File	Doc	Category	Size	Date	Package
MemoryIndex.java	API Doc	Apache Lucene 2.1.0	44006	Wed Feb 14 10:46:24 GMT 2007	org.apache.lucene.index.memory

MemoryIndex

• java.lang.Object

float score = search(String text, Query query)

Analyzer analyzer = PatternAnalyzer.DEFAULT_ANALYZER;
//Analyzer analyzer = new SimpleAnalyzer();
MemoryIndex index = new MemoryIndex();
index.addField("content", "Readings about Salmons and other select Alaska fishing Manuals", analyzer);
index.addField("author", "Tales of James", analyzer);
QueryParser parser = new QueryParser("content", analyzer);
float score = index.search(parser.parse("+author:james +salmon~ +fish* manual~"));
if (score > 0.0f) {
System.out.println("it's a match");
} else {
System.out.println("no match found");
}
System.out.println("indexData=" + index.toString());

(: An XQuery that finds all books authored by James that have something to do with "salmon fishing manuals", sorted by relevance :)
declare namespace lucene = "java:nux.xom.pool.FullTextUtil";
declare variable $query := "+salmon~ +fish* manual~"; (: any arbitrary Lucene query can go here :)

for $book in /books/book[author="James" and lucene:match(abstract, $query) > 0.0]
let $score := lucene:match($book/abstract, $query)
order by $score descending
return $book

MemoryIndex index = ...
synchronized (index) {
// read and/or write index (i.e. add fields and/or query)
}

Fields Summary
private final HashMap	fields info for each field: Map
private transient Map$Entry[]	sortedFields fields sorted ascending by fieldName; lazily computed on demand
private final int	stride pos: positions[3*i], startOffset: positions[3*i +1], endOffset: positions[3*i +2]
private static final float	docBoost Could be made configurable; See {@link Document#setBoost(float)}
private static final long	serialVersionUID
private static final boolean	DEBUG
private static final Comparator	termComparator Sorts term entries into ascending order; also works for Arrays.binarySearch() and Arrays.sort()
private static final Term	MATCH_ALL_TERM

Constructors Summary
public MemoryIndex() Constructs an empty instance. this(false);
private MemoryIndex(boolean storeOffsets) Constructs an empty instance that can optionally store the start and end character offset of each token term in the text. This can be useful for highlighting of hit locations with the Lucene highlighter package. Private until the highlighter package matures, so that this can actually be meaningfully integrated. param storeOffsets whether or not to store the start and end character offset of each token term in the text this.stride = storeOffsets ? 3 : 1;

