/// <summary>
/// Returns <see cref="DocsEnum"/> for the specified term.
/// This will return <c>null</c> if either the field or
/// term does not exist.
/// </summary>
/// <seealso cref="TermsEnum.Docs(IBits, DocsEnum)"/>
public DocsEnum GetTermDocsEnum(Term term) // LUCENENET specific: Renamed from TermDocsEnum()
{
Debug.Assert(term.Field != null);
Debug.Assert(term.Bytes != null);
Fields fields = Fields;
if (fields != null)
{
Terms terms = fields.GetTerms(term.Field);
if (terms != null)
{
TermsEnum termsEnum = terms.GetIterator(null);
if (termsEnum.SeekExact(term.Bytes))
{
return(termsEnum.Docs(LiveDocs, null));
}
}
}
return(null);
}
/// <summary>
/// checks the terms enum sequentially
/// if deep is false, it does a 'shallow' test that doesnt go down to the docsenums
/// </summary>
public void AssertTermsEnumEquals(string info, IndexReader leftReader, TermsEnum leftTermsEnum, TermsEnum rightTermsEnum, bool deep)
{
BytesRef term;
Bits randomBits = new RandomBits(leftReader.MaxDoc, Random().NextDouble(), Random());
DocsAndPositionsEnum leftPositions = null;
DocsAndPositionsEnum rightPositions = null;
DocsEnum leftDocs = null;
DocsEnum rightDocs = null;
while ((term = leftTermsEnum.Next()) != null)
{
Assert.AreEqual(term, rightTermsEnum.Next(), info);
AssertTermStatsEquals(info, leftTermsEnum, rightTermsEnum);
if (deep)
{
AssertDocsAndPositionsEnumEquals(info, leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions));
AssertDocsAndPositionsEnumEquals(info, leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions));
AssertPositionsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions));
AssertPositionsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions));
// with freqs:
AssertDocsEnumEquals(info, leftDocs = leftTermsEnum.Docs(null, leftDocs), rightDocs = rightTermsEnum.Docs(null, rightDocs), true);
AssertDocsEnumEquals(info, leftDocs = leftTermsEnum.Docs(randomBits, leftDocs), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs), true);
// w/o freqs:
AssertDocsEnumEquals(info, leftDocs = leftTermsEnum.Docs(null, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(null, rightDocs, DocsEnum.FLAG_NONE), false);
AssertDocsEnumEquals(info, leftDocs = leftTermsEnum.Docs(randomBits, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs, DocsEnum.FLAG_NONE), false);
// with freqs:
AssertDocsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(null, leftDocs), rightDocs = rightTermsEnum.Docs(null, rightDocs), true);
AssertDocsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(randomBits, leftDocs), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs), true);
// w/o freqs:
AssertDocsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(null, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(null, rightDocs, DocsEnum.FLAG_NONE), false);
AssertDocsSkippingEquals(info, leftReader, leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(randomBits, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs, DocsEnum.FLAG_NONE), false);
}
}
Assert.IsNull(rightTermsEnum.Next(), info);
}
/// <summary>
/// checks the terms enum sequentially
/// if deep is false, it does a 'shallow' test that doesnt go down to the docsenums
/// </summary>
public virtual void AssertTermsEnum(TermsEnum leftTermsEnum, TermsEnum rightTermsEnum, bool deep)
{
BytesRef term;
Bits randomBits = new RandomBits(MAXDOC, Random().NextDouble(), Random());
DocsAndPositionsEnum leftPositions = null;
