/*
* listPostings displays the first n postings for a term in a
* field in an index (specified by reader). Set n to MAX_VALUE
* to display all postings.
*/
static void listPostings(IndexReader reader, String termString, String field, Integer n)
throws IOException {
System.out.println("\nPostings: " + termString + " " + field);
/*
* Prepare to access the index.
*/
BytesRef termBytes = new BytesRef(termString);
Term term = new Term(field, termBytes);
Bits liveDocs = MultiFields.getLiveDocs(reader);
/*
* Lookup the collection term frequency (ctf).
*/
long df = reader.docFreq(term);
System.out.println("\tdf: " + df);
long ctf = reader.totalTermFreq(term);
System.out.println("\tctf: " + ctf);
if (df < 1) return;
/*
* Lookup the inverted list.
*/
DocsAndPositionsEnum postings =
MultiFields.getTermPositionsEnum(reader, liveDocs, field, termBytes);
/*
* Iterate through the first n postings.
*/
long count = 0;
while ((count < n) && (postings.nextDoc() != DocIdSetIterator.NO_MORE_DOCS)) {
System.out.println("\tdocid: " + postings.docID());
int tf = postings.freq();
System.out.println("\ttf: " + tf);
System.out.print("\tPositions: ");
for (int j = 0; j < tf; j++) {
int pos = postings.nextPosition();
System.out.print(pos + " ");
}
System.out.println("");
count++;
}
;
return;
}
public void testSetPosition() throws Exception {
Analyzer analyzer =
new Analyzer() {
@Override
public TokenStreamComponents createComponents(String fieldName) {
return new TokenStreamComponents(
new Tokenizer() {
// TODO: use CannedTokenStream
private final String[] TOKENS = {"1", "2", "3", "4", "5"};
private final int[] INCREMENTS = {1, 2, 1, 0, 1};
private int i = 0;
PositionIncrementAttribute posIncrAtt =
addAttribute(PositionIncrementAttribute.class);
CharTermAttribute termAtt = addAttribute(CharTermAttribute.class);
OffsetAttribute offsetAtt = addAttribute(OffsetAttribute.class);
@Override
public boolean incrementToken() {
if (i == TOKENS.length) return false;
clearAttributes();
termAtt.append(TOKENS[i]);
offsetAtt.setOffset(i, i);
posIncrAtt.setPositionIncrement(INCREMENTS[i]);
i++;
return true;
}
@Override
public void reset() throws IOException {
super.reset();
this.i = 0;
}
});
}
};
Directory store = newDirectory();
RandomIndexWriter writer = new RandomIndexWriter(random(), store, analyzer);
Document d = new Document();
d.add(newTextField("field", "bogus", Field.Store.YES));
writer.addDocument(d);
IndexReader reader = writer.getReader();
writer.close();
IndexSearcher searcher = newSearcher(reader);
PostingsEnum pos =
MultiFields.getTermPositionsEnum(searcher.getIndexReader(), "field", new BytesRef("1"));
pos.nextDoc();
// first token should be at position 0
assertEquals(0, pos.nextPosition());
pos = MultiFields.getTermPositionsEnum(searcher.getIndexReader(), "field", new BytesRef("2"));
pos.nextDoc();
// second token should be at position 2
assertEquals(2, pos.nextPosition());
PhraseQuery q;
ScoreDoc[] hits;
q = new PhraseQuery("field", "1", "2");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
// same as previous, using the builder with implicit positions
PhraseQuery.Builder builder = new PhraseQuery.Builder();
builder.add(new Term("field", "1"));
builder.add(new Term("field", "2"));
q = builder.build();
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
// same as previous, just specify positions explicitely.
builder = new PhraseQuery.Builder();
builder.add(new Term("field", "1"), 0);
builder.add(new Term("field", "2"), 1);
q = builder.build();
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
// specifying correct positions should find the phrase.
builder = new PhraseQuery.Builder();
builder.add(new Term("field", "1"), 0);
builder.add(new Term("field", "2"), 2);
q = builder.build();
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "2", "3");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "3", "4");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
// phrase query would find it when correct positions are specified.
builder = new PhraseQuery.Builder();
builder.add(new Term("field", "3"), 0);
builder.add(new Term("field", "4"), 0);
q = builder.build();
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
// phrase query should fail for non existing searched term
// even if there exist another searched terms in the same searched position.
builder = new PhraseQuery.Builder();
builder.add(new Term("field", "3"), 0);
builder.add(new Term("field", "9"), 0);
q = builder.build();
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
// multi-phrase query should succed for non existing searched term
// because there exist another searched terms in the same searched position.
MultiPhraseQuery mq = new MultiPhraseQuery();
mq.add(new Term[] {new Term("field", "3"), new Term("field", "9")}, 0);
hits = searcher.search(mq, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "2", "4");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "3", "5");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "4", "5");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(1, hits.length);
q = new PhraseQuery("field", "2", "5");
hits = searcher.search(q, 1000).scoreDocs;
assertEquals(0, hits.length);
reader.close();
store.close();
}