private static Map<String, List<String>> generate_result(Directory directory) {
Map<String, List<String>> result_map = new HashMap<String, List<String>>();
try {
IndexReader reader = IndexReader.open(directory);
TermEnum termEnum = reader.terms();
while (termEnum.next()) {
String termEnumString = termEnum.term().toString();
if (termEnumString.startsWith("content:")) {
String term = termEnumString.substring(termEnumString.lastIndexOf(":") + 1);
TermDocs termDocs = reader.termDocs(termEnum.term());
while (termDocs.next()) {
Document doc = reader.document(termDocs.doc());
String relative_path = doc.get("relative_path");
if (result_map.containsKey(relative_path)) {
result_map.get(relative_path).add(term + termDocs.freq());
} else {
result_map.put(relative_path, new ArrayList<String>());
}
}
}
}
} catch (IOException e) {
e.printStackTrace();
} finally {
}
return result_map;
}
public void getIndexInfo(String indexdir, int freqThreshold) {
IndexReader reader = null;
try {
Directory dir = FSDirectory.open(new File(indexdir));
System.out.println(dir);
reader = IndexReader.open(dir);
System.out.println("document num:" + reader.numDocs());
System.out.println("======================");
TermEnum terms = reader.terms();
sortedTermQueue.clear();
maxDocNum = reader.maxDoc();
linkMap.clear();
termList.clear();
while (terms.next()) {
// System.out.print(terms.term() + "\tDocFreq:" +
TermDocs termDocs = reader.termDocs(terms.term());
MyTerm temp = new MyTerm(terms.term(), termDocs, maxDocNum);
if (temp.totalFreq < freqThreshold) {
continue;
} /*
* if(temp.originTrem.text().length()==1){ continue; }
*/
linkMap.put(temp.originTrem.text(), temp);
sortedTermQueue.add(temp);
termList.add(temp);
}
System.out.println("total Size:" + sortedTermQueue.size());
System.out.println("mapsize:" + linkMap.keySet().size());
// System.exit(0);
int num = 0;
this.maxFreq = sortedTermQueue.peek().totalFreq;
while (!sortedTermQueue.isEmpty()) {
num++;
System.out.println(num + ":" + sortedTermQueue.poll());
}
System.out.println("read index info done");
} catch (IOException e) {
e.printStackTrace();
} finally {
try {
reader.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
public void seek(TermEnum termEnum) throws IOException {
seek(termEnum.term());
if (termEnum instanceof MultiTermEnum) {
tenum = (MultiTermEnum) termEnum;
if (topReader != tenum.topReader) tenum = null;
}
}
/**
* Tests the IndexReader.getFieldNames implementation
*
* @throws Exception on error
*/
public void testFilterIndexReader() throws Exception {
Directory directory = newDirectory();
IndexWriter writer =
new IndexWriter(
directory, newIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(random)));
Document d1 = new Document();
d1.add(newField("default", "one two", Field.Store.YES, Field.Index.ANALYZED));
writer.addDocument(d1);
Document d2 = new Document();
d2.add(newField("default", "one three", Field.Store.YES, Field.Index.ANALYZED));
writer.addDocument(d2);
Document d3 = new Document();
d3.add(newField("default", "two four", Field.Store.YES, Field.Index.ANALYZED));
writer.addDocument(d3);
writer.close();
IndexReader reader = new TestReader(IndexReader.open(directory, true));
TermEnum terms = reader.terms();
while (terms.next()) {
assertTrue(terms.term().text().indexOf('e') != -1);
}
terms.close();
TermPositions positions = reader.termPositions(new Term("default", "one"));
while (positions.next()) {
assertTrue((positions.doc() % 2) == 1);
}
int NUM_DOCS = 3;
TermDocs td = reader.termDocs(null);
for (int i = 0; i < NUM_DOCS; i++) {
assertTrue(td.next());
assertEquals(i, td.doc());
assertEquals(1, td.freq());
}
td.close();
reader.close();
directory.close();
}
public MultiTermEnum(IndexReader topReader, IndexReader[] readers, int[] starts, Term t)
throws IOException {
this.topReader = topReader;
queue = new SegmentMergeQueue(readers.length);
matchingSegments = new SegmentMergeInfo[readers.length + 1];
for (int i = 0; i < readers.length; i++) {
IndexReader reader = readers[i];
TermEnum termEnum;
if (t != null) {
termEnum = reader.terms(t);
} else termEnum = reader.terms();
SegmentMergeInfo smi = new SegmentMergeInfo(starts[i], termEnum, reader);
smi.ord = i;
if (t == null ? smi.next() : termEnum.term() != null) queue.add(smi); // initialize queue
else smi.close();
}
if (t != null && queue.size() > 0) {
next();
}
}
public void seek(TermEnum terms) throws IOException {
original.seek(terms);
docFreq = terms.docFreq();
pointer = -1;
if (docFreq > postingMaps.length) { // grow postingsMap
PostingMap[] newMap = new PostingMap[docFreq];
System.arraycopy(postingMaps, 0, newMap, 0, postingMaps.length);
for (int i = postingMaps.length; i < docFreq; i++) {
newMap[i] = new PostingMap();
}
postingMaps = newMap;
}
out.reset();
int i = 0;
while (original.next()) {
PostingMap map = postingMaps[i++];
map.newDoc = oldToNew[original.doc()]; // remap the newDoc id
map.offset = out.getFilePointer(); // save pointer to buffer
final int tf = original.freq(); // buffer tf & positions
out.writeVInt(tf);
int prevPosition = 0;
for (int j = tf; j > 0; j--) { // delta encode positions
int p = original.nextPosition();
out.writeVInt(p - prevPosition);
prevPosition = p;
}
}
out.flush();
docFreq = i; // allow for deletions
Arrays.sort(postingMaps, 0, docFreq); // resort by mapped doc ids
// HeapSorter.sort(postingMaps,docFreq); // TODO MC - due to the lack of space
// NOTE: this might be substantially faster if RAMInputStream were public
// and supported a reset() operation.
