public void TestBasic()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetMergePolicy(NewLogMergePolicy());
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, iwc);
KeyValuePair<List<string>, IDictionary<string, Document>> res = GenerateIndexDocuments(AtLeast(1000), true, false);
IDictionary<string, Document> docs = res.Value;
List<String> invalidDocTerms = res.Key;
foreach (Document doc in docs.Values)
{
writer.AddDocument(doc);
}
writer.Commit();
writer.Dispose();
IndexReader ir = DirectoryReader.Open(dir);
IDictionary dictionary = new DocumentDictionary(ir, FIELD_NAME, WEIGHT_FIELD_NAME, PAYLOAD_FIELD_NAME);
IInputIterator inputIterator = dictionary.EntryIterator;
BytesRef f;
while ((f = inputIterator.Next()) != null)
{
string field = f.Utf8ToString();
Document doc = docs.ContainsKey(field) ? docs[field] : null;
docs.Remove(field);
//Document doc = docs.Remove(f.Utf8ToString());
assertTrue(f.equals(new BytesRef(doc.Get(FIELD_NAME))));
IndexableField weightField = doc.GetField(WEIGHT_FIELD_NAME);
assertEquals(inputIterator.Weight, (weightField != null) ? Convert.ToInt64(weightField.NumericValue) : 0);
assertTrue(inputIterator.Payload.Equals(doc.GetField(PAYLOAD_FIELD_NAME).BinaryValue));
}
foreach (string invalidTerm in invalidDocTerms)
{
var invalid = docs[invalidTerm];
docs.Remove(invalidTerm);
assertNotNull(invalid);
}
assertTrue(!docs.Any());
ir.Dispose();
dir.Dispose();
}
public void TestEmptyReader()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random));
iwc.SetMergePolicy(NewLogMergePolicy());
// Make sure the index is created?
RandomIndexWriter writer = new RandomIndexWriter(Random, dir, iwc);
writer.Commit();
writer.Dispose();
IndexReader ir = DirectoryReader.Open(dir);
IDictionary dictionary = new DocumentDictionary(ir, FIELD_NAME, WEIGHT_FIELD_NAME, PAYLOAD_FIELD_NAME);
IInputIterator inputIterator = dictionary.GetEntryIterator();
assertNull(inputIterator.Next());
assertEquals(inputIterator.Weight, 0);
assertNull(inputIterator.Payload);
ir.Dispose();
dir.Dispose();
}
public void TestWithDeletions()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetMergePolicy(NewLogMergePolicy());
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, iwc);
KeyValuePair <List <string>, IDictionary <string, Document> > res = GenerateIndexDocuments(AtLeast(1000), false, false);
IDictionary <string, Document> docs = res.Value;
List <String> invalidDocTerms = res.Key;
Random rand = Random();
List <string> termsToDel = new List <string>();
foreach (Document doc in docs.Values)
{
IIndexableField f2 = doc.GetField(FIELD_NAME);
if (rand.nextBoolean() && f2 != null && !invalidDocTerms.Contains(f2.GetStringValue()))
{
termsToDel.Add(doc.Get(FIELD_NAME));
}
writer.AddDocument(doc);
}
writer.Commit();
Term[] delTerms = new Term[termsToDel.size()];
for (int i = 0; i < termsToDel.size(); i++)
{
delTerms[i] = new Term(FIELD_NAME, termsToDel[i]);
}
foreach (Term delTerm in delTerms)
{
writer.DeleteDocuments(delTerm);
}
writer.Commit();
writer.Dispose();
foreach (string termToDel in termsToDel)
{
var toDel = docs[termToDel];
assertTrue(toDel != null);
docs.Remove(termToDel);
}
IndexReader ir = DirectoryReader.Open(dir);
assertEquals(ir.NumDocs, docs.size());
IDictionary dictionary = new DocumentDictionary(ir, FIELD_NAME, WEIGHT_FIELD_NAME);
IInputIterator inputIterator = dictionary.GetEntryIterator();
BytesRef f;
while ((f = inputIterator.Next()) != null)
{
var field = f.Utf8ToString();
Document doc = docs.ContainsKey(field) ? docs[field] : null;
docs.Remove(field);
assertTrue(f.equals(new BytesRef(doc.Get(FIELD_NAME))));
IIndexableField weightField = doc.GetField(WEIGHT_FIELD_NAME);
assertEquals(inputIterator.Weight, (weightField != null) ? Convert.ToInt64(weightField.GetNumericValue()) : 0);
assertEquals(inputIterator.Payload, null);
}
foreach (string invalidTerm in invalidDocTerms)
{
var invalid = docs[invalidTerm];
docs.Remove(invalidTerm);
assertNotNull(invalid);
}
assertTrue(!docs.Any());
ir.Dispose();
dir.Dispose();
}
public void TestEmptyReader()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetMergePolicy(NewLogMergePolicy());
// Make sure the index is created?
