SegmentReader

Namespace: Rowles.LeanLucene.Index.Segment

Assembly: Rowles.LeanLucene.dll

Reads a single immutable segment from disc via MMapDirectory.

public sealed class SegmentReader : IDisposable

SegmentReader

Constructors

SegmentReader(MMapDirectory, SegmentInfo): Initialises a new SegmentReader for the given segment. Opens all required files and validates the segment's on-disk data.

Properties

DocBase: Gets or sets the document base offset for this reader within the global document namespace.

HasDeletions: True when this segment has no deleted documents, allowing callers to skip per-doc IsLive checks.

HasTermVectors: Whether this segment has term vector files.

HasVectors: Returns whether this segment has vector data.

Info: Gets the segment metadata for this reader.

MaxDoc: Gets the total number of documents in this segment, including deleted documents.

TermOffsetCacheCount

TermOffsetCacheHits

VectorFieldNames: Returns the field names with vector data in this segment.

Methods

Dispose(): Performs application-defined tasks associated with freeing, releasing, or resetting unmanaged resources.

GetAllTermsForField(string): Returns all terms for a given field.

GetDocFreq(string): Returns the document frequency for a pre-built qualified term string.

GetDocFreq(string, string): Returns the document frequency for a term (count only, no full decode).

GetDocFreqByQualified(string): Returns the document frequency using a pre-built qualified term string.

GetDocIds(string): Returns document IDs for a pre-built qualified term string.

GetDocIds(string, string): Returns document IDs matching the given field and term.

GetFieldLength(int): Returns an approximate field length using the first available field (for non-field-specific queries).

GetFieldLength(int, string): Returns an approximate field length for BM25 for a specific field, derived from the stored norm.

GetFuzzyMatches(string, ReadOnlySpan<char>, int, int): Returns terms within Levenshtein distance of queryTerm, with edit distances.

GetHnswGraph(string): Returns the (lazy-loaded) HNSW graph for the given vector field, or null if no graph exists. Thread-safe; the first caller materialises the graph and subsequent callers reuse it.

GetNorm(int): Returns the quantised norm value for a document using the first available field.

GetNorm(int, string): Returns the quantised norm value for a document in a specific field (0..1 range).

GetNumericDocValues(string): Returns the NumericDocValues array for a field, or null if unavailable.

GetNumericRange(string, double, double): Returns all document IDs that have a numeric value in the given field within the specified range.

GetParentBitSet(): Returns the parent bitset for block-join indexing, or null if this segment has no block documents.

GetPositions(string, int): Returns positional data for a pre-built qualified term string.

GetPositions(string, string, int): Returns positional data for a term in a specific document, or null if unavailable.

GetPostingsEnum(string): Returns a PostingsEnum cursor for the given qualified term (field\0term). Decodes the postings list once; caller must dispose.

GetPostingsEnumAtOffset(long): Returns a PostingsEnum at a known postings offset, skipping the dictionary lookup. Use when the offset was already obtained from a term scan (e.g. prefix/wildcard).

GetPostingsEnumWithPositions(string): Returns a PostingsEnum with decoded positions for phrase queries.

GetSortedDocValues(string): Returns the SortedDocValues array for a field, or null if unavailable.

GetStoredFields(int): Returns all stored fields for the specified document as a read-only dictionary.

GetTermFrequency(string, int): Returns the term frequency for a pre-built qualified term string.

GetTermFrequency(string, string, int): Returns the term frequency for a given term in a specific document.

GetTermVectors(int): Returns term vectors for a document, or null if term vectors are not stored for this segment. Lazily opens the .tvd/.tvx files on first access.

GetTermsInRange(string, string?, string?, bool, bool): Returns terms in lexicographic range [lower, upper] for a field.

GetTermsMatching(string, ReadOnlySpan<char>): Returns all terms for a field matching a wildcard pattern.

GetTermsMatchingRegex(string, Regex): Returns terms for a field matching the compiled regex.

GetTermsWithPrefix(string): Returns all terms matching a qualified prefix.

GetVector(int): Reads the vector for a given document from the first available vector field (legacy convenience).

GetVector(string, int): Reads the vector for a given document on the named vector field.

IntersectAutomaton(string, IAutomaton): Intersects the term dictionary with an automaton, returning matching terms.

IsLive(int): Returns true if the document with the given ID has not been deleted.

ReadDocFreqAtOffset(long): Reads docFreq directly from a known postings file offset (no dictionary lookup).

TryGetFieldLengths(string, out int[]?): Retrieves the raw field-length array for a given field, allowing callers to resolve the array once and index by docId directly in tight loops.

TryGetNumericValue(string, int, out double): Tries to get a numeric field value for a document from the .num index.

TryGetSortedDocValue(string, int, out string): Tries to get a string DocValues field value for a document.

Edit this page