in = tempDir.openInput(TEMP_FILE);
}
public static void main(String[] args) throws Exception {
// the IndexReader object is the main handle that will give you
// all the documents, terms and inverted index
IndexReader r = IndexReader.open(FSDirectory.open(new File("index")));
// You can figure out the number of documents using the maxDoc() function
System.out.println("The number of documents in this index is: " + r.maxDoc());
int i = 0;
// You can find out all the terms that have been indexed using the terms() function
TermEnum t = r.terms();
while (t.next()) {
// Since there are so many terms, let us try printing only term #100000-#100010
if (i > 100000) System.out.println("[" + i + "] " + t.term().text());
if (++i > 100010) break;
}
// You can create your own query terms by calling the Term constructor, with the field
// 'contents'
// In the following example, the query term is 'brute'
Term te = new Term("contents", "brute");
// You can also quickly find out the number of documents that have term t
System.out.println("Number of documents with the word 'brute' is: " + r.docFreq(te));
// You can use the inverted index to find out all the documents that contain the term 'brute'
// by using the termDocs function
TermDocs td = r.termDocs(te);
while (td.next()) {
System.out.println(
"Document number ["
+ td.doc()
+ "] contains the term 'brute' "
+ td.freq()
+ " time(s).");
}
// You can find the URL of the a specific document number using the document() function
// For example, the URL for document number 14191 is:
Document d = r.document(14191);
String url =
d.getFieldable("path")
.stringValue(); // the 'path' field of the Document object holds the URL
System.out.println(url.replace("%%", "/"));
// -------- Now let us use all of the functions above to make something useful --------
// The following bit of code is a worked out example of how to get a bunch of documents
// in response to a query and show them (without ranking them according to TF/IDF)
Scanner sc = new Scanner(System.in);
String str = "";
System.out.print("query> ");
while (!(str = sc.nextLine()).equals("quit")) {
String[] terms = str.split("\\s+");
for (String word : terms) {
Term term = new Term("contents", word);
TermDocs tdocs = r.termDocs(term);
while (tdocs.next()) {
String d_url =
r.document(tdocs.doc()).getFieldable("path").stringValue().replace("%%", "/");
System.out.println("[" + tdocs.doc() + "] " + d_url);
}
}
System.out.print("query> ");
}
}
public static void main(String[] args) throws Exception {
if (args.length < 2) {
System.err.println("TermDumper [-c|-v value] field <index...>");
System.exit(1);
}
boolean count = false;
String value = null;
boolean all = false;
int i = 0;
for (; i < args.length; i++) {
String arg = args[i];
if ("-h".equals(arg) || "--help".equals(arg)) {
System.err.println("TermDumper [-c|-v value] field <index...>");
System.exit(1);
} else if ("-c".equals(arg) || "--count".equals(arg)) {
count = true;
} else if ("-v".equals(arg) || "--vaue".equals(arg)) {
value = args[++i];
} else if ("-a".equals(arg) || "--all".equals(arg)) {
all = true;
} else {
break;
}
}
String field = args[i++];
java.util.ArrayList<IndexReader> readers =
new java.util.ArrayList<IndexReader>(args.length - 1);
for (; i < args.length; i++) {
String arg = args[i];
try {
IndexReader reader = IndexReader.open(new MMapDirectory(new File(arg)), true);
readers.add(reader);
} catch (IOException ioe) {
System.err.println("Error reading: " + arg);
}
}
for (IndexReader reader : readers) {
TermDocs termDocs = reader.termDocs();
TermEnum termEnum = reader.terms(new Term(field));
try {
do {
Term term = termEnum.term();
if (term == null || !field.equals(term.field())) break;
if (value == null) {
if (count) {
termDocs.seek(termEnum);
int c = 0;
for (; termDocs.next(); c++) ;
System.out.print(c + " ");
}
System.out.println(term.text());
} else if (value.equals(term.text())) {
termDocs.seek(termEnum);
while (termDocs.next()) {
if (all) {
Document d = reader.document(termDocs.doc());
System.out.println(termDocs.doc());
for (Object o : d.getFields()) {
Field f = (Field) o;
System.out.println(f.name() + " " + d.get(f.name()));
}
} else {
System.out.println(
termDocs.doc() + " " + reader.document(termDocs.doc()).get("url"));
}
}
}
} while (termEnum.next());
} finally {
termDocs.close();
termEnum.close();
}
}
}
