/// <summary>
/// Frees persistent resources used by this <see cref="Analyzer"/> </summary>
public virtual void Dispose()
{
if (storedValue != null)
{
storedValue.Dispose();
storedValue = null;
}
}
/// <summary>
/// Frees persistent resources used by this <see cref="Analyzer"/>
/// </summary>
protected virtual void Dispose(bool disposing)
{
if (disposing)
{
if (storedValue != null)
{
storedValue.Dispose();
storedValue = null;
}
}
}
/// <summary>Set the configuration parameters of this doc maker.</summary>
public virtual void SetConfig(Config config, ContentSource source)
{
this.m_config = config;
this.m_source = source;
bool stored = config.Get("doc.stored", false);
bool bodyStored = config.Get("doc.body.stored", stored);
bool tokenized = config.Get("doc.tokenized", true);
bool bodyTokenized = config.Get("doc.body.tokenized", tokenized);
bool norms = config.Get("doc.tokenized.norms", false);
bool bodyNorms = config.Get("doc.body.tokenized.norms", true);
bool termVec = config.Get("doc.term.vector", false);
bool termVecPositions = config.Get("doc.term.vector.positions", false);
bool termVecOffsets = config.Get("doc.term.vector.offsets", false);
m_valType = new FieldType(TextField.TYPE_NOT_STORED);
m_valType.IsStored = stored;
m_valType.IsTokenized = tokenized;
m_valType.OmitNorms = !norms;
m_valType.StoreTermVectors = termVec;
m_valType.StoreTermVectorPositions = termVecPositions;
m_valType.StoreTermVectorOffsets = termVecOffsets;
m_valType.Freeze();
m_bodyValType = new FieldType(TextField.TYPE_NOT_STORED);
m_bodyValType.IsStored = bodyStored;
m_bodyValType.IsTokenized = bodyTokenized;
m_bodyValType.OmitNorms = !bodyNorms;
m_bodyValType.StoreTermVectors = termVec;
m_bodyValType.StoreTermVectorPositions = termVecPositions;
m_bodyValType.StoreTermVectorOffsets = termVecOffsets;
m_bodyValType.Freeze();
storeBytes = config.Get("doc.store.body.bytes", false);
m_reuseFields = config.Get("doc.reuse.fields", true);
// In a multi-rounds run, it is important to reset DocState since settings
// of fields may change between rounds, and this is the only way to reset
// the cache of all threads.
docState = new DisposableThreadLocal <DocState>();
m_indexProperties = config.Get("doc.index.props", false);
updateDocIDLimit = config.Get("doc.random.id.limit", -1);
if (updateDocIDLimit != -1)
{
r = new Random(179);
}
}
internal SegmentCoreReaders(SegmentReader owner, Directory dir, SegmentCommitInfo si, IOContext context, int termsIndexDivisor)
{
fieldsReaderLocal = new AnonymousFieldsReaderLocal(this);
termVectorsLocal = new AnonymousTermVectorsLocal(this);
if (termsIndexDivisor == 0)
{
throw new ArgumentException("indexDivisor must be < 0 (don't load terms index) or greater than 0 (got 0)");
}
Codec codec = si.Info.Codec;
Directory cfsDir; // confusing name: if (cfs) its the cfsdir, otherwise its the segment's directory.
bool success = false;
try
{
if (si.Info.UseCompoundFile)
{
cfsDir = cfsReader = new CompoundFileDirectory(dir, IndexFileNames.SegmentFileName(si.Info.Name, "", IndexFileNames.COMPOUND_FILE_EXTENSION), context, false);
}
else
{
cfsReader = null;
cfsDir = dir;
}
FieldInfos fieldInfos = owner.FieldInfos;
this.termsIndexDivisor = termsIndexDivisor;
PostingsFormat format = codec.PostingsFormat;
SegmentReadState segmentReadState = new SegmentReadState(cfsDir, si.Info, fieldInfos, context, termsIndexDivisor);
// Ask codec for its Fields
fields = format.FieldsProducer(segmentReadState);
Debug.Assert(fields != null);
// ask codec for its Norms:
// TODO: since we don't write any norms file if there are no norms,
// kinda jaky to assume the codec handles the case of no norms file at all gracefully?!
