indexer

package

v0.11.0 Latest Latest Go to latest Published: Sep 3, 2026 License: MIT Imports: 19 Imported by: 0

Details

Rendered for

This section is empty.

This section is empty.

This section is empty.

type Embedder struct {
	// contains filtered or unexported fields
}

Embedder generates and atomically stores embeddings for chunks whose vector/metadata pair is missing, stale, or invalid.

func NewEmbedder(database *db.DB, provider providers.EmbeddingProvider, providerTag, fingerprint string) *Embedder

func (e *Embedder) EmbedCollection(ctx context.Context, collection string) error

EmbedCollection repairs every chunk that does not have a valid vector and matching current metadata.

type Indexer struct {
	// contains filtered or unexported fields
}

Indexer walks a collection and upserts documents into the DB.

func New(database *db.DB, chunkSize int) *Indexer

func (idx *Indexer) Index(ctx context.Context, col config.Collection) (Stats, error)

Index indexes all files in a collection.

type Stats struct {
	FilesScanned int
	FilesAdded   int
	FilesUpdated int
	FilesRemoved int
	FilesSkipped int
	Duration     time.Duration
}

Stats summarises an index run.