DocsAndPositionsEnum rightPositions = null;
DocsEnum leftDocs = null;
DocsEnum rightDocs = null;
while ((term = leftTermsEnum.Next()) != null)
{
Assert.AreEqual(term, rightTermsEnum.Next());
AssertTermStats(leftTermsEnum, rightTermsEnum);
if (deep)
{
// with payloads + off
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions));
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions));
// with payloads only
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsAndPositionsEnum.FLAG_PAYLOADS), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsAndPositionsEnum.FLAG_PAYLOADS));
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsAndPositionsEnum.FLAG_PAYLOADS), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsAndPositionsEnum.FLAG_PAYLOADS));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsAndPositionsEnum.FLAG_PAYLOADS), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsAndPositionsEnum.FLAG_PAYLOADS));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsAndPositionsEnum.FLAG_PAYLOADS), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsAndPositionsEnum.FLAG_PAYLOADS));
// with offsets only
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsAndPositionsEnum.FLAG_OFFSETS), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsAndPositionsEnum.FLAG_OFFSETS));
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsAndPositionsEnum.FLAG_OFFSETS), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsAndPositionsEnum.FLAG_OFFSETS));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsAndPositionsEnum.FLAG_OFFSETS), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsAndPositionsEnum.FLAG_OFFSETS));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsAndPositionsEnum.FLAG_OFFSETS), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsAndPositionsEnum.FLAG_OFFSETS));
// with positions only
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsEnum.FLAG_NONE), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsEnum.FLAG_NONE));
AssertDocsAndPositionsEnum(leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsEnum.FLAG_NONE), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsEnum.FLAG_NONE));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(null, leftPositions, DocsEnum.FLAG_NONE), rightPositions = rightTermsEnum.DocsAndPositions(null, rightPositions, DocsEnum.FLAG_NONE));
AssertPositionsSkipping(leftTermsEnum.DocFreq(), leftPositions = leftTermsEnum.DocsAndPositions(randomBits, leftPositions, DocsEnum.FLAG_NONE), rightPositions = rightTermsEnum.DocsAndPositions(randomBits, rightPositions, DocsEnum.FLAG_NONE));
// with freqs:
AssertDocsEnum(leftDocs = leftTermsEnum.Docs(null, leftDocs), rightDocs = rightTermsEnum.Docs(null, rightDocs));
AssertDocsEnum(leftDocs = leftTermsEnum.Docs(randomBits, leftDocs), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs));
// w/o freqs:
AssertDocsEnum(leftDocs = leftTermsEnum.Docs(null, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(null, rightDocs, DocsEnum.FLAG_NONE));
AssertDocsEnum(leftDocs = leftTermsEnum.Docs(randomBits, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs, DocsEnum.FLAG_NONE));
// with freqs:
AssertDocsSkipping(leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(null, leftDocs), rightDocs = rightTermsEnum.Docs(null, rightDocs));
AssertDocsSkipping(leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(randomBits, leftDocs), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs));