if (fieldInfos.HasNorms)
{
normsProducer = codec.NormsFormat.NormsProducer(segmentReadState);
Debug.Assert(normsProducer != null);
}
else
{
normsProducer = null;
}
fieldsReaderOrig = si.Info.Codec.StoredFieldsFormat.FieldsReader(cfsDir, si.Info, fieldInfos, context);
if (fieldInfos.HasVectors) // open term vector files only as needed
{
termVectorsReaderOrig = si.Info.Codec.TermVectorsFormat.VectorsReader(cfsDir, si.Info, fieldInfos, context);
}
else
{
termVectorsReaderOrig = null;
}
success = true;
}
finally
{
if (!success)
{
DecRef();
}
}
}
internal SegmentCoreReaders(SegmentReader owner, Directory dir, SegmentCommitInfo si, IOContext context, int termsIndexDivisor)
{
fieldsReaderLocal = new AnonymousFieldsReaderLocal(this);
termVectorsLocal = new AnonymousTermVectorsLocal(this);
if (termsIndexDivisor == 0)
{
throw new System.ArgumentException("indexDivisor must be < 0 (don't load terms index) or greater than 0 (got 0)");
}
Codec codec = si.Info.Codec;
Directory cfsDir; // confusing name: if (cfs) its the cfsdir, otherwise its the segment's directory.
bool success = false;
try
{
if (si.Info.UseCompoundFile)
{
cfsDir = cfsReader = new CompoundFileDirectory(dir, IndexFileNames.SegmentFileName(si.Info.Name, "", IndexFileNames.COMPOUND_FILE_EXTENSION), context, false);
}
else
{
cfsReader = null;
cfsDir = dir;
}
FieldInfos fieldInfos = owner.FieldInfos;
this.termsIndexDivisor = termsIndexDivisor;
PostingsFormat format = codec.PostingsFormat;
SegmentReadState segmentReadState = new SegmentReadState(cfsDir, si.Info, fieldInfos, context, termsIndexDivisor);
// Ask codec for its Fields
fields = format.FieldsProducer(segmentReadState);
Debug.Assert(fields != null);
// ask codec for its Norms:
// TODO: since we don't write any norms file if there are no norms,
// kinda jaky to assume the codec handles the case of no norms file at all gracefully?!
if (fieldInfos.HasNorms)
{
normsProducer = codec.NormsFormat.NormsProducer(segmentReadState);
Debug.Assert(normsProducer != null);
}
else
{
normsProducer = null;
}
// LUCENENET TODO: EXCEPTIONS Not sure why this catch block is swallowing AccessViolationException,
// because it didn't exist in Lucene. Is it really needed? AVE is for protected memory...could
// this be needed because we are using unchecked??
#if !NETSTANDARD
try
{
#endif
fieldsReaderOrig = si.Info.Codec.StoredFieldsFormat.FieldsReader(cfsDir, si.Info, fieldInfos, context);
#if !NETSTANDARD
}
#pragma warning disable 168
catch (System.AccessViolationException ave)
#pragma warning restore 168
{
}
#endif
if (fieldInfos.HasVectors) // open term vector files only as needed
{
termVectorsReaderOrig = si.Info.Codec.TermVectorsFormat.VectorsReader(cfsDir, si.Info, fieldInfos, context);
}
else
{
termVectorsReaderOrig = null;
}
success = true;
}
finally
{
if (!success)
{
DecRef();
}
}
}
protected virtual void AssertEquals(RandomTokenStream tk, FieldType ft, Terms terms)
{
Assert.AreEqual(1, terms.DocCount);
int termCount = new JCG.HashSet <string>(tk.terms).Count;
Assert.AreEqual((long)termCount, terms.Count); // LUCENENET specific - cast required because types don't match (xUnit checks this)
Assert.AreEqual((long)termCount, terms.SumDocFreq); // LUCENENET specific - cast required because types don't match (xUnit checks this)
Assert.AreEqual(ft.StoreTermVectorPositions, terms.HasPositions);
Assert.AreEqual(ft.StoreTermVectorOffsets, terms.HasOffsets);
Assert.AreEqual(ft.StoreTermVectorPayloads && tk.HasPayloads(), terms.HasPayloads);
ISet <BytesRef> uniqueTerms = new JCG.HashSet <BytesRef>();
foreach (string term in tk.freqs.Keys)
{
uniqueTerms.Add(new BytesRef(term));
