chunker

type CharacterSplitter struct {
	ChunkSize    int
	ChunkOverlap int
	Separators   []string
}

type TextSplitter interface {
	// SplitText turns a raw string into meaningful chunk strings
	SplitText(text string) ([]string, error)

	// SplitDocument extends raw string logic with ID mapping and metadata (Chunk = Node in LlamaIndex)
	SplitDocument(ctx context.Context, doc *core.Document) ([]*core.Chunk, error)
}