Methods Summary
public void	addField(java.lang.String fieldName, java.lang.String text, org.apache.lucene.analysis.Analyzer analyzer) Convenience method; Tokenizes the given field text and adds the resulting terms to the index; Equivalent to adding an indexed non-keyword Lucene {@link org.apache.lucene.document.Field} that is {@link org.apache.lucene.document.Field.Index#TOKENIZED tokenized}, {@link org.apache.lucene.document.Field.Store#NO not stored}, {@link org.apache.lucene.document.Field.TermVector#WITH_POSITIONS termVectorStored with positions} (or {@link org.apache.lucene.document.Field.TermVector#WITH_POSITIONS termVectorStored with positions and offsets}), param fieldName a name to be associated with the text param text the text to tokenize and index. param analyzer the analyzer to use for tokenization if (fieldName == null) throw new IllegalArgumentException("fieldName must not be null"); if (text == null) throw new IllegalArgumentException("text must not be null"); if (analyzer == null) throw new IllegalArgumentException("analyzer must not be null"); TokenStream stream; if (analyzer instanceof PatternAnalyzer) { stream = ((PatternAnalyzer) analyzer).tokenStream(fieldName, text); } else { stream = analyzer.tokenStream(fieldName, new PatternAnalyzer.FastStringReader(text)); } addField(fieldName, stream);
public void	addField(java.lang.String fieldName, org.apache.lucene.analysis.TokenStream stream) Equivalent to addField(fieldName, stream, 1.0f). param fieldName a name to be associated with the text param stream the token stream to retrieve tokens from addField(fieldName, stream, 1.0f);
public void	addField(java.lang.String fieldName, org.apache.lucene.analysis.TokenStream stream, float boost) Iterates over the given token stream and adds the resulting terms to the index; Equivalent to adding a tokenized, indexed, termVectorStored, unstored, Lucene {@link org.apache.lucene.document.Field}. Finally closes the token stream. Note that untokenized keywords can be added with this method via {@link #keywordTokenStream(Collection)}, the Lucene contrib KeywordTokenizer or similar utilities. param fieldName a name to be associated with the text param stream the token stream to retrieve tokens from. param boost the boost factor for hits for this field see Field#setBoost(float) /* * Note that this method signature avoids having a user call new * o.a.l.d.Field(...) which would be much too expensive due to the * String.intern() usage of that class. * * More often than not, String.intern() leads to serious performance * degradations rather than improvements! If you're curious why, check * out the JDK's native code, see how it oscillates multiple times back * and forth between Java code and native code on each intern() call, * only to end up using a plain vanilla java.util.HashMap on the Java * heap for it's interned strings! String.equals() has a small cost * compared to String.intern(), trust me. Application level interning * (e.g. a HashMap per Directory/Index) typically leads to better * solutions than frequent hidden low-level calls to String.intern(). * * Perhaps with some luck, Lucene's Field.java (and Term.java) and * cousins could be fixed to not use String.intern(). Sigh :-( */ try { if (fieldName == null) throw new IllegalArgumentException("fieldName must not be null"); if (stream == null) throw new IllegalArgumentException("token stream must not be null"); if (boost <= 0.0f) throw new IllegalArgumentException("boost factor must be greater than 0.0"); if (fields.get(fieldName) != null) throw new IllegalArgumentException("field must not be added more than once"); HashMap terms = new HashMap(); int numTokens = 0; int pos = -1; Token token; while ((token = stream.next()) != null) { String term = token.termText(); if (term.length() == 0) continue; // nothing to do // if (DEBUG) System.err.println("token='" + term + "'"); numTokens++; pos += token.getPositionIncrement(); ArrayIntList positions = (ArrayIntList) terms.get(term); if (positions == null) { // term not seen before positions = new ArrayIntList(stride); terms.put(term, positions); } if (stride == 1) { positions.add(pos); } else { positions.add(pos, token.startOffset(), token.endOffset()); } } // ensure infos.numTokens > 0 invariant; needed for correct operation of terms() if (numTokens > 0) { boost = boost * docBoost; // see DocumentWriter.addDocument(...) fields.put(fieldName, new Info(terms, numTokens, boost)); sortedFields = null; // invalidate sorted view, if any } } catch (IOException e) { // can never happen throw new RuntimeException(e); } finally { try { if (stream != null) stream.close(); } catch (IOException e2) { throw new RuntimeException(e2); } }
public org.apache.lucene.search.IndexSearcher	createSearcher() Creates and returns a searcher that can be used to execute arbitrary Lucene queries and to collect the resulting query results as hits. return a searcher MemoryIndexReader reader = new MemoryIndexReader(); IndexSearcher searcher = new IndexSearcher(reader); // ensures no auto-close !! reader.setSearcher(searcher); // to later get hold of searcher.getSimilarity() return searcher;
