Java Code Examples for org.apache.lucene.document.StoredField#TYPE

public void testIllegalVectorsWithoutIndexed() throws Exception {
  Directory dir = newDirectory();
  MockAnalyzer a = new MockAnalyzer(random());
  a.setEnableChecks(false);
  RandomIndexWriter w = new RandomIndexWriter(random(), dir, a);
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectors(true);
  Document doc = new Document();
  doc.add(new Field("field", "value", ft));
  
  IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
    w.addDocument(doc);
  });
  assertEquals("cannot store term vectors for a field that is not indexed (field=\"field\")", expected.getMessage());
  
  w.close();
  dir.close();
}
 

public void testIllegalVectorPositionsWithoutIndexed() throws Exception {
  Directory dir = newDirectory();
  MockAnalyzer a = new MockAnalyzer(random());
  a.setEnableChecks(false);
  RandomIndexWriter w = new RandomIndexWriter(random(), dir, a);
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectorPositions(true);
  Document doc = new Document();
  doc.add(new Field("field", "value", ft));
  
  IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
    w.addDocument(doc);
  });
  assertEquals("cannot store term vector positions for a field that is not indexed (field=\"field\")", expected.getMessage());
  
  w.close();
  dir.close();
}
 

public void testIllegalVectorOffsetsWithoutIndexed() throws Exception {
  Directory dir = newDirectory();
  MockAnalyzer a = new MockAnalyzer(random());
  a.setEnableChecks(false);
  RandomIndexWriter w = new RandomIndexWriter(random(), dir, a);
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectorOffsets(true);
  Document doc = new Document();
  doc.add(new Field("field", "value", ft));
  
  IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
    w.addDocument(doc);
  });
  assertEquals("cannot store term vector offsets for a field that is not indexed (field=\"field\")", expected.getMessage());
  
  w.close();
  dir.close();
}
 

public void testIllegalVectorPayloadsWithoutIndexed() throws Exception {
  Directory dir = newDirectory();
  MockAnalyzer a = new MockAnalyzer(random());
  a.setEnableChecks(false);
  RandomIndexWriter w = new RandomIndexWriter(random(), dir, a);
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectorPayloads(true);
  Document doc = new Document();
  doc.add(new Field("field", "value", ft));
  
  IllegalArgumentException expected = expectThrows(IllegalArgumentException.class, () -> {
    w.addDocument(doc);
  });
  assertEquals("cannot store term vector payloads for a field that is not indexed (field=\"field\")", expected.getMessage());

  w.close();
  dir.close();
}
 

/** test a null string value doesn't abort the entire segment */
public void testNullStoredFieldReuse() throws Exception {
  Directory dir = newDirectory();
  Analyzer analyzer = new MockAnalyzer(random());
  IndexWriter iw = new IndexWriter(dir, new IndexWriterConfig(analyzer));
  // add good document
  Document doc = new Document();
  Field theField = new StoredField("foo", "hello", StoredField.TYPE);
  doc.add(theField);
  iw.addDocument(doc);
  expectThrows(IllegalArgumentException.class, () -> {
    // set to null value
    theField.setStringValue(null);
    iw.addDocument(doc);
  });

  assertNull(iw.getTragicException());
  iw.close();
  // make sure we see our good doc
  DirectoryReader r = DirectoryReader.open(dir);
  assertEquals(1, r.numDocs());
  r.close();
  dir.close();
}
 

public void testNoAbortOnBadTVSettings() throws Exception {
  Directory dir = newDirectory();
  // Don't use RandomIndexWriter because we want to be sure both docs go to 1 seg:
  IndexWriterConfig iwc = new IndexWriterConfig(new MockAnalyzer(random()));
  IndexWriter iw = new IndexWriter(dir, iwc);

  Document doc = new Document();
  iw.addDocument(doc);
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectors(true);
  ft.freeze();
  doc.add(new Field("field", "value", ft));

  expectThrows(IllegalArgumentException.class, () -> {
    iw.addDocument(doc);
  });

  IndexReader r = DirectoryReader.open(iw);

