private AppendingDeltaPackedLongBuffer PendingCounts; // termIDs per doc
#endregion Fields
#region Constructors
public SortedSetDocValuesWriter(FieldInfo fieldInfo, Counter iwBytesUsed)
{
this.FieldInfo = fieldInfo;
this.IwBytesUsed = iwBytesUsed;
Hash = new BytesRefHash(new ByteBlockPool(new ByteBlockPool.DirectTrackingAllocator(iwBytesUsed)), BytesRefHash.DEFAULT_CAPACITY, new DirectBytesStartArray(BytesRefHash.DEFAULT_CAPACITY, iwBytesUsed));
Pending = new AppendingPackedLongBuffer(PackedInts.COMPACT);
PendingCounts = new AppendingDeltaPackedLongBuffer(PackedInts.COMPACT);
BytesUsed = Pending.RamBytesUsed() + PendingCounts.RamBytesUsed();
iwBytesUsed.AddAndGet(BytesUsed);
}
public TermsHashPerField(DocInverterPerField docInverterPerField, TermsHash termsHash, TermsHash nextTermsHash, FieldInfo fieldInfo)
{
IntPool = termsHash.IntPool;
BytePool = termsHash.BytePool;
TermBytePool = termsHash.TermBytePool;
DocState = termsHash.DocState;
this.TermsHash = termsHash;
BytesUsed = termsHash.BytesUsed;
FieldState = docInverterPerField.FieldState;
this.Consumer = termsHash.Consumer.AddField(this, fieldInfo);
PostingsBytesStartArray byteStarts = new PostingsBytesStartArray(this, BytesUsed);
BytesHash = new BytesRefHash(TermBytePool, HASH_INIT_SIZE, byteStarts);
StreamCount = Consumer.StreamCount;
NumPostingInt = 2 * StreamCount;
this.FieldInfo = fieldInfo;
if (nextTermsHash != null)
{
NextPerField = (TermsHashPerField)nextTermsHash.AddField(docInverterPerField, fieldInfo);
}
else
{
NextPerField = null;
}
}
public void TestRandomSortedBytes()
{
Directory dir = NewDirectory();
IndexWriterConfig cfg = NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
if (!DefaultCodecSupportsDocsWithField())
{
// if the codec doesnt support missing, we expect missing to be mapped to byte[]
// by the impersonator, but we have to give it a chance to merge them to this
cfg.SetMergePolicy(NewLogMergePolicy());
}
RandomIndexWriter w = new RandomIndexWriter(Random(), dir, cfg);
int numDocs = AtLeast(100);
BytesRefHash hash = new BytesRefHash();
IDictionary<string, string> docToString = new Dictionary<string, string>();
int maxLength = TestUtil.NextInt(Random(), 1, 50);
for (int i = 0; i < numDocs; i++)
{
Document doc = new Document();
doc.Add(NewTextField("id", "" + i, Field.Store.YES));
string @string = TestUtil.RandomRealisticUnicodeString(Random(), 1, maxLength);
BytesRef br = new BytesRef(@string);
doc.Add(new SortedDocValuesField("field", br));
hash.Add(br);
docToString["" + i] = @string;
w.AddDocument(doc);
}
if (Rarely())
{
w.Commit();
}
int numDocsNoValue = AtLeast(10);
for (int i = 0; i < numDocsNoValue; i++)
{
Document doc = new Document();
doc.Add(NewTextField("id", "noValue", Field.Store.YES));
w.AddDocument(doc);
}
if (!DefaultCodecSupportsDocsWithField())
{
BytesRef bytesRef = new BytesRef();
hash.Add(bytesRef); // add empty value for the gaps
}
if (Rarely())
{
w.Commit();
}
if (!DefaultCodecSupportsDocsWithField())
{
// if the codec doesnt support missing, we expect missing to be mapped to byte[]
// by the impersonator, but we have to give it a chance to merge them to this
w.ForceMerge(1);
}
for (int i = 0; i < numDocs; i++)
{
Document doc = new Document();
string id = "" + i + numDocs;
doc.Add(NewTextField("id", id, Field.Store.YES));
string @string = TestUtil.RandomRealisticUnicodeString(Random(), 1, maxLength);
BytesRef br = new BytesRef(@string);
hash.Add(br);
docToString[id] = @string;
doc.Add(new SortedDocValuesField("field", br));
w.AddDocument(doc);
}
w.Commit();
IndexReader reader = w.Reader;
SortedDocValues docValues = MultiDocValues.GetSortedValues(reader, "field");
int[] sort = hash.Sort(BytesRef.UTF8SortedAsUnicodeComparer);
BytesRef expected = new BytesRef();
BytesRef actual = new BytesRef();
Assert.AreEqual(hash.Size(), docValues.ValueCount);
for (int i = 0; i < hash.Size(); i++)
{
hash.Get(sort[i], expected);
docValues.LookupOrd(i, actual);
Assert.AreEqual(expected.Utf8ToString(), actual.Utf8ToString());
int ord = docValues.LookupTerm(expected);
Assert.AreEqual(i, ord);
}
AtomicReader slowR = SlowCompositeReaderWrapper.Wrap(reader);
ISet<KeyValuePair<string, string>> entrySet = docToString.EntrySet();
foreach (KeyValuePair<string, string> entry in entrySet)
{
// pk lookup
DocsEnum termDocsEnum = slowR.TermDocsEnum(new Term("id", entry.Key));
int docId = termDocsEnum.NextDoc();
expected = new BytesRef(entry.Value);
docValues.Get(docId, actual);
Assert.AreEqual(expected, actual);
}
reader.Dispose();
w.Dispose();
dir.Dispose();
}
