/* * Utility function to display a term vector. */ static void termVectorDisplay(Terms terms) throws IOException { if ((terms == null) || (terms.size() == -1)) System.out.println(" The field is not stored."); else { /* * The terms for this field are stored. */ System.out.println(" Vocabulary size: " + terms.size() + " terms"); TermsEnum ithTerm = terms.iterator(null); /* * Iterate over the terms in this document. * Information about a term's occurrences (tf and * positions) is accessed via the indexing API, which * returns inverted lists that describe (only) the * current document. */ while (ithTerm.next() != null) { System.out.format( " %10d %-20s %d ", ithTerm.ord(), ithTerm.term().utf8ToString(), ithTerm.totalTermFreq()); DocsAndPositionsEnum currDoc = ithTerm.docsAndPositions(null, null); currDoc.nextDoc(); for (int jthPosition = 0; jthPosition < ithTerm.totalTermFreq(); jthPosition++) System.out.print(currDoc.nextPosition() + " "); System.out.println(); } ; } ; }
public void testSortedTermsEnum() throws IOException { Directory directory = newDirectory(); Analyzer analyzer = new MockAnalyzer(random()); IndexWriterConfig iwconfig = newIndexWriterConfig(TEST_VERSION_CURRENT, analyzer); iwconfig.setMergePolicy(newLogMergePolicy()); RandomIndexWriter iwriter = new RandomIndexWriter(random(), directory, iwconfig); Document doc = new Document(); doc.add(new StringField("field", "hello", Field.Store.NO)); iwriter.addDocument(doc); doc = new Document(); doc.add(new StringField("field", "world", Field.Store.NO)); iwriter.addDocument(doc); doc = new Document(); doc.add(new StringField("field", "beer", Field.Store.NO)); iwriter.addDocument(doc); iwriter.forceMerge(1); DirectoryReader ireader = iwriter.getReader(); iwriter.close(); AtomicReader ar = getOnlySegmentReader(ireader); SortedSetDocValues dv = FieldCache.DEFAULT.getDocTermOrds(ar, "field"); assertEquals(3, dv.getValueCount()); TermsEnum termsEnum = dv.termsEnum(); // next() assertEquals("beer", termsEnum.next().utf8ToString()); assertEquals(0, termsEnum.ord()); assertEquals("hello", termsEnum.next().utf8ToString()); assertEquals(1, termsEnum.ord()); assertEquals("world", termsEnum.next().utf8ToString()); assertEquals(2, termsEnum.ord()); // seekCeil() assertEquals(SeekStatus.NOT_FOUND, termsEnum.seekCeil(new BytesRef("ha!"))); assertEquals("hello", termsEnum.term().utf8ToString()); assertEquals(1, termsEnum.ord()); assertEquals(SeekStatus.FOUND, termsEnum.seekCeil(new BytesRef("beer"))); assertEquals("beer", termsEnum.term().utf8ToString()); assertEquals(0, termsEnum.ord()); assertEquals(SeekStatus.END, termsEnum.seekCeil(new BytesRef("zzz"))); // seekExact() assertTrue(termsEnum.seekExact(new BytesRef("beer"), true)); assertEquals("beer", termsEnum.term().utf8ToString()); assertEquals(0, termsEnum.ord()); assertTrue(termsEnum.seekExact(new BytesRef("hello"), true)); assertEquals("hello", termsEnum.term().utf8ToString()); assertEquals(1, termsEnum.ord()); assertTrue(termsEnum.seekExact(new BytesRef("world"), true)); assertEquals("world", termsEnum.term().utf8ToString()); assertEquals(2, termsEnum.ord()); assertFalse(termsEnum.seekExact(new BytesRef("bogus"), true)); // seek(ord) termsEnum.seekExact(0); assertEquals("beer", termsEnum.term().utf8ToString()); assertEquals(0, termsEnum.ord()); termsEnum.seekExact(1); assertEquals("hello", termsEnum.term().utf8ToString()); assertEquals(1, termsEnum.ord()); termsEnum.seekExact(2); assertEquals("world", termsEnum.term().utf8ToString()); assertEquals(2, termsEnum.ord()); ireader.close(); directory.close(); }
private void verify(AtomicReader r, int[][] idToOrds, BytesRef[] termsArray, BytesRef prefixRef) throws Exception { final DocTermOrds dto = new DocTermOrds( r, r.getLiveDocs(), "field", prefixRef, Integer.MAX_VALUE, _TestUtil.nextInt(random(), 2, 10)); final FieldCache.Ints docIDToID = FieldCache.DEFAULT.getInts(r, "id", false); /* for(int docID=0;docID<subR.maxDoc();docID++) { System.out.println(" docID=" + docID + " id=" + docIDToID[docID]); } */ if (VERBOSE) { System.out.println( "TEST: verify prefix=" + (prefixRef == null ? "null" : prefixRef.utf8ToString())); System.out.println("TEST: all TERMS:"); TermsEnum allTE = MultiFields.getTerms(r, "field").iterator(null); int ord = 0; while (allTE.next() != null) { System.out.println(" ord=" + (ord++) + " term=" + allTE.term().utf8ToString()); } } // final TermsEnum te = subR.fields().terms("field").iterator(); final TermsEnum te = dto.getOrdTermsEnum(r); if (dto.numTerms() == 0) { if (prefixRef == null) { assertNull(MultiFields.getTerms(r, "field")); } else { Terms terms = MultiFields.getTerms(r, "field"); if (terms != null) { TermsEnum termsEnum = terms.iterator(null); TermsEnum.SeekStatus result = termsEnum.seekCeil(prefixRef, false); if (result != TermsEnum.SeekStatus.END) { assertFalse( "term=" + termsEnum.term().utf8ToString() + " matches prefix=" + prefixRef.utf8ToString(), StringHelper.startsWith(termsEnum.term(), prefixRef)); } else { // ok } } else { // ok } } return; } if (VERBOSE) { System.out.println("TEST: TERMS:"); te.seekExact(0); while (true) { System.out.println(" ord=" + te.ord() + " term=" + te.term().utf8ToString()); if (te.next() == null) { break; } } } SortedSetDocValues iter = dto.iterator(r); for (int docID = 0; docID < r.maxDoc(); docID++) { if (VERBOSE) { System.out.println( "TEST: docID=" + docID + " of " + r.maxDoc() + " (id=" + docIDToID.get(docID) + ")"); } iter.setDocument(docID); final int[] answers = idToOrds[docIDToID.get(docID)]; int upto = 0; long ord; while ((ord = iter.nextOrd()) != SortedSetDocValues.NO_MORE_ORDS) { te.seekExact(ord); final BytesRef expected = termsArray[answers[upto++]]; if (VERBOSE) { System.out.println( " exp=" + expected.utf8ToString() + " actual=" + te.term().utf8ToString()); } assertEquals( "expected=" + expected.utf8ToString() + " actual=" + te.term().utf8ToString() + " ord=" + ord, expected, te.term()); } assertEquals(answers.length, upto); } }