CharArraySet set_Renamed) { if (set_Renamed == null) throw new System.NullReferenceException("Given set is null"); /* * Instead of delegating calls to the given set copy the low-level values to * the unmodifiable Subclass */ return new UnmodifiableCharArraySet(set_Renamed.entries, set_Renamed.ignoreCase, set_Renamed.count); }
private void InitBlock(CharArraySet enclosingInstance) { this.enclosingInstance = enclosingInstance; }
public StopFilter(bool enablePositionIncrements, TokenStream in_Renamed, System.String[] stopWords, bool ignoreCase):base(in_Renamed) { this.stopWords = (CharArraySet) MakeStopSet(stopWords, ignoreCase); this.enablePositionIncrements = enablePositionIncrements; Init(); }
/// <summary> </summary> /// <param name="stopWords">A List of Strings representing the stopwords /// </param> /// <param name="ignoreCase">if true, all words are lower cased first /// </param> /// <returns> A Set containing the words /// </returns> public static System.Collections.Hashtable MakeStopSet(System.Collections.IList stopWords, bool ignoreCase) { CharArraySet stopSet = new CharArraySet(stopWords.Count, ignoreCase); stopSet.AddAll(stopWords); return stopSet; }
/// <summary> </summary> /// <param name="stopWords">An array of stopwords /// </param> /// <param name="ignoreCase">If true, all words are lower cased first. /// </param> /// <returns> a Set containing the words /// </returns> public static System.Collections.Hashtable MakeStopSet(System.String[] stopWords, bool ignoreCase) { CharArraySet stopSet = new CharArraySet(stopWords.Length, ignoreCase); stopSet.AddAll(new System.Collections.ArrayList(stopWords)); return stopSet; }
/// <summary> Construct a token stream filtering the given input. /// If <code>stopWords</code> is an instance of {@link CharArraySet} (true if /// <code>makeStopSet()</code> was used to construct the set) it will be directly used /// and <code>ignoreCase</code> will be ignored since <code>CharArraySet</code> /// directly controls case sensitivity. /// <p/> /// If <code>stopWords</code> is not an instance of {@link CharArraySet}, /// a new CharArraySet will be constructed and <code>ignoreCase</code> will be /// used to specify the case sensitivity of that set. /// /// </summary> /// <param name="enablePositionIncrements">true if token positions should record the removed stop words /// </param> /// <param name="input">Input TokenStream /// </param> /// <param name="stopWords">The set of Stop Words. /// </param> /// <param name="ignoreCase">-Ignore case when stopping. /// </param> public StopFilter(bool enablePositionIncrements, TokenStream input, System.Collections.Hashtable stopWords, bool ignoreCase):base(input) { if (stopWords is CharArraySet) { this.stopWords = (CharArraySet) stopWords; } else { this.stopWords = new CharArraySet(stopWords.Count, ignoreCase); this.stopWords.Add(stopWords); } this.enablePositionIncrements = enablePositionIncrements; Init(); }
static StopAnalyzer() { { System.String[] stopWords = new System.String[]{"a", "an", "and", "are", "as", "at", "be", "but", "by", "for", "if", "in", "into", "is", "it", "no", "not", "of", "on", "or", "such", "that", "the", "their", "then", "there", "these", "they", "this", "to", "was", "will", "with"}; CharArraySet stopSet = new CharArraySet(stopWords.Length, false); stopSet.AddAll(new System.Collections.ArrayList(stopWords)); ENGLISH_STOP_WORDS_SET = CharArraySet.UnmodifiableSet(stopSet); } }