public SegmentTermDocs(IndexInput freqStream, TermInfosReader tis, FieldInfos fieldInfos)
{
this.FreqStream = (IndexInput)freqStream.Clone();
this.Tis = tis;
this.FieldInfos = fieldInfos;
SkipInterval = tis.SkipInterval;
MaxSkipLevels = tis.MaxSkipLevels;
}
public Lucene3xFields(Directory dir, FieldInfos fieldInfos, SegmentInfo info, IOContext context, int indexDivisor)
{
Si = info;
// NOTE: we must always load terms index, even for
// "sequential" scan during merging, because what is
// sequential to merger may not be to TermInfosReader
// since we do the surrogates dance:
if (indexDivisor < 0)
{
indexDivisor = -indexDivisor;
}
bool success = false;
try
{
TermInfosReader r = new TermInfosReader(dir, info.Name, fieldInfos, context, indexDivisor);
if (indexDivisor == -1)
{
TisNoIndex = r;
}
else
{
TisNoIndex = null;
Tis = r;
}
this.Context = context;
this.FieldInfos = fieldInfos;
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
FreqStream = dir.OpenInput(IndexFileNames.SegmentFileName(info.Name, "", Lucene3xPostingsFormat.FREQ_EXTENSION), context);
bool anyProx = false;
foreach (FieldInfo fi in fieldInfos)
{
if (fi.Indexed)
{
Fields[fi.Name] = fi;
PreTerms_[fi.Name] = new PreTerms(this, fi);
if (fi.FieldIndexOptions == FieldInfo.IndexOptions.DOCS_AND_FREQS_AND_POSITIONS)
{
anyProx = true;
}
}
}
if (anyProx)
{
ProxStream = dir.OpenInput(IndexFileNames.SegmentFileName(info.Name, "", Lucene3xPostingsFormat.PROX_EXTENSION), context);
}
else
{
ProxStream = null;
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
Dispose();
}
}
this.Dir = dir;
}
//private Directory cfsReader; // LUCENENET NOTE: cfsReader not used
public Lucene3xFields(Directory dir, FieldInfos fieldInfos, SegmentInfo info, IOContext context, int indexDivisor)
{
si = info;
// NOTE: we must always load terms index, even for
// "sequential" scan during merging, because what is
// sequential to merger may not be to TermInfosReader
// since we do the surrogates dance:
if (indexDivisor < 0)
{
indexDivisor = -indexDivisor;
}
bool success = false;
try
{
var r = new TermInfosReader(dir, info.Name, fieldInfos, context, indexDivisor);
if (indexDivisor == -1)
{
TisNoIndex = r;
}
else
{
TisNoIndex = null;
Tis = r;
}
this.context = context;
this.fieldInfos = fieldInfos;
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
FreqStream = dir.OpenInput(IndexFileNames.SegmentFileName(info.Name, "", Lucene3xPostingsFormat.FREQ_EXTENSION), context);
bool anyProx = false;
foreach (FieldInfo fi in fieldInfos)
{
if (fi.IsIndexed)
{
fields[fi.Name] = fi;
preTerms[fi.Name] = new PreTerms(this, fi);
if (fi.IndexOptions == IndexOptions.DOCS_AND_FREQS_AND_POSITIONS)
{
anyProx = true;
}
}
}
if (anyProx)
{
ProxStream = dir.OpenInput(IndexFileNames.SegmentFileName(info.Name, "", Lucene3xPostingsFormat.PROX_EXTENSION), context);
}
else
{
ProxStream = null;
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
Dispose();
}
}
this.dir = dir;
}
public override SeekStatus SeekCeil(BytesRef term)
{
if (DEBUG_SURROGATES)
{
Console.WriteLine("TE.seek target=" + UnicodeUtil.ToHexString(term.Utf8ToString()));
}
skipNext = false;
TermInfosReader tis = outerInstance.TermsDict;
Term t0 = new Term(fieldInfo.Name, term);
Debug.Assert(termEnum != null);
tis.SeekEnum(termEnum, t0, false);
Term t = termEnum.Term();
if (t != null && t.Field == internedFieldName && term.BytesEquals(t.Bytes))
{
// If we found an exact match, no need to do the
// surrogate dance
if (DEBUG_SURROGATES)
{
Console.WriteLine(" seek exact match");
}
current = t.Bytes;
return(SeekStatus.FOUND);
}
else if (t == null || t.Field != internedFieldName)
{
// TODO: maybe we can handle this like the next()
// into null? set term as prevTerm then dance?
if (DEBUG_SURROGATES)
{
Console.WriteLine(" seek hit EOF");
}
// We hit EOF; try end-case surrogate dance: if we
// find an E, try swapping in S, backwards:
scratchTerm.CopyBytes(term);
Debug.Assert(scratchTerm.Offset == 0);
for (int i = scratchTerm.Length - 1; i >= 0; i--)
{
if (IsHighBMPChar(scratchTerm.Bytes, i))
{
if (DEBUG_SURROGATES)
{
Console.WriteLine(" found E pos=" + i + "; try seek");
}
if (SeekToNonBMP(seekTermEnum, scratchTerm, i))
{
scratchTerm.CopyBytes(seekTermEnum.Term().Bytes);
outerInstance.TermsDict.SeekEnum(termEnum, seekTermEnum.Term(), false);
newSuffixStart = 1 + i;
DoPushes();
// Found a match
// TODO: faster seek?
current = termEnum.Term().Bytes;
return(SeekStatus.NOT_FOUND);
}
}
}
if (DEBUG_SURROGATES)
{
Console.WriteLine(" seek END");
}
current = null;
return(SeekStatus.END);
}
else
{
// We found a non-exact but non-null term; this one
// is fun -- just treat it like next, by pretending
// requested term was prev:
prevTerm.CopyBytes(term);
if (DEBUG_SURROGATES)
{
Console.WriteLine(" seek hit non-exact term=" + UnicodeUtil.ToHexString(t.Text()));
}
BytesRef br = t.Bytes;
Debug.Assert(br.Offset == 0);
SetNewSuffixStart(term, br);
SurrogateDance();
Term t2 = termEnum.Term();
if (t2 == null || t2.Field != internedFieldName)
{
// PreFlex codec interns field names; verify:
Debug.Assert(t2 == null || !t2.Field.Equals(internedFieldName, StringComparison.Ordinal));
current = null;
return(SeekStatus.END);
}
else
{
current = t2.Bytes;
Debug.Assert(!unicodeSortOrder || term.CompareTo(current) < 0, "term=" + UnicodeUtil.ToHexString(term.Utf8ToString()) + " vs current=" + UnicodeUtil.ToHexString(current.Utf8ToString()));
return(SeekStatus.NOT_FOUND);
}
}
}
/*
* SegmentTermPositions(SegmentReader p) {
* super(p);
* this.proxStream = null; // the proxStream will be cloned lazily when nextPosition() is called for the first time
* }
*/
public SegmentTermPositions(IndexInput freqStream, IndexInput proxStream, TermInfosReader tis, FieldInfos fieldInfos)
: base(freqStream, tis, fieldInfos)
{
this.proxStreamOrig = proxStream; // the proxStream will be cloned lazily when nextPosition() is called for the first time
}