}
BytesRef[] sortedTerms = uniqueTerms.ToArray(/*new BytesRef[0]*/);
Array.Sort(sortedTerms, terms.Comparer);
TermsEnum termsEnum = terms.GetEnumerator(Random.NextBoolean() ? null : this.termsEnum.Value);
this.termsEnum.Value = termsEnum;
for (int i = 0; i < sortedTerms.Length; ++i)
{
Assert.IsTrue(termsEnum.MoveNext());
Assert.AreEqual(sortedTerms[i], termsEnum.Term);
Assert.AreEqual(1, termsEnum.DocFreq);
FixedBitSet bits = new FixedBitSet(1);
DocsEnum docsEnum = termsEnum.Docs(bits, Random.NextBoolean() ? null : this.docsEnum.Value);
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsEnum.NextDoc());
bits.Set(0);
docsEnum = termsEnum.Docs(Random.NextBoolean() ? bits : null, Random.NextBoolean() ? null : docsEnum);
Assert.IsNotNull(docsEnum);
Assert.AreEqual(0, docsEnum.NextDoc());
Assert.AreEqual(0, docsEnum.DocID);
Assert.AreEqual(tk.freqs[termsEnum.Term.Utf8ToString()], docsEnum.Freq);
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsEnum.NextDoc());
this.docsEnum.Value = docsEnum;
bits.Clear(0);
DocsAndPositionsEnum docsAndPositionsEnum = termsEnum.DocsAndPositions(bits, Random.NextBoolean() ? null : this.docsAndPositionsEnum.Value);
Assert.AreEqual(ft.StoreTermVectorOffsets || ft.StoreTermVectorPositions, docsAndPositionsEnum != null);
if (docsAndPositionsEnum != null)
{
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsAndPositionsEnum.NextDoc());
}
bits.Set(0);
docsAndPositionsEnum = termsEnum.DocsAndPositions(Random.NextBoolean() ? bits : null, Random.NextBoolean() ? null : docsAndPositionsEnum);
Assert.AreEqual(ft.StoreTermVectorOffsets || ft.StoreTermVectorPositions, docsAndPositionsEnum != null);
if (terms.HasPositions || terms.HasOffsets)
{
Assert.AreEqual(0, docsAndPositionsEnum.NextDoc());
int freq = docsAndPositionsEnum.Freq;
Assert.AreEqual(tk.freqs[termsEnum.Term.Utf8ToString()], freq);
if (docsAndPositionsEnum != null)
{
for (int k = 0; k < freq; ++k)
{
int position = docsAndPositionsEnum.NextPosition();
ISet <int> indexes;
if (terms.HasPositions)
{
indexes = tk.positionToTerms[position];
Assert.IsNotNull(indexes);
}
else
{
indexes = tk.startOffsetToTerms[docsAndPositionsEnum.StartOffset];
Assert.IsNotNull(indexes);
}
if (terms.HasPositions)
{
bool foundPosition = false;
foreach (int index in indexes)
{
if (tk.termBytes[index].Equals(termsEnum.Term) && tk.positions[index] == position)
{
foundPosition = true;
break;
}
}
Assert.IsTrue(foundPosition);
}
if (terms.HasOffsets)
{
bool foundOffset = false;
foreach (int index in indexes)
{
if (tk.termBytes[index].Equals(termsEnum.Term) && tk.startOffsets[index] == docsAndPositionsEnum.StartOffset && tk.endOffsets[index] == docsAndPositionsEnum.EndOffset)
{
foundOffset = true;
break;
}
}
Assert.IsTrue(foundOffset);
}
if (terms.HasPayloads)
{
bool foundPayload = false;
foreach (int index in indexes)
{
if (tk.termBytes[index].Equals(termsEnum.Term) && Equals(tk.payloads[index], docsAndPositionsEnum.GetPayload()))
{
foundPayload = true;
break;
}
}
Assert.IsTrue(foundPayload);
}
}
try
{
docsAndPositionsEnum.NextPosition();
Assert.Fail();
}
catch (Exception e) when(e.IsException())
{
// ok
}
catch (Exception e) when(e.IsAssertionError())
{
// ok
}
}
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsAndPositionsEnum.NextDoc());
}
this.docsAndPositionsEnum.Value = docsAndPositionsEnum;
}
Assert.IsFalse(termsEnum.MoveNext());
for (int i = 0; i < 5; ++i)
{
if (Random.NextBoolean())
{
Assert.IsTrue(termsEnum.SeekExact(RandomPicks.RandomFrom(Random, tk.termBytes)));
}
else
{
Assert.AreEqual(SeekStatus.FOUND, termsEnum.SeekCeil(RandomPicks.RandomFrom(Random, tk.termBytes)));
}
}
}