// w/o freqs:
AssertDocsSkipping(leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(null, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(null, rightDocs, DocsEnum.FLAG_NONE));
AssertDocsSkipping(leftTermsEnum.DocFreq(), leftDocs = leftTermsEnum.Docs(randomBits, leftDocs, DocsEnum.FLAG_NONE), rightDocs = rightTermsEnum.Docs(randomBits, rightDocs, DocsEnum.FLAG_NONE));
}
}
Assert.IsNull(rightTermsEnum.Next());
}
public virtual void TestPositionReader()
{
TermVectorsReader reader = Codec.Default.TermVectorsFormat().VectorsReader(Dir, Seg.Info, FieldInfos, NewIOContext(Random()));
BytesRef[] terms;
Terms vector = reader.Get(0).Terms(TestFields[0]);
Assert.IsNotNull(vector);
Assert.AreEqual(TestTerms.Length, vector.Size());
TermsEnum termsEnum = vector.Iterator(null);
DocsAndPositionsEnum dpEnum = null;
for (int i = 0; i < TestTerms.Length; i++)
{
BytesRef text = termsEnum.Next();
Assert.IsNotNull(text);
string term = text.Utf8ToString();
//System.out.println("Term: " + term);
Assert.AreEqual(TestTerms[i], term);
dpEnum = termsEnum.DocsAndPositions(null, dpEnum);
Assert.IsNotNull(dpEnum);
int doc = dpEnum.DocID();
Assert.AreEqual(-1, doc);
Assert.IsTrue(dpEnum.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(dpEnum.Freq(), Positions[i].Length);
for (int j = 0; j < Positions[i].Length; j++)
{
Assert.AreEqual(Positions[i][j], dpEnum.NextPosition());
}
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dpEnum.NextDoc());
dpEnum = termsEnum.DocsAndPositions(null, dpEnum);
doc = dpEnum.DocID();
Assert.AreEqual(-1, doc);
Assert.IsTrue(dpEnum.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.IsNotNull(dpEnum);
Assert.AreEqual(dpEnum.Freq(), Positions[i].Length);
for (int j = 0; j < Positions[i].Length; j++)
{
Assert.AreEqual(Positions[i][j], dpEnum.NextPosition());
Assert.AreEqual(j * 10, dpEnum.StartOffset());
Assert.AreEqual(j * 10 + TestTerms[i].Length, dpEnum.EndOffset());
}
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dpEnum.NextDoc());
}
Terms freqVector = reader.Get(0).Terms(TestFields[1]); //no pos, no offset
Assert.IsNotNull(freqVector);
Assert.AreEqual(TestTerms.Length, freqVector.Size());
termsEnum = freqVector.Iterator(null);
Assert.IsNotNull(termsEnum);
for (int i = 0; i < TestTerms.Length; i++)
{
BytesRef text = termsEnum.Next();
Assert.IsNotNull(text);
string term = text.Utf8ToString();
//System.out.println("Term: " + term);
Assert.AreEqual(TestTerms[i], term);
Assert.IsNotNull(termsEnum.Docs(null, null));
Assert.IsNull(termsEnum.DocsAndPositions(null, null)); // no pos
}
reader.Dispose();
}
private void VerifyEnum(ThreadState threadState, string field, BytesRef term, TermsEnum termsEnum, FieldInfo.IndexOptions maxTestOptions, FieldInfo.IndexOptions maxIndexOptions, ISet<Option> options, bool alwaysTestMax)
// Maximum options (docs/freqs/positions/offsets) to test:
{
if (VERBOSE)
{
Console.WriteLine(" verifyEnum: options=" + options + " maxTestOptions=" + maxTestOptions);
}
// Make sure TermsEnum really is positioned on the
// expected term:
Assert.AreEqual(term, termsEnum.Term());
// 50% of the time time pass liveDocs:
bool useLiveDocs = options.Contains(Option.LIVE_DOCS) && Random().NextBoolean();
Bits liveDocs;
if (useLiveDocs)
{
liveDocs = GlobalLiveDocs;
if (VERBOSE)
{
Console.WriteLine(" use liveDocs");
}
}
else
{
liveDocs = null;
if (VERBOSE)
{
Console.WriteLine(" no liveDocs");
}
}
