public override DocIdSetIterator GetIterator()
{
DocsEnum docsEnum = m_reader.GetTermDocsEnum(m_term);
if (docsEnum == null)
{
return(EmptyDocIdSet.Instance.GetIterator());
}
return(docsEnum);
}
public void TestNonExistingsField()
{
MemoryIndex mindex = new MemoryIndex(Random.nextBoolean(), Random.nextInt(50) * 1024 * 1024);
MockAnalyzer mockAnalyzer = new MockAnalyzer(Random);
mindex.AddField("field", "the quick brown fox", mockAnalyzer);
AtomicReader reader = (AtomicReader)mindex.CreateSearcher().IndexReader;
assertNull(reader.GetNumericDocValues("not-in-index"));
assertNull(reader.GetNormValues("not-in-index"));
assertNull(reader.GetTermDocsEnum(new Term("not-in-index", "foo")));
assertNull(reader.GetTermPositionsEnum(new Term("not-in-index", "foo")));
assertNull(reader.GetTerms("not-in-index"));
}
public override DocIdSet GetDocIdSet(AtomicReaderContext context, IBits acceptDocs)
{
bool nullBitset = Random.Next(10) == 5;
AtomicReader reader = context.AtomicReader;
DocsEnum termDocsEnum = reader.GetTermDocsEnum(new Term("field", "0"));
if (termDocsEnum == null)
{
return(null); // no docs -- return null
}
BitSet bitSet = new BitSet(reader.MaxDoc);
int d;
while ((d = termDocsEnum.NextDoc()) != DocsEnum.NO_MORE_DOCS)
{
bitSet.Set(d);
}
return(new DocIdSetAnonymousClass(this, nullBitset, reader, bitSet));
}
public override DocIdSetIterator GetIterator()
{
Assert.IsTrue(nullBitset, "iterator should not be called if bitset is present");
return(reader.GetTermDocsEnum(new Term("field", "0")));
}
/// <summary>
/// loads multi-value facet data. This method uses a workarea to prepare loading.
/// </summary>
/// <param name="fieldName"></param>
/// <param name="reader"></param>
/// <param name="listFactory"></param>
/// <param name="workArea"></param>
public virtual void Load(string fieldName, AtomicReader reader, TermListFactory listFactory, BoboSegmentReader.WorkArea workArea)
{
#if FEATURE_STRING_INTERN
string field = string.Intern(fieldName);
#else
string field = fieldName;
#endif
int maxdoc = reader.MaxDoc;
BigNestedInt32Array.BufferedLoader loader = GetBufferedLoader(maxdoc, workArea);
ITermValueList list = (listFactory == null ? (ITermValueList) new TermStringList() : listFactory.CreateTermList());
List <int> minIDList = new List <int>();
List <int> maxIDList = new List <int>();
List <int> freqList = new List <int>();
OpenBitSet bitset = new OpenBitSet(maxdoc + 1);
int negativeValueCount = GetNegativeValueCount(reader, field);
int t = 1; // valid term id starts from 1
list.Add(null);
minIDList.Add(-1);
maxIDList.Add(-1);
freqList.Add(0);
m_overflow = false;
Terms terms = reader.GetTerms(field);
if (terms != null)
{
TermsEnum termsEnum = terms.GetIterator(null);
BytesRef text;
while ((text = termsEnum.Next()) != null)
{
string strText = text.Utf8ToString();
list.Add(strText);
Term term = new Term(field, strText);
DocsEnum docsEnum = reader.GetTermDocsEnum(term);
int df = 0;
int minID = -1;
int maxID = -1;
int docID = -1;
int valId = (t - 1 < negativeValueCount) ? (negativeValueCount - t + 1) : t;
while ((docID = docsEnum.NextDoc()) != DocsEnum.NO_MORE_DOCS)
{
df++;
if (!loader.Add(docID, valId))
