Documentation
¶
Overview ¶
Package memstore provides in-memory reference implementations of the persistence ports. It defines the reference semantics every other engine must match (via the internal/conformance suites) and is a perfectly serviceable backend for small corpora: brute-force cosine over float32 slices handles ~100k chunks in single-digit milliseconds.
Index ¶
- type AnswerCache
- type CollectionRepository
- func (r *CollectionRepository) Create(_ context.Context, c *domain.Collection) error
- func (r *CollectionRepository) Delete(_ context.Context, name string) error
- func (r *CollectionRepository) Get(_ context.Context, name string) (*domain.Collection, error)
- func (r *CollectionRepository) List(_ context.Context) ([]*domain.Collection, error)
- func (r *CollectionRepository) RecordSource(_ context.Context, name, source string) error
- type DocumentRepository
- func (r *DocumentRepository) Delete(_ context.Context, collection string, id domain.DocumentID) ([]domain.ChunkID, error)
- func (r *DocumentRepository) DeleteChunks(_ context.Context, collection string, ids []domain.ChunkID) ([]domain.ChunkID, error)
- func (r *DocumentRepository) DeleteCollection(_ context.Context, collection string) ([]domain.ChunkID, error)
- func (r *DocumentRepository) GetBySource(_ context.Context, collection, sourceURI string) (*domain.Document, error)
- func (r *DocumentRepository) GetChunks(_ context.Context, ids []domain.ChunkID) ([]domain.Chunk, error)
- func (r *DocumentRepository) GetChunksByDocument(_ context.Context, collection string, id domain.DocumentID) ([]domain.Chunk, error)
- func (r *DocumentRepository) GetChunksByIDs(_ context.Context, collection string, ids []string) ([]domain.Chunk, error)
- func (r *DocumentRepository) GetDocuments(_ context.Context, ids []domain.DocumentID) ([]*domain.Document, error)
- func (r *DocumentRepository) ListDocuments(_ context.Context, collection string) ([]*domain.Document, error)
- func (r *DocumentRepository) Upsert(_ context.Context, doc *domain.Document, chunks []domain.Chunk) error
- type LexicalIndex
- func (x *LexicalIndex) Delete(_ context.Context, collection string, ids []domain.ChunkID) error
- func (x *LexicalIndex) Search(_ context.Context, collection, query string, k int, filter domain.Predicate) ([]domain.ChunkID, error)
- func (x *LexicalIndex) Upsert(_ context.Context, collection string, docs []app.LexicalDoc) error
- type VectorIndex
- func (x *VectorIndex) Delete(_ context.Context, collection string, ids []domain.ChunkID) error
- func (x *VectorIndex) Entries(_ context.Context, collection string) ([]app.VectorEntry, error)
- func (x *VectorIndex) Search(_ context.Context, collection string, query []float32, k int, ...) ([]domain.VectorMatch, error)
- func (x *VectorIndex) Upsert(_ context.Context, collection string, entries []app.VectorEntry) error
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type AnswerCache ¶
type AnswerCache struct {
// contains filtered or unexported fields
}
AnswerCache is a thread-safe, in-memory app.AnswerCache (the reference impl, and what the "memory" storage backend uses). It is process-local, so it only helps within a single run.
func (*AnswerCache) Get ¶
func (c *AnswerCache) Get(_ context.Context, key string, notBefore time.Time) (app.Answer, bool, error)
Get returns the cached answer for key when present and stored at or after notBefore; otherwise ok is false.
type CollectionRepository ¶
type CollectionRepository struct {
// contains filtered or unexported fields
}
CollectionRepository is a thread-safe, in-memory store of Collection aggregates, keyed by name. It owns its data: it stores and returns copies so callers cannot alias the stored aggregates.
func NewCollectionRepository ¶
func NewCollectionRepository() *CollectionRepository
NewCollectionRepository returns an empty repository.
func (*CollectionRepository) Create ¶
func (r *CollectionRepository) Create(_ context.Context, c *domain.Collection) error
Create stores a copy of c, failing with ErrAlreadyExists if the name is taken.
func (*CollectionRepository) Delete ¶
func (r *CollectionRepository) Delete(_ context.Context, name string) error
Delete removes the named collection, or fails with ErrNotFound. The cross-aggregate cascade (its documents and vectors) is the use case's job.
func (*CollectionRepository) Get ¶
func (r *CollectionRepository) Get(_ context.Context, name string) (*domain.Collection, error)
Get returns a copy of the named collection, or ErrNotFound.
func (*CollectionRepository) List ¶
func (r *CollectionRepository) List(_ context.Context) ([]*domain.Collection, error)
List returns copies of every collection, in unspecified order.
func (*CollectionRepository) RecordSource ¶
func (r *CollectionRepository) RecordSource(_ context.Context, name, source string) error
RecordSource appends source to the collection's Sources, idempotently, or fails with ErrNotFound.
type DocumentRepository ¶
type DocumentRepository struct {
// contains filtered or unexported fields
}
DocumentRepository is a thread-safe, in-memory store of Documents and their Chunks. Chunk data lives in byChunkID (the source of truth for GetChunks); docChunks tracks each document's chunk IDs so Upsert can replace and Delete can cascade. Values are stored and returned by copy.