FieldInfo fieldInfo = CurrentFieldInfos.FieldInfo(field);
// NOTE: can be empty list if we are using liveDocs:
SeedPostings expected = GetSeedPostings(term.Utf8ToString(), Fields[field][term], useLiveDocs, maxIndexOptions);
Assert.AreEqual(expected.DocFreq, termsEnum.DocFreq());
bool allowFreqs = fieldInfo.FieldIndexOptions >= FieldInfo.IndexOptions.DOCS_AND_FREQS && maxTestOptions.CompareTo(FieldInfo.IndexOptions.DOCS_AND_FREQS) >= 0;
bool doCheckFreqs = allowFreqs && (alwaysTestMax || Random().Next(3) <= 2);
bool allowPositions = fieldInfo.FieldIndexOptions >= FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS && maxTestOptions.CompareTo(FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS) >= 0;
bool doCheckPositions = allowPositions && (alwaysTestMax || Random().Next(3) <= 2);
bool allowOffsets = fieldInfo.FieldIndexOptions >= FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS && maxTestOptions.CompareTo(FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS) >= 0;
bool doCheckOffsets = allowOffsets && (alwaysTestMax || Random().Next(3) <= 2);
bool doCheckPayloads = options.Contains(Option.PAYLOADS) && allowPositions && fieldInfo.HasPayloads() && (alwaysTestMax || Random().Next(3) <= 2);
DocsEnum prevDocsEnum = null;
DocsEnum docsEnum;
DocsAndPositionsEnum docsAndPositionsEnum;
if (!doCheckPositions)
{
if (allowPositions && Random().Next(10) == 7)
{
// 10% of the time, even though we will not check positions, pull a DocsAndPositions enum
if (options.Contains(Option.REUSE_ENUMS) && Random().Next(10) < 9)
{
prevDocsEnum = threadState.ReuseDocsAndPositionsEnum;
}
int flags = 0;
if (alwaysTestMax || Random().NextBoolean())
{
flags |= DocsAndPositionsEnum.FLAG_OFFSETS;
}
if (alwaysTestMax || Random().NextBoolean())
{
flags |= DocsAndPositionsEnum.FLAG_PAYLOADS;
}
if (VERBOSE)
{
Console.WriteLine(" get DocsAndPositionsEnum (but we won't check positions) flags=" + flags);
}
threadState.ReuseDocsAndPositionsEnum = termsEnum.DocsAndPositions(liveDocs, (DocsAndPositionsEnum)prevDocsEnum, flags);
docsEnum = threadState.ReuseDocsAndPositionsEnum;
docsAndPositionsEnum = threadState.ReuseDocsAndPositionsEnum;
}
else
{
if (VERBOSE)
{
Console.WriteLine(" get DocsEnum");
}
if (options.Contains(Option.REUSE_ENUMS) && Random().Next(10) < 9)
{
prevDocsEnum = threadState.ReuseDocsEnum;
}
threadState.ReuseDocsEnum = termsEnum.Docs(liveDocs, prevDocsEnum, doCheckFreqs ? DocsEnum.FLAG_FREQS : DocsEnum.FLAG_NONE);
docsEnum = threadState.ReuseDocsEnum;
docsAndPositionsEnum = null;
}
}
else
{
if (options.Contains(Option.REUSE_ENUMS) && Random().Next(10) < 9)
{
prevDocsEnum = threadState.ReuseDocsAndPositionsEnum;
}
int flags = 0;
if (alwaysTestMax || doCheckOffsets || Random().Next(3) == 1)
{
flags |= DocsAndPositionsEnum.FLAG_OFFSETS;
}
if (alwaysTestMax || doCheckPayloads || Random().Next(3) == 1)
{
flags |= DocsAndPositionsEnum.FLAG_PAYLOADS;
}
if (VERBOSE)
{
Console.WriteLine(" get DocsAndPositionsEnum flags=" + flags);
}
threadState.ReuseDocsAndPositionsEnum = termsEnum.DocsAndPositions(liveDocs, (DocsAndPositionsEnum)prevDocsEnum, flags);
docsEnum = threadState.ReuseDocsAndPositionsEnum;
docsAndPositionsEnum = threadState.ReuseDocsAndPositionsEnum;
}
Assert.IsNotNull(docsEnum, "null DocsEnum");
int initialDocID = docsEnum.DocID();
Assert.AreEqual(-1, initialDocID, "inital docID should be -1" + docsEnum);
if (VERBOSE)
{
if (prevDocsEnum == null)
{
Console.WriteLine(" got enum=" + docsEnum);