{
LogOverflow(fieldName);
}
minID = docID;
bitset.FastSet(docID);
while (docsEnum.NextDoc() != DocsEnum.NO_MORE_DOCS)
{
docID = docsEnum.DocID;
df++;
if (!loader.Add(docID, valId))
{
LogOverflow(fieldName);
}
bitset.FastSet(docID);
}
maxID = docID;
}
freqList.Add(df);
minIDList.Add(minID);
maxIDList.Add(maxID);
t++;
}
}
list.Seal();
try
{
m_nestedArray.Load(maxdoc + 1, loader);
}
catch (Exception e)
{
throw new RuntimeException("failed to load due to " + e.ToString(), e);
}
this.m_valArray = list;
this.m_freqs = freqList.ToArray();
this.m_minIDs = minIDList.ToArray();
this.m_maxIDs = maxIDList.ToArray();
int doc = 0;
while (doc < maxdoc && !m_nestedArray.Contains(doc, 0, true))
{
++doc;
}
if (doc < maxdoc)
{
this.m_minIDs[0] = doc;
doc = maxdoc - 1;
while (doc >= 0 && !m_nestedArray.Contains(doc, 0, true))
{
--doc;
}
this.m_maxIDs[0] = doc;
}
this.m_freqs[0] = maxdoc - (int)bitset.Cardinality();
}
public virtual void Load(string fieldName, AtomicReader reader, TermListFactory listFactory)
{
#if FEATURE_STRING_INTERN
string field = string.Intern(fieldName);
#else
string field = fieldName;
#endif
int maxDoc = reader.MaxDoc;
int dictValueCount = GetDictValueCount(reader, fieldName);
BigSegmentedArray order = NewInstance(dictValueCount, maxDoc);
this.m_orderArray = order;
List <int> minIDList = new List <int>();
List <int> maxIDList = new List <int>();
List <int> freqList = new List <int>();
int length = maxDoc + 1;
ITermValueList list = listFactory == null ? (ITermValueList) new TermStringList() : listFactory.CreateTermList();
int negativeValueCount = GetNegativeValueCount(reader, field);
int t = 1; // valid term id starts from 1
list.Add(null);
minIDList.Add(-1);
maxIDList.Add(-1);
freqList.Add(0);
int totalFreq = 0;
Terms terms = reader.GetTerms(field);
if (terms != null)
{
TermsEnum termsEnum = terms.GetIterator(null);
BytesRef text;
while ((text = termsEnum.Next()) != null)
{
// store term text
// we expect that there is at most one term per document
if (t >= length)
{
throw new RuntimeException("there are more terms than "
+ "documents in field \"" + field + "\", but it's impossible to sort on "
+ "tokenized fields");
}
string strText = text.Utf8ToString();
list.Add(strText);
Term term = new Term(field, strText);
DocsEnum docsEnum = reader.GetTermDocsEnum(term);
// freqList.add(termEnum.docFreq()); // doesn't take into account
// deldocs
int minID = -1;
int maxID = -1;
int docID = -1;
int df = 0;
int valId = (t - 1 < negativeValueCount) ? (negativeValueCount - t + 1) : t;
while ((docID = docsEnum.NextDoc()) != DocsEnum.NO_MORE_DOCS)
{
df++;
order.Add(docID, valId);
minID = docID;
while (docsEnum.NextDoc() != DocsEnum.NO_MORE_DOCS)
{
docID = docsEnum.DocID;
df++;
order.Add(docID, valId);
}
maxID = docID;
}
freqList.Add(df);
totalFreq += df;
minIDList.Add(minID);
maxIDList.Add(maxID);
t++;
}
}
list.Seal();
this.m_valArray = list;
this.m_freqs = freqList.ToArray();
this.m_minIDs = minIDList.ToArray();
this.m_maxIDs = maxIDList.ToArray();
int doc = 0;
while (doc < maxDoc && order.Get(doc) != 0)
{
++doc;
}
if (doc < maxDoc)
{
this.m_minIDs[0] = doc;
// Try to get the max
