Lucene++ - a full-featured, c++ search engine
API Documentation
Go to the documentation of this file.
7 #ifndef TERMVECTORSREADER_H
8 #define TERMVECTORSREADER_H
Collection< int64_t > readTvfPointers(int32_t fieldCount)
Reads the long[] offsets into TVF; you have to pre-seek tvx/tvd to the right point.
bool storingPositions
Definition: TermVectorsReader.h:132
IndexInputPtr tvd
Definition: TermVectorsReader.h:47
#define LUCENE_CLASS(Name)
Definition: LuceneObject.h:24
static const uint8_t STORE_POSITIONS_WITH_TERMVECTOR
Definition: TermVectorsReader.h:40
ParallelArrayTermVectorMapper()
IndexInputPtr tvf
Definition: TermVectorsReader.h:48
boost::shared_ptr< LuceneObject > LuceneObjectPtr
Definition: LuceneTypes.h:539
void seekTvx(int32_t docNum)
boost::shared_ptr< TermVectorMapper > TermVectorMapperPtr
Definition: LuceneTypes.h:254
IndexInputPtr getTvdStream()
Used for bulk copy when merging.
TermFreqVectorPtr materializeVector()
Construct the vector.
Collection< TermFreqVectorPtr > readTermVectors(int32_t docNum, Collection< String > fields, Collection< int64_t > tvfPointers)
boost::shared_ptr< FieldInfos > FieldInfosPtr
Definition: LuceneTypes.h:127
IndexInputPtr tvx
Definition: TermVectorsReader.h:46
void ConstructReader(const DirectoryPtr &d, const String &segment, const FieldInfosPtr &fieldInfos, int32_t readBufferSize, int32_t docStoreOffset, int32_t size)
virtual ~ParallelArrayTermVectorMapper()
IndexInputPtr getTvfStream()
Used for bulk copy when merging.
TermFreqVectorPtr get(int32_t docNum, const String &field)
Retrieve the term vector for the given document and field.
static const int32_t FORMAT_UTF8_LENGTH_IN_BYTES
Changed strings to UTF8 with length-in-bytes not length-in-chars.
Definition: TermVectorsReader.h:32
void get(int32_t docNum, const String &field, const TermVectorMapperPtr &mapper)
virtual ~TermVectorsReader()
Definition: AbstractAllTermDocs.h:12
Collection< Collection< int32_t > > positions
Definition: TermVectorsReader.h:128
int32_t _size
Definition: TermVectorsReader.h:49
Base class for all Lucene classes.
Definition: LuceneObject.h:31
Collection< Collection< TermVectorOffsetInfoPtr > > offsets
Definition: TermVectorsReader.h:129
void get(int32_t docNumber, const TermVectorMapperPtr &mapper)
Collection< String > terms
Definition: TermVectorsReader.h:123
boost::shared_ptr< IndexInput > IndexInputPtr
Definition: LuceneTypes.h:493
virtual void setExpectations(const String &field, int32_t numTerms, bool storeOffsets, bool storePositions)
Tell the mapper what to expect in regards to field, number of terms, offset and position storage....
int32_t currentPosition
Definition: TermVectorsReader.h:130
static const int32_t FORMAT_VERSION
NOTE: if you make a new format, it must be larger than the current format.
Definition: TermVectorsReader.h:22
boost::shared_ptr< TermFreqVector > TermFreqVectorPtr
Definition: LuceneTypes.h:237
String field
Definition: TermVectorsReader.h:133
virtual LuceneObjectPtr clone(const LuceneObjectPtr &other=LuceneObjectPtr())
Return clone of this object.
virtual void map(const String &term, int32_t frequency, Collection< TermVectorOffsetInfoPtr > offsets, Collection< int32_t > positions)
Map the Term Vector information into your own structure.
static const int32_t FORMAT_SIZE
The size in bytes that the FORMAT_VERSION will take up at the beginning of each file.
Definition: TermVectorsReader.h:38
static const int32_t FORMAT_VERSION2
Changes to speed up bulk merging of term vectors.
Definition: TermVectorsReader.h:29
int32_t format
Definition: TermVectorsReader.h:55
int32_t docStoreOffset
The docID offset where our docs begin in the index file. This will be 0 if we have our own private fi...
Definition: TermVectorsReader.h:53
TermVectorsReader(const DirectoryPtr &d, const String &segment, const FieldInfosPtr &fieldInfos, int32_t readBufferSize, int32_t docStoreOffset=-1, int32_t size=0)
void readTermVector(const String &field, int64_t tvfPointer, const TermVectorMapperPtr &mapper)
Models the existing parallel array structure.
Definition: TermVectorsReader.h:118
boost::shared_ptr< Directory > DirectoryPtr
Definition: LuceneTypes.h:489
static const int32_t FORMAT_CURRENT
NOTE: always change this if you switch to a new format.
Definition: TermVectorsReader.h:35
Collection< int32_t > termFreqs
Definition: TermVectorsReader.h:127
TermVectorsReader(const DirectoryPtr &d, const String &segment, const FieldInfosPtr &fieldInfos)
The TermVectorMapper can be used to map Term Vectors into your own structure instead of the parallel ...
Definition: TermVectorMapper.h:18
bool storingOffsets
Definition: TermVectorsReader.h:131
void readTermVectors(Collection< String > fields, Collection< int64_t > tvfPointers, const TermVectorMapperPtr &mapper)
FieldInfosPtr fieldInfos
Definition: TermVectorsReader.h:44
Collection< TermFreqVectorPtr > get(int32_t docNum)
Return all term vectors stored for this document or null if the could not be read in.
Definition: TermVectorsReader.h:14
int32_t checkValidFormat(const IndexInputPtr &in)
void rawDocs(Collection< int32_t > tvdLengths, Collection< int32_t > tvfLengths, int32_t startDocID, int32_t numDocs)
Retrieve the length (in bytes) of the tvd and tvf entries for the next numDocs starting with startDoc...
static const uint8_t STORE_OFFSET_WITH_TERMVECTOR
Definition: TermVectorsReader.h:41
int32_t numTotalDocs
Definition: TermVectorsReader.h:50
Collection< String > readFields(int32_t fieldCount)
Reads the String[] fields; you have to pre-seek tvd to the right point.
clucene.sourceforge.net