func NewDocumentRepository ¶
func NewDocumentRepository() *DocumentRepository
NewDocumentRepository returns an empty repository.
func (*DocumentRepository) Delete ¶
func (r *DocumentRepository) Delete(_ context.Context, collection string, id domain.DocumentID) ([]domain.ChunkID, error)
Delete removes the document and its chunks, returning the removed chunk IDs, or fails with ErrNotFound. Their vectors live in the VectorIndex and are removed by the use case.
func (*DocumentRepository) DeleteChunks ¶
func (r *DocumentRepository) DeleteChunks(_ context.Context, collection string, ids []domain.ChunkID) ([]domain.ChunkID, error)
DeleteChunks removes the given chunks that belong to the collection, returning the IDs actually removed. The owning documents are left in place (their record stands even after losing chunks). IDs absent from the collection are skipped.
func (*DocumentRepository) DeleteCollection ¶
func (r *DocumentRepository) DeleteCollection(_ context.Context, collection string) ([]domain.ChunkID, error)
DeleteCollection removes every document and its chunks in the collection, returning all removed chunk IDs. An empty collection is a no-op.
func (*DocumentRepository) GetBySource ¶
func (r *DocumentRepository) GetBySource(_ context.Context, collection, sourceURI string) (*domain.Document, error)
GetBySource returns a copy of the document for (collection, sourceURI), or ErrNotFound.
func (*DocumentRepository) GetChunks ¶
func (r *DocumentRepository) GetChunks(_ context.Context, ids []domain.ChunkID) ([]domain.Chunk, error)
GetChunks hydrates chunks by ID, preserving input order and skipping IDs with no stored chunk (the result may be shorter than the input).
func (*DocumentRepository) GetChunksByDocument ¶
func (r *DocumentRepository) GetChunksByDocument(_ context.Context, collection string, id domain.DocumentID) ([]domain.Chunk, error)
GetChunksByDocument returns all chunks of one document in seq order. An unknown document yields no chunks and no error.
func (*DocumentRepository) GetChunksByIDs ¶
func (r *DocumentRepository) GetChunksByIDs(_ context.Context, collection string, ids []string) ([]domain.Chunk, error)
GetChunksByIDs returns the chunks with the given IDs that belong to the collection, in input order. IDs absent from the collection (unknown, or owned by another collection) are omitted. An unknown collection yields no chunks.
func (*DocumentRepository) GetDocuments ¶
func (r *DocumentRepository) GetDocuments(_ context.Context, ids []domain.DocumentID) ([]*domain.Document, error)
GetDocuments hydrates documents by ID, preserving input order and skipping IDs with no stored document (the result may be shorter than the input).
func (*DocumentRepository) ListDocuments ¶
func (r *DocumentRepository) ListDocuments(_ context.Context, collection string) ([]*domain.Document, error)
ListDocuments returns every document in the collection (order unspecified). An unknown or empty collection yields no documents and no error.
type LexicalIndex ¶ added in v1.0.0
type LexicalIndex struct {
// contains filtered or unexported fields
}
LexicalIndex is a thread-safe in-memory BM25 keyword index — the reference semantics for the lexical half of hybrid retrieval. Corpus statistics (idf, average document length) are recomputed per search; for the small corpora memstore targets this is cheap and keeps the store free of incremental bookkeeping.
func NewLexicalIndex ¶ added in v1.0.0
func NewLexicalIndex() *LexicalIndex
NewLexicalIndex returns an empty index.
func (*LexicalIndex) Delete ¶ added in v1.0.0
Delete removes the given chunk IDs; absent IDs are a no-op.
func (*LexicalIndex) Search ¶ added in v1.0.0
func (x *LexicalIndex) Search(_ context.Context, collection, query string, k int, filter domain.Predicate) ([]domain.ChunkID, error)
Search ranks the metadata-matching chunks that contain at least one query term by BM25, best first, returning up to k chunk IDs.
func (*LexicalIndex) Upsert ¶ added in v1.0.0
func (x *LexicalIndex) Upsert(_ context.Context, collection string, docs []app.LexicalDoc) error
Upsert tokenizes and stores each document, replacing entries with the same ChunkID.
type VectorIndex ¶
type VectorIndex struct {
// contains filtered or unexported fields
}
VectorIndex is a thread-safe, brute-force cosine similarity index.
func (*VectorIndex) Entries ¶
func (x *VectorIndex) Entries(_ context.Context, collection string) ([]app.VectorEntry, error)
Entries returns a copy of every stored vector (and its metadata) for the collection (order unspecified). An unknown collection yields no entries and no error.
func (*VectorIndex) Search ¶
func (x *VectorIndex) Search(_ context.Context, collection string, query []float32, k int, filter domain.Predicate) ([]domain.VectorMatch, error)
Search returns up to k matches by cosine similarity, best first, considering only entries whose metadata satisfies filter (the zero predicate matches all).
func (*VectorIndex) Upsert ¶
func (x *VectorIndex) Upsert(_ context.Context, collection string, entries []app.VectorEntry) error
Upsert stores copies of the vectors and their metadata, replacing entries with the same ChunkID.
Source Files