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, iwc);
writer.Commit();
writer.Dispose();
IndexReader ir = DirectoryReader.Open(dir);
IDictionary dictionary = new DocumentDictionary(ir, FIELD_NAME, WEIGHT_FIELD_NAME, PAYLOAD_FIELD_NAME);
IInputIterator inputIterator = dictionary.EntryIterator;
assertNull(inputIterator.Next());
assertEquals(inputIterator.Weight, 0);
assertNull(inputIterator.Payload);
ir.Dispose();
dir.Dispose();
}
public void TestWithDeletions()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetMergePolicy(NewLogMergePolicy());
RandomIndexWriter writer = new RandomIndexWriter(Random(), dir, iwc);
KeyValuePair<List<string>, IDictionary<string, Document>> res = GenerateIndexDocuments(AtLeast(1000), false, false);
IDictionary<string, Document> docs = res.Value;
List<String> invalidDocTerms = res.Key;
Random rand = Random();
List<string> termsToDel = new List<string>();
foreach (Document doc in docs.Values)
{
IndexableField f2 = doc.GetField(FIELD_NAME);
if (rand.nextBoolean() && f2 != null && !invalidDocTerms.Contains(f2.StringValue))
{
termsToDel.Add(doc.Get(FIELD_NAME));
}
writer.AddDocument(doc);
}
writer.Commit();
Term[] delTerms = new Term[termsToDel.size()];
for (int i = 0; i < termsToDel.size(); i++)
{
delTerms[i] = new Term(FIELD_NAME, termsToDel[i]);
}
foreach (Term delTerm in delTerms)
{
writer.DeleteDocuments(delTerm);
}
writer.Commit();
writer.Dispose();
foreach (string termToDel in termsToDel)
{
var toDel = docs[termToDel];
assertTrue(toDel != null);
docs.Remove(termToDel);
}
IndexReader ir = DirectoryReader.Open(dir);
assertEquals(ir.NumDocs, docs.size());
IDictionary dictionary = new DocumentDictionary(ir, FIELD_NAME, WEIGHT_FIELD_NAME);
IInputIterator inputIterator = dictionary.EntryIterator;
BytesRef f;
while ((f = inputIterator.Next()) != null)
{
var field = f.Utf8ToString();
Document doc = docs.ContainsKey(field) ? docs[field] : null;
docs.Remove(field);
assertTrue(f.equals(new BytesRef(doc.Get(FIELD_NAME))));
IndexableField weightField = doc.GetField(WEIGHT_FIELD_NAME);
assertEquals(inputIterator.Weight, (weightField != null) ? Convert.ToInt64(weightField.NumericValue) : 0);
assertEquals(inputIterator.Payload, null);
}
foreach (string invalidTerm in invalidDocTerms)
{
var invalid = docs[invalidTerm];
docs.Remove(invalidTerm);
assertNotNull(invalid);
}
assertTrue(!docs.Any());
ir.Dispose();
dir.Dispose();
}
/// <summary>
/// Creates an iterator over term, weight and payload fields from the lucene
/// index. setting <code>withPayload</code> to false, implies an iterator
/// over only term and weight.
/// </summary>
public DocumentInputIterator(DocumentDictionary outerInstance, bool hasPayloads, bool hasContexts)
{
this.outerInstance = outerInstance;
this.hasPayloads = hasPayloads;
this.hasContexts = hasContexts;
docCount = outerInstance.reader.MaxDoc() - 1;
weightValues = (outerInstance.weightField != null) ? MultiDocValues.GetNumericValues(outerInstance.reader, outerInstance.weightField) : null;
liveDocs = (outerInstance.reader.Leaves().Count > 0) ? MultiFields.GetLiveDocs(outerInstance.reader) : null;
relevantFields = GetRelevantFields(new string[] { outerInstance.field, outerInstance.weightField, outerInstance.payloadField, outerInstance.contextsField });
}