doc = maxDoc - 1;
while (doc >= 0 && order.Get(doc) != 0)
{
--doc;
}
this.m_maxIDs[0] = doc;
}
this.m_freqs[0] = reader.NumDocs - totalFreq;
}
public override void Load(string fieldName, AtomicReader reader, TermListFactory listFactory, BoboSegmentReader.WorkArea workArea)
{
#if FEATURE_STRING_INTERN
string field = string.Intern(fieldName);
#else
string field = fieldName;
#endif
int maxdoc = reader.MaxDoc;
BigNestedInt32Array.BufferedLoader loader = GetBufferedLoader(maxdoc, workArea);
BigNestedInt32Array.BufferedLoader weightLoader = GetBufferedLoader(maxdoc, null);
var list = (listFactory == null ? new TermStringList() : listFactory.CreateTermList());
List <int> minIDList = new List <int>();
List <int> maxIDList = new List <int>();
List <int> freqList = new List <int>();
OpenBitSet bitset = new OpenBitSet(maxdoc + 1);
int negativeValueCount = GetNegativeValueCount(reader, field);
int t = 1; // valid term id starts from 1
list.Add(null);
minIDList.Add(-1);
maxIDList.Add(-1);
freqList.Add(0);
m_overflow = false;
string pre = null;
int df = 0;
int minID = -1;
int maxID = -1;
int docID = -1;
int valId = 0;
Terms terms = reader.GetTerms(field);
if (terms != null)
{
TermsEnum termsEnum = terms.GetIterator(null);
BytesRef text;
while ((text = termsEnum.Next()) != null)
{
string strText = text.Utf8ToString();
string val = null;
int weight = 0;
string[] split = strText.Split(new char[] { '\0' }, StringSplitOptions.RemoveEmptyEntries);
if (split.Length > 1)
{
val = split[0];
weight = int.Parse(split[split.Length - 1]);
}
else
{
continue;
}
if (pre == null || !val.Equals(pre))
{
if (pre != null)
{
freqList.Add(df);
minIDList.Add(minID);
maxIDList.Add(maxID);
}
list.Add(val);
df = 0;
minID = -1;
maxID = -1;
valId = (t - 1 < negativeValueCount) ? (negativeValueCount - t + 1) : t;
t++;
}
Term term = new Term(field, strText);
DocsEnum docsEnum = reader.GetTermDocsEnum(term);
if (docsEnum != null)
{
while ((docID = docsEnum.NextDoc()) != DocsEnum.NO_MORE_DOCS)
{
df++;
if (!loader.Add(docID, valId))
{
LogOverflow(fieldName);
}
else
{
weightLoader.Add(docID, weight);
}
if (docID < minID)
{
minID = docID;
}
bitset.FastSet(docID);
while (docsEnum.NextDoc() != DocsEnum.NO_MORE_DOCS)
{
docID = docsEnum.DocID;
df++;
if (!loader.Add(docID, valId))
{
LogOverflow(fieldName);
}
else
{
weightLoader.Add(docID, weight);
}
bitset.FastSet(docID);
}
if (docID > maxID)
{
maxID = docID;
}
}
}
pre = val;
}
if (pre != null)
{
freqList.Add(df);
minIDList.Add(minID);
maxIDList.Add(maxID);
}
}
list.Seal();
try
{
m_nestedArray.Load(maxdoc + 1, loader);
m_weightArray.Load(maxdoc + 1, weightLoader);
}
catch (Exception e)
{
throw new RuntimeException("failed to load due to " + e.ToString(), e);
}
this.m_valArray = list;
this.m_freqs = freqList.ToArray();
this.m_minIDs = minIDList.ToArray();
this.m_maxIDs = maxIDList.ToArray();
int doc = 0;
while (doc < maxdoc && !m_nestedArray.Contains(doc, 0, true))
{
++doc;
}
if (doc < maxdoc)
{
this.m_minIDs[0] = doc;
doc = maxdoc - 1;
while (doc >= 0 && !m_nestedArray.Contains(doc, 0, true))
{
--doc;
}
this.m_maxIDs[0] = doc;
}
this.m_freqs[0] = maxdoc - (int)bitset.Cardinality();
}