}
else if (prevDocsEnum == docsEnum)
{
Console.WriteLine(" got reuse enum=" + docsEnum);
}
else
{
Console.WriteLine(" got enum=" + docsEnum + " (reuse of " + prevDocsEnum + " failed)");
}
}
// 10% of the time don't consume all docs:
int stopAt;
if (!alwaysTestMax && options.Contains(Option.PARTIAL_DOC_CONSUME) && expected.DocFreq > 1 && Random().Next(10) == 7)
{
stopAt = Random().Next(expected.DocFreq - 1);
if (VERBOSE)
{
Console.WriteLine(" will not consume all docs (" + stopAt + " vs " + expected.DocFreq + ")");
}
}
else
{
stopAt = expected.DocFreq;
if (VERBOSE)
{
Console.WriteLine(" consume all docs");
}
}
double skipChance = alwaysTestMax ? 0.5 : Random().NextDouble();
int numSkips = expected.DocFreq < 3 ? 1 : TestUtil.NextInt(Random(), 1, Math.Min(20, expected.DocFreq / 3));
int skipInc = expected.DocFreq / numSkips;
int skipDocInc = MaxDoc / numSkips;
// Sometimes do 100% skipping:
bool doAllSkipping = options.Contains(Option.SKIPPING) && Random().Next(7) == 1;
double freqAskChance = alwaysTestMax ? 1.0 : Random().NextDouble();
double payloadCheckChance = alwaysTestMax ? 1.0 : Random().NextDouble();
double offsetCheckChance = alwaysTestMax ? 1.0 : Random().NextDouble();
if (VERBOSE)
{
if (options.Contains(Option.SKIPPING))
{
Console.WriteLine(" skipChance=" + skipChance + " numSkips=" + numSkips);
}
else
{
Console.WriteLine(" no skipping");
}
if (doCheckFreqs)
{
Console.WriteLine(" freqAskChance=" + freqAskChance);
}
if (doCheckPayloads)
{
Console.WriteLine(" payloadCheckChance=" + payloadCheckChance);
}
if (doCheckOffsets)
{
Console.WriteLine(" offsetCheckChance=" + offsetCheckChance);
}
}
while (expected.Upto <= stopAt)
{
if (expected.Upto == stopAt)
{
if (stopAt == expected.DocFreq)
{
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsEnum.NextDoc(), "DocsEnum should have ended but didn't");
// Common bug is to forget to set this.Doc=NO_MORE_DOCS in the enum!:
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsEnum.DocID(), "DocsEnum should have ended but didn't");
}
break;
}
if (options.Contains(Option.SKIPPING) && (doAllSkipping || Random().NextDouble() <= skipChance))
{
int targetDocID = -1;
if (expected.Upto < stopAt && Random().NextBoolean())
{
// Pick target we know exists:
int skipCount = TestUtil.NextInt(Random(), 1, skipInc);
for (int skip = 0; skip < skipCount; skip++)
{
if (expected.NextDoc() == DocsEnum.NO_MORE_DOCS)
{
break;
}
}
}
else
{
// Pick random target (might not exist):
int skipDocIDs = TestUtil.NextInt(Random(), 1, skipDocInc);
if (skipDocIDs > 0)
{
targetDocID = expected.DocID() + skipDocIDs;
expected.Advance(targetDocID);
}
}
if (expected.Upto >= stopAt)
{
int target = Random().NextBoolean() ? MaxDoc : DocsEnum.NO_MORE_DOCS;
if (VERBOSE)
{
Console.WriteLine(" now advance to end (target=" + target + ")");
}
Assert.AreEqual(DocsEnum.NO_MORE_DOCS, docsEnum.Advance(target), "DocsEnum should have ended but didn't");
break;
}
else
{
if (VERBOSE)
{
if (targetDocID != -1)
{
Console.WriteLine(" now advance to random target=" + targetDocID + " (" + expected.Upto + " of " + stopAt + ") current=" + docsEnum.DocID());
}
else
{
Console.WriteLine(" now advance to known-exists target=" + expected.DocID() + " (" + expected.Upto + " of " + stopAt + ") current=" + docsEnum.DocID());
}
}
int docID = docsEnum.Advance(targetDocID != -1 ? targetDocID : expected.DocID());
Assert.AreEqual(expected.DocID(), docID, "docID is wrong");
}
}
else
{
