document.Document

A document containing text content to be chunked and indexed.

Usage

Source

document.Document(
    content,
    origin=None,
    attributes=None,
)

Documents are the primary input for RAG stores. Each document holds the full text to be indexed plus an optional origin identifying where it came from. A document is chunked into a ChunkedDocument before being embedded and written to a store. Document is the base type; MarkdownDocument is the common concrete variant produced by read_as_markdown().

Parameters

content: str

The full text content of the document.

origin: Optional[str] = None

Unique origin identifier for the document. This can be None or an empty string while preparing a document object, but stores require a populated origin for upsert operations.

attributes: Optional[dict[str, Any]] = None
Optional user-defined attributes applied at document insertion time. Document-level attributes can be inherited by chunks and returned during retrieval for filtering and downstream prompt/context use.

Methods

Name Description
from_any() Convert any document-like or IntoDocument object to a raghilda Document.
to_chunked() Return a ChunkedDocument with the same fields and supplied chunks.

from_any()

Convert any document-like or IntoDocument object to a raghilda Document.

Usage

Source

from_any(doc)

This conversion only accepts unchunked inputs. If the source object already carries chunks, use ChunkedDocument.from_any() instead.

Parameters
doc: Union[DocumentLike, IntoDocument]
An object that implements the DocumentLike protocol or has a to_document() method.
Returns
Document
A raghilda Document instance.

to_chunked()

Return a ChunkedDocument with the same fields and supplied chunks.

Usage

Source

to_chunked(chunks)