term vectors format.
Very similarly to
Lucene90StoredFieldsFormat, this format is based on compressed
chunks of data, with document-level granularity so that a document can never span across distinct
chunks. Moreover, data is made as compact as possible:
- textual data is compressed using the very light, LZ4 compression algorithm,
- binary data is written using fixed-size blocks of
Term vectors are stored using two files
- a data file where terms, frequencies, positions, offsets and payloads are stored,
- an index file, loaded into memory, used to locate specific documents in the data file.
A vector metadata file (extension
- VectorMeta (.tvm) --> <Header>, PackedIntsVersion, ChunkSize, ChunkIndexMetadata, ChunkCount, DirtyChunkCount, DirtyDocsCount, Footer
- Header -->
- PackedIntsVersion, ChunkSize -->
- ChunkCount, DirtyChunkCount, DirtyDocsCount -->
- ChunkIndexMetadata -->
- Footer -->
- PackedIntsVersion is
- ChunkSize is the number of bytes of terms to accumulate before flushing.
- ChunkCount is not known in advance and is the number of chunks necessary to store all document of the segment.
- DirtyChunkCount is the number of prematurely flushed chunks in the .tvd file.
A vector data file (extension
.tvd). This file stores terms, frequencies, positions, offsets and payloads for every document. Upon writing a new segment, it accumulates data into memory until the buffer used to store terms and payloads grows beyond 4KB. Then it flushes all metadata, terms and positions to disk using LZ4 compression for terms and payloads and
blocks of packed intsfor positions.
Here is a more detailed description of the field data file format:
- VectorData (.tvd) --> <Header>, <Chunk>ChunkCount, Footer
- Header -->
- Chunk --> DocBase, ChunkDocs, < NumFields >, < FieldNums >, < FieldNumOffs >, < Flags >, < NumTerms >, < TermLengths >, < TermFreqs >, < Positions >, < StartOffsets >, < Lengths >, < PayloadLengths >, < TermAndPayloads >
- NumFields --> DocNumFieldsChunkDocs
- FieldNums --> FieldNumDeltaTotalDistincFields
- Flags --> Bit < FieldFlags >
- FieldFlags --> if Bit==1: FlagTotalDistinctFields else FlagTotalFields
- NumTerms --> FieldNumTermsTotalFields
- TermLengths --> PrefixLengthTotalTerms SuffixLengthTotalTerms
- TermFreqs --> TermFreqMinus1TotalTerms
- Positions --> PositionDeltaTotalPositions
- StartOffsets --> (AvgCharsPerTermTotalDistinctFields) StartOffsetDeltaTotalOffsets
- Lengths --> LengthMinusTermLengthTotalOffsets
- PayloadLengths --> PayloadLengthTotalPayloads
- TermAndPayloads --> LZ4-compressed representation of < FieldTermsAndPayLoads >TotalFields
- FieldTermsAndPayLoads --> Terms (Payloads)
- DocBase, ChunkDocs, DocNumFields (with ChunkDocs==1) -->
- AvgCharsPerTerm -->
- DocNumFields (with ChunkDocs>=1), FieldNumOffs -->
- FieldNumTerms, PrefixLength, SuffixLength, TermFreqMinus1, PositionDelta,
StartOffsetDelta, LengthMinusTermLength, PayloadLength -->
blocks of 64 packed ints
- Footer -->
- DocBase is the ID of the first doc of the chunk.
- ChunkDocs is the number of documents in the chunk.
- DocNumFields is the number of fields for each doc.
- FieldNums is a delta-encoded list of the sorted unique field numbers present in the chunk.
- FieldNumOffs is the array of FieldNumOff; array size is the total number of fields in the chunk.
- FieldNumOff is the offset of the field number in FieldNums.
- TotalFields is the total number of fields (sum of the values of NumFields).
- Bit in Flags is a single bit which when true means that fields have the same options for every document in the chunk.
- Flag: a 3-bits int where:
- the first bit means that the field has positions
- the second bit means that the field has offsets
- the third bit means that the field has payloads
- FieldNumTerms is the number of terms for each field.
- TotalTerms is the total number of terms (sum of NumTerms).
- PrefixLength is 0 for the first term of a field, the common prefix with the previous term otherwise.
- SuffixLength is the length of the term minus PrefixLength for every term using.
- TermFreqMinus1 is (frequency - 1) for each term.
- TotalPositions is the sum of frequencies of terms of all fields that have positions.
- PositionDelta is the absolute position for the first position of a term, and the difference with the previous positions for following positions.
- TotalOffsets is the sum of frequencies of terms of all fields that have offsets.
- AvgCharsPerTerm is the average number of chars per term, encoded as a float on 4 bytes. They are not present if no field has both positions and offsets enabled.
- StartOffsetDelta is the (startOffset - previousStartOffset - AvgCharsPerTerm * PositionDelta). previousStartOffset is 0 for the first offset and AvgCharsPerTerm is 0 if the field has no positions.
- LengthMinusTermLength is (endOffset - startOffset - termLength).
- TotalPayloads is the sum of frequencies of terms of all fields that have payloads.
- PayloadLength is the payload length encoded.
- Terms is term bytes.
- Payloads is payload bytes (if the field has payloads).
An index file (extension
- WARNING: This API is experimental and might change in incompatible ways in the next release.
Lucene90TermVectorsFormatpublic Lucene90TermVectorsFormat()Sole constructor.