expected.NextDoc();
if (VERBOSE)
{
Console.WriteLine(" now nextDoc to " + expected.DocID() + " (" + expected.Upto + " of " + stopAt + ")");
}
int docID = docsEnum.NextDoc();
Assert.AreEqual(expected.DocID(), docID, "docID is wrong");
if (docID == DocsEnum.NO_MORE_DOCS)
{
break;
}
}
if (doCheckFreqs && Random().NextDouble() <= freqAskChance)
{
if (VERBOSE)
{
Console.WriteLine(" now freq()=" + expected.Freq());
}
int freq = docsEnum.Freq();
Assert.AreEqual(expected.Freq(), freq, "freq is wrong");
}
if (doCheckPositions)
{
int freq = docsEnum.Freq();
int numPosToConsume;
if (!alwaysTestMax && options.Contains(Option.PARTIAL_POS_CONSUME) && Random().Next(5) == 1)
{
numPosToConsume = Random().Next(freq);
}
else
{
numPosToConsume = freq;
}
for (int i = 0; i < numPosToConsume; i++)
{
int pos = expected.NextPosition();
if (VERBOSE)
{
Console.WriteLine(" now nextPosition to " + pos);
}
Assert.AreEqual(pos, docsAndPositionsEnum.NextPosition(), "position is wrong");
if (doCheckPayloads)
{
BytesRef expectedPayload = expected.Payload;
if (Random().NextDouble() <= payloadCheckChance)
{
if (VERBOSE)
{
Console.WriteLine(" now check expectedPayload length=" + (expectedPayload == null ? 0 : expectedPayload.Length));
}
if (expectedPayload == null || expectedPayload.Length == 0)
{
Assert.IsNull(docsAndPositionsEnum.Payload, "should not have payload");
}
else
{
BytesRef payload = docsAndPositionsEnum.Payload;
Assert.IsNotNull(payload, "should have payload but doesn't");
Assert.AreEqual(expectedPayload.Length, payload.Length, "payload length is wrong");
for (int byteUpto = 0; byteUpto < expectedPayload.Length; byteUpto++)
{
Assert.AreEqual(expectedPayload.Bytes[expectedPayload.Offset + byteUpto], payload.Bytes[payload.Offset + byteUpto], "payload bytes are wrong");
}
// make a deep copy
payload = BytesRef.DeepCopyOf(payload);
Assert.AreEqual(payload, docsAndPositionsEnum.Payload, "2nd call to getPayload returns something different!");
}
}
else
{
if (VERBOSE)
{
Console.WriteLine(" skip check payload length=" + (expectedPayload == null ? 0 : expectedPayload.Length));
}
}
}
if (doCheckOffsets)
{
if (Random().NextDouble() <= offsetCheckChance)
{
if (VERBOSE)
{
Console.WriteLine(" now check offsets: startOff=" + expected.StartOffset() + " endOffset=" + expected.EndOffset());
}
Assert.AreEqual(expected.StartOffset(), docsAndPositionsEnum.StartOffset(), "startOffset is wrong");
Assert.AreEqual(expected.EndOffset(), docsAndPositionsEnum.EndOffset(), "endOffset is wrong");
}
else
{
if (VERBOSE)
{
Console.WriteLine(" skip check offsets");
}
}
}
else if (fieldInfo.FieldIndexOptions < FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS)
{
if (VERBOSE)
{
Console.WriteLine(" now check offsets are -1");
}
Assert.AreEqual(-1, docsAndPositionsEnum.StartOffset(), "startOffset isn't -1");
Assert.AreEqual(-1, docsAndPositionsEnum.EndOffset(), "endOffset isn't -1");
}
}
}
}
}
public override DocsEnum Docs(Bits bits, DocsEnum reuse, int flags)
{
return(Tenum.Docs(bits, reuse, flags));
}
public override DocsEnum Docs(IBits bits, DocsEnum reuse, DocsFlags flags)
{
return(tenum.Docs(bits, reuse, flags));
}
public virtual void TestPositionReader()
{
TermVectorsReader reader = Codec.Default.TermVectorsFormat.VectorsReader(dir, seg.Info, fieldInfos, NewIOContext(Random));
//BytesRef[] terms; // LUCENENET NOTE: Not used in Lucene
Terms vector = reader.Get(0).GetTerms(testFields[0]);
Assert.IsNotNull(vector);