  // Make sure the exc didn't lose our first document:
  assertEquals(1, r.numDocs());
  iw.close();
  r.close();
  dir.close();
}
 

@Override
public IndexableFieldType fieldType() {
  FieldType ft = new FieldType(StoredField.TYPE);
  ft.setStoreTermVectors(true);
  ft.freeze();
  return ft;
}
 

@Override
public IndexableFieldType fieldType() {
  return StoredField.TYPE;
}
 

@Override
public IndexableFieldType fieldType() {
  return StoredField.TYPE;
}
 

public void testIndexStoreCombos() throws Exception {
  Directory dir = newDirectory();
  IndexWriter w = new IndexWriter(dir, newIndexWriterConfig(new MockAnalyzer(random())));
  byte[] b = new byte[50];
  for(int i=0;i<50;i++)
    b[i] = (byte) (i+77);

  Document doc = new Document();

  FieldType customType = new FieldType(StoredField.TYPE);
  customType.setTokenized(true);

  Field f = new Field("binary", b, 10, 17, customType);
  // TODO: this is evil, changing the type after creating the field:
  customType.setIndexOptions(IndexOptions.DOCS);
  final MockTokenizer doc1field1 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc1field1.setReader(new StringReader("doc1field1"));
  f.setTokenStream(doc1field1);

  FieldType customType2 = new FieldType(TextField.TYPE_STORED);

  Field f2 = newField("string", "value", customType2);
  final MockTokenizer doc1field2 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc1field2.setReader(new StringReader("doc1field2"));
  f2.setTokenStream(doc1field2);
  doc.add(f);
  doc.add(f2);
  w.addDocument(doc);

  // add 2 docs to test in-memory merging
  final MockTokenizer doc2field1 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc2field1.setReader(new StringReader("doc2field1"));
  f.setTokenStream(doc2field1);
  final MockTokenizer doc2field2 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc2field2.setReader(new StringReader("doc2field2"));
  f2.setTokenStream(doc2field2);
  w.addDocument(doc);

  // force segment flush so we can force a segment merge with doc3 later.
  w.commit();

  final MockTokenizer doc3field1 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc3field1.setReader(new StringReader("doc3field1"));
  f.setTokenStream(doc3field1);
  final MockTokenizer doc3field2 = new MockTokenizer(MockTokenizer.WHITESPACE, false);
  doc3field2.setReader(new StringReader("doc3field2"));
  f2.setTokenStream(doc3field2);

  w.addDocument(doc);
  w.commit();
  w.forceMerge(1);   // force segment merge.
  w.close();

  IndexReader ir = DirectoryReader.open(dir);
  Document doc2 = ir.document(0);
  IndexableField f3 = doc2.getField("binary");
  b = f3.binaryValue().bytes;
  assertTrue(b != null);
  assertEquals(17, b.length, 17);
  assertEquals(87, b[0]);

  assertTrue(ir.document(0).getField("binary").binaryValue()!=null);
  assertTrue(ir.document(1).getField("binary").binaryValue()!=null);
  assertTrue(ir.document(2).getField("binary").binaryValue()!=null);

  assertEquals("value", ir.document(0).get("string"));
  assertEquals("value", ir.document(1).get("string"));
  assertEquals("value", ir.document(2).get("string"));


  // test that the terms were indexed.
  assertTrue(TestUtil.docs(random(), ir, "binary", new BytesRef("doc1field1"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertTrue(TestUtil.docs(random(), ir, "binary", new BytesRef("doc2field1"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertTrue(TestUtil.docs(random(), ir, "binary", new BytesRef("doc3field1"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertTrue(TestUtil.docs(random(), ir, "string", new BytesRef("doc1field2"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertTrue(TestUtil.docs(random(), ir, "string", new BytesRef("doc2field2"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
  assertTrue(TestUtil.docs(random(), ir, "string", new BytesRef("doc3field2"), null, PostingsEnum.NONE).nextDoc() != DocIdSetIterator.NO_MORE_DOCS);

  ir.close();
  dir.close();

}