public int	getMemorySize() Returns a reasonable approximation of the main memory [bytes] consumed by this instance. Useful for smart memory sensititive caches/pools. Assumes fieldNames are interned, whereas tokenized terms are memory-overlaid. return the main memory consumption // for example usage in a smart cache see nux.xom.pool.Pool int PTR = VM.PTR; int INT = VM.INT; int size = 0; size += VM.sizeOfObject(2*PTR + INT); // memory index if (sortedFields != null) size += VM.sizeOfObjectArray(sortedFields.length); size += VM.sizeOfHashMap(fields.size()); Iterator iter = fields.entrySet().iterator(); while (iter.hasNext()) { // for each Field Info Map.Entry entry = (Map.Entry) iter.next(); Info info = (Info) entry.getValue(); size += VM.sizeOfObject(2*INT + 3*PTR); // Info instance vars if (info.sortedTerms != null) size += VM.sizeOfObjectArray(info.sortedTerms.length); int len = info.terms.size(); size += VM.sizeOfHashMap(len); Iterator iter2 = info.terms.entrySet().iterator(); while (--len >= 0) { // for each term Map.Entry e = (Map.Entry) iter2.next(); size += VM.sizeOfObject(PTR + 3*INT); // assumes substring() memory overlay // size += STR + 2 * ((String) e.getKey()).length(); ArrayIntList positions = (ArrayIntList) e.getValue(); size += VM.sizeOfArrayIntList(positions.size()); } } return size;
public org.apache.lucene.analysis.TokenStream	keywordTokenStream(java.util.Collection keywords) Convenience method; Creates and returns a token stream that generates a token for each keyword in the given collection, "as is", without any transforming text analysis. The resulting token stream can be fed into {@link #addField(String, TokenStream)}, perhaps wrapped into another {@link org.apache.lucene.analysis.TokenFilter}, as desired. param keywords the keywords to generate tokens for return the corresponding token stream // TODO: deprecate & move this method into AnalyzerUtil? if (keywords == null) throw new IllegalArgumentException("keywords must not be null"); return new TokenStream() { private Iterator iter = keywords.iterator(); private int start = 0; public Token next() { if (!iter.hasNext()) return null; Object obj = iter.next(); if (obj == null) throw new IllegalArgumentException("keyword must not be null"); String term = obj.toString(); Token token = new Token(term, start, start + term.length()); start += term.length() + 1; // separate words by 1 (blank) character return token; } };
private int	numPositions(org.apache.lucene.index.memory.MemoryIndex$ArrayIntList positions) return positions.size() / stride;
public float	search(org.apache.lucene.search.Query query) Convenience method that efficiently returns the relevance score by matching this index against the given Lucene query expression. param query an arbitrary Lucene query to run against this index return the relevance score of the matchmaking; A number in the range [0.0 .. 1.0], with 0.0 indicating no match. The higher the number the better the match. see org.apache.lucene.queryParser.QueryParser#parse(String) if (query == null) throw new IllegalArgumentException("query must not be null"); Searcher searcher = createSearcher(); try { final float[] scores = new float[1]; // inits to 0.0f (no match) searcher.search(query, new HitCollector() { public void collect(int doc, float score) { scores[0] = score; } }); float score = scores[0]; return score; } catch (IOException e) { // can never happen (RAMDirectory) throw new RuntimeException(e); } finally { // searcher.close(); /* * Note that it is harmless and important for good performance to * NOT close the index reader!!! This avoids all sorts of * unnecessary baggage and locking in the Lucene IndexReader * superclass, all of which is completely unnecessary for this main * memory index data structure without thread-safety claims. * * Wishing IndexReader would be an interface... * * Actually with the new tight createSearcher() API auto-closing is now * made impossible, hence searcher.close() would be harmless and also * would not degrade performance... */ }
private static java.util.Map$Entry[]	sort(java.util.HashMap map) returns a view of the given map's entries, sorted ascending by key int size = map.size(); Map.Entry[] entries = new Map.Entry[size]; Iterator iter = map.entrySet().iterator(); for (int i=0; i < size; i++) { entries[i] = (Map.Entry) iter.next(); } if (size > 1) Arrays.sort(entries, termComparator); return entries;
private void	sortFields() sorts into ascending order (on demand), reusing memory along the way if (sortedFields == null) sortedFields = sort(fields);
public java.lang.String	toString() Returns a String representation of the index data for debugging purposes. return the string representation StringBuffer result = new StringBuffer(256); sortFields(); int sumChars = 0; int sumPositions = 0; int sumTerms = 0; for (int i=0; i < sortedFields.length; i++) { Map.Entry entry = sortedFields[i]; String fieldName = (String) entry.getKey(); Info info = (Info) entry.getValue(); info.sortTerms(); result.append(fieldName + ":\n"); int numChars = 0; int numPositions = 0; for (int j=0; j < info.sortedTerms.length; j++) { Map.Entry e = info.sortedTerms[j]; String term = (String) e.getKey(); ArrayIntList positions = (ArrayIntList) e.getValue(); result.append("\t'" + term + "':" + numPositions(positions) + ":"); result.append(positions.toString(stride)); // ignore offsets result.append("\n"); numPositions += numPositions(positions); numChars += term.length(); } result.append("\tterms=" + info.sortedTerms.length); result.append(", positions=" + numPositions); result.append(", Kchars=" + (numChars/1000.0f)); result.append("\n"); sumPositions += numPositions; sumChars += numChars; sumTerms += info.sortedTerms.length; } result.append("\nfields=" + sortedFields.length); result.append(", terms=" + sumTerms); result.append(", positions=" + sumPositions); result.append(", Kchars=" + (sumChars/1000.0f)); return result.toString();