/// <summary>Retrieve the length (in bytes) of the tvd and tvf
/// entries for the next numDocs starting with
/// startDocID. This is used for bulk copying when
/// merging segments, if the field numbers are
/// congruent. Once this returns, the tvf & tvd streams
/// are seeked to the startDocID.
/// </summary>
internal void RawDocs(int[] tvdLengths, int[] tvfLengths, int startDocID, int numDocs)
{
if (tvx == null)
{
for (int i = 0; i < tvdLengths.Length; i++)
{
tvdLengths[i] = 0;
}
for (int i = 0; i < tvfLengths.Length; i++)
{
tvfLengths[i] = 0;
}
return;
}
// SegmentMerger calls canReadRawDocs() first and should
// not call us if that returns false.
if (format < FORMAT_VERSION2)
{
throw new System.SystemException("cannot read raw docs with older term vector formats");
}
SeekTvx(startDocID);
long tvdPosition = tvx.ReadLong();
tvd.Seek(tvdPosition);
long tvfPosition = tvx.ReadLong();
tvf.Seek(tvfPosition);
long lastTvdPosition = tvdPosition;
long lastTvfPosition = tvfPosition;
int count = 0;
while (count < numDocs)
{
int docID = docStoreOffset + startDocID + count + 1;
System.Diagnostics.Debug.Assert(docID <= numTotalDocs);
if (docID < numTotalDocs)
{
tvdPosition = tvx.ReadLong();
tvfPosition = tvx.ReadLong();
}
else
{
tvdPosition = tvd.Length();
tvfPosition = tvf.Length();
System.Diagnostics.Debug.Assert(count == numDocs - 1);
}
tvdLengths[count] = (int)(tvdPosition - lastTvdPosition);
tvfLengths[count] = (int)(tvfPosition - lastTvfPosition);
count++;
lastTvdPosition = tvdPosition;
lastTvfPosition = tvfPosition;
}
}
private void SkipPayload()
{
if (needToLoadPayload && payloadLength > 0)
{
proxStream.Seek(proxStream.GetFilePointer() + payloadLength);
}
needToLoadPayload = false;
}
private void SeekTvx(int docNum)
{
if (format < FORMAT_VERSION2)
{
tvx.Seek((docNum + docStoreOffset) * 8L + FORMAT_SIZE);
}
else
{
tvx.Seek((docNum + docStoreOffset) * 16L + FORMAT_SIZE);
}
}
internal void Seek(long pointer, long p, Term t, TermInfo ti)
{
input.Seek(pointer);
position = p;
termBuffer.Set(t);
prevBuffer.Reset();
termInfo.Set(ti);
}
/// <summary>Expert: implements buffer refill. Reads bytes from the current
/// position in the input.
/// </summary>
/// <param name="b">the array to read bytes into
/// </param>
/// <param name="offset">the offset in the array to start storing bytes
/// </param>
/// <param name="len">the number of bytes to read
/// </param>
public override void ReadInternal(byte[] b, int offset, int len)
{
long start = GetFilePointer();
if (start + len > length)
{
throw new System.IO.IOException("read past EOF");
}
base_Renamed.Seek(fileOffset + start);
base_Renamed.ReadBytes(b, offset, len, false);
}
public override byte[] GetBinaryValue(byte[] result)
{
Enclosing_Instance.EnsureOpen();
if (isBinary)
{
if (fieldsData == null)
{
// Allocate new buffer if result is null or too small
byte[] b;
if (result == null || result.Length < toRead)
{
b = new byte[toRead];
}
else
{
b = result;
}
IndexInput localFieldsStream = GetFieldStream();
// Throw this IOException since IndexReader.document does so anyway, so probably not that big of a change for people
// since they are already handling this exception when getting the document
try
{
localFieldsStream.Seek(pointer);
localFieldsStream.ReadBytes(b, 0, toRead);
if (isCompressed == true)
{
fieldsData = Enclosing_Instance.Uncompress(b);
}
else
{
fieldsData = b;
}
}
catch (System.IO.IOException e)
{
throw new FieldReaderException(e);
}
binaryOffset = 0;
binaryLength = toRead;
}
return((byte[])fieldsData);
}
else
{
return(null);
}
}
internal virtual void Seek(TermInfo ti, Term term)
{
count = 0;
FieldInfo fi = parent.core.fieldInfos.FieldInfo(term.field);
currentFieldOmitTermFreqAndPositions = (fi != null)?fi.omitTermFreqAndPositions:false;
currentFieldStoresPayloads = (fi != null)?fi.storePayloads:false;
if (ti == null)
{
df = 0;
}
else
{
df = ti.docFreq;
doc = 0;
freqBasePointer = ti.freqPointer;
proxBasePointer = ti.proxPointer;
skipPointer = freqBasePointer + ti.skipOffset;
freqStream.Seek(freqBasePointer);
haveSkipped = false;
}
}
/// <summary>The value of the field as a String, or null. If null, the Reader value,
/// binary value, or TokenStream value is used. Exactly one of stringValue(),
/// readerValue(), binaryValue(), and tokenStreamValue() must be set.
/// </summary>
public override System.String StringValue()
{
Enclosing_Instance.EnsureOpen();
if (isBinary)
{
return(null);
}
else
{
if (fieldsData == null)
{
IndexInput localFieldsStream = GetFieldStream();
try
{
localFieldsStream.Seek(pointer);
if (isCompressed)
{
byte[] b = new byte[toRead];
localFieldsStream.ReadBytes(b, 0, b.Length);
fieldsData = System.Text.Encoding.GetEncoding("UTF-8").GetString(Enclosing_Instance.Uncompress(b));
}
else
{
if (Enclosing_Instance.format >= FieldsWriter.FORMAT_VERSION_UTF8_LENGTH_IN_BYTES)
{
byte[] bytes = new byte[toRead];
localFieldsStream.ReadBytes(bytes, 0, toRead);
fieldsData = System.Text.Encoding.GetEncoding("UTF-8").GetString(bytes);
}
else
{
//read in chars b/c we already know the length we need to read
char[] chars = new char[toRead];
localFieldsStream.ReadChars(chars, 0, toRead);
fieldsData = new System.String(chars);
}
}
}
catch (System.IO.IOException e)
{
throw new FieldReaderException(e);
}
}
return((System.String)fieldsData);
}
}
/// <summary> Construct a FieldInfos object using the directory and the name of the file
/// IndexInput
/// </summary>
/// <param name="d">The directory to open the IndexInput from
/// </param>
/// <param name="name">The name of the file to open the IndexInput from in the Directory
/// </param>
/// <throws> IOException </throws>
public /*internal*/ FieldInfos(Directory d, System.String name)
{
IndexInput input = d.OpenInput(name);
try
{
try
{
Read(input, name);
}
catch (System.IO.IOException ioe)
{
if (format == FORMAT_PRE)
{
// LUCENE-1623: FORMAT_PRE (before there was a
// format) may be 2.3.2 (pre-utf8) or 2.4.x (utf8)
// encoding; retry with input set to pre-utf8
input.Seek(0);
input.SetModifiedUTF8StringsMode();
byNumber.Clear();
byName.Clear();
try
{
Read(input, name);
}
catch (System.Exception t)
{
// Ignore any new exception & throw original IOE
throw ioe;
}
}
else
{
// The IOException cannot be caused by
// LUCENE-1623, so re-throw it
throw ioe;
}
}
}
finally
{
input.Close();
}
}