Assert.AreEqual(testTerms.Length, vector.Count);
TermsEnum termsEnum = vector.GetEnumerator();
DocsAndPositionsEnum dpEnum = null;
for (int i = 0; i < testTerms.Length; i++)
{
Assert.IsTrue(termsEnum.MoveNext());
BytesRef text = termsEnum.Term;
string term = text.Utf8ToString();
//System.out.println("Term: " + term);
Assert.AreEqual(testTerms[i], term);
dpEnum = termsEnum.DocsAndPositions(null, dpEnum);
Assert.IsNotNull(dpEnum);
int doc = dpEnum.DocID;
Assert.AreEqual(-1, doc);
Assert.IsTrue(dpEnum.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(dpEnum.Freq, positions[i].Length);
for (int j = 0; j < positions[i].Length; j++)
{
Assert.AreEqual(positions[i][j], dpEnum.NextPosition());
}
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dpEnum.NextDoc());
dpEnum = termsEnum.DocsAndPositions(null, dpEnum);
doc = dpEnum.DocID;
Assert.AreEqual(-1, doc);
Assert.IsTrue(dpEnum.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.IsNotNull(dpEnum);
Assert.AreEqual(dpEnum.Freq, positions[i].Length);
for (int j = 0; j < positions[i].Length; j++)
{
Assert.AreEqual(positions[i][j], dpEnum.NextPosition());
Assert.AreEqual(j * 10, dpEnum.StartOffset);
Assert.AreEqual(j * 10 + testTerms[i].Length, dpEnum.EndOffset);
}
Assert.AreEqual(DocIdSetIterator.NO_MORE_DOCS, dpEnum.NextDoc());
}
Terms freqVector = reader.Get(0).GetTerms(testFields[1]); //no pos, no offset
Assert.IsNotNull(freqVector);
Assert.AreEqual(testTerms.Length, freqVector.Count);
termsEnum = freqVector.GetEnumerator();
Assert.IsNotNull(termsEnum);
for (int i = 0; i < testTerms.Length; i++)
{
Assert.IsTrue(termsEnum.MoveNext());
BytesRef text = termsEnum.Term;
string term = text.Utf8ToString();
//System.out.println("Term: " + term);
Assert.AreEqual(testTerms[i], term);
Assert.IsNotNull(termsEnum.Docs(null, null));
Assert.IsNull(termsEnum.DocsAndPositions(null, null)); // no pos
}
reader.Dispose();
}
protected virtual void FillDocsAndScores(FixedBitSet matchingDocs, Bits acceptDocs,
TermsEnum termsEnum)
{
BytesRef spare = new BytesRef();
DocsEnum docsEnum = null;
for (int i = 0; i < outerInstance._terms.Size(); i++)
{
if (termsEnum.SeekExact(outerInstance._terms.Get(outerInstance._ords[i], spare)))
{
docsEnum = termsEnum.Docs(acceptDocs, docsEnum, FLAG_NONE);
float score = outerInstance._scores[outerInstance._ords[i]];
for (int doc = docsEnum.NextDoc();
doc != NO_MORE_DOCS;
doc = docsEnum.NextDoc())
{
matchingDocs.Set(doc);
// In the case the same doc is also related to a another doc, a score might be overwritten. I think this
// can only happen in a many-to-many relation
scores[doc] = score;
}
}
}
}
protected override void FillDocsAndScores(FixedBitSet matchingDocs, Bits acceptDocs,
TermsEnum termsEnum)
{
BytesRef spare = new BytesRef();
DocsEnum docsEnum = null;
for (int i = 0; i < outerInstance._terms.Size(); i++)
{
if (termsEnum.SeekExact(outerInstance._terms.Get(outerInstance._ords[i], spare)))
{
docsEnum = termsEnum.Docs(acceptDocs, docsEnum, FLAG_NONE);
float score = outerInstance._scores[outerInstance._ords[i]];
for (int doc = docsEnum.NextDoc();
doc != NO_MORE_DOCS;
doc = docsEnum.NextDoc())
{
// I prefer this:
/*if (scores[doc] < score) {
scores[doc] = score;
matchingDocs.set(doc);
}*/
// But this behaves the same as MVInnerScorer and only then the tests will pass:
if (!matchingDocs.Get(doc))
{
scores[doc] = score;
matchingDocs.Set(doc);
}
}
}
}
}
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)));
}
}
}