private TermDocs termDocs(int i) throws IOException {
TermDocs result = readerTermDocs[i];
if (result == null) result = readerTermDocs[i] = termDocs(readers[i]);
if (smi != null) {
assert (smi.ord == i);
assert (smi.termEnum.term().equals(term));
result.seek(smi.termEnum);
} else {
result.seek(term);
}
return result;
}
public void testSkipTo(int indexDivisor) throws IOException {
Directory dir = new RAMDirectory();
IndexWriter writer =
new IndexWriter(dir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
Term ta = new Term("content", "aaa");
for (int i = 0; i < 10; i++) addDoc(writer, "aaa aaa aaa aaa");
Term tb = new Term("content", "bbb");
for (int i = 0; i < 16; i++) addDoc(writer, "bbb bbb bbb bbb");
Term tc = new Term("content", "ccc");
for (int i = 0; i < 50; i++) addDoc(writer, "ccc ccc ccc ccc");
// assure that we deal with a single segment
writer.optimize();
writer.close();
IndexReader reader = IndexReader.open(dir);
reader.setTermInfosIndexDivisor(indexDivisor);
assertEquals(indexDivisor, reader.getTermInfosIndexDivisor());
TermDocs tdocs = reader.termDocs();
// without optimization (assumption skipInterval == 16)
// with next
tdocs.seek(ta);
assertTrue(tdocs.next());
assertEquals(0, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.next());
assertEquals(1, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.skipTo(0));
assertEquals(2, tdocs.doc());
assertTrue(tdocs.skipTo(4));
assertEquals(4, tdocs.doc());
assertTrue(tdocs.skipTo(9));
assertEquals(9, tdocs.doc());
assertFalse(tdocs.skipTo(10));
// without next
tdocs.seek(ta);
assertTrue(tdocs.skipTo(0));
assertEquals(0, tdocs.doc());
assertTrue(tdocs.skipTo(4));
assertEquals(4, tdocs.doc());
assertTrue(tdocs.skipTo(9));
assertEquals(9, tdocs.doc());
assertFalse(tdocs.skipTo(10));
// exactly skipInterval documents and therefore with optimization
// with next
tdocs.seek(tb);
assertTrue(tdocs.next());
assertEquals(10, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.next());
assertEquals(11, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.skipTo(5));
assertEquals(12, tdocs.doc());
assertTrue(tdocs.skipTo(15));
assertEquals(15, tdocs.doc());
assertTrue(tdocs.skipTo(24));
assertEquals(24, tdocs.doc());
assertTrue(tdocs.skipTo(25));
assertEquals(25, tdocs.doc());
assertFalse(tdocs.skipTo(26));
// without next
tdocs.seek(tb);
assertTrue(tdocs.skipTo(5));
assertEquals(10, tdocs.doc());
assertTrue(tdocs.skipTo(15));
assertEquals(15, tdocs.doc());
assertTrue(tdocs.skipTo(24));
assertEquals(24, tdocs.doc());
assertTrue(tdocs.skipTo(25));
assertEquals(25, tdocs.doc());
assertFalse(tdocs.skipTo(26));
// much more than skipInterval documents and therefore with optimization
// with next
tdocs.seek(tc);
assertTrue(tdocs.next());
assertEquals(26, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.next());
assertEquals(27, tdocs.doc());
assertEquals(4, tdocs.freq());
assertTrue(tdocs.skipTo(5));
assertEquals(28, tdocs.doc());
assertTrue(tdocs.skipTo(40));
assertEquals(40, tdocs.doc());
assertTrue(tdocs.skipTo(57));
assertEquals(57, tdocs.doc());
assertTrue(tdocs.skipTo(74));
assertEquals(74, tdocs.doc());
assertTrue(tdocs.skipTo(75));
assertEquals(75, tdocs.doc());
assertFalse(tdocs.skipTo(76));
// without next
tdocs.seek(tc);
assertTrue(tdocs.skipTo(5));
assertEquals(26, tdocs.doc());
assertTrue(tdocs.skipTo(40));
assertEquals(40, tdocs.doc());
assertTrue(tdocs.skipTo(57));
assertEquals(57, tdocs.doc());
assertTrue(tdocs.skipTo(74));
assertEquals(74, tdocs.doc());
assertTrue(tdocs.skipTo(75));
assertEquals(75, tdocs.doc());
assertFalse(tdocs.skipTo(76));
tdocs.close();
reader.close();
dir.close();
}
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();
}
}
}