REST Resource: projects.locations.dataStores.branches.documents.chunks

Resource: Chunk

Chunk captures all raw metadata information of items to be recommended or searched in the chunk mode.

JSON representation
{
  "name": string,
  "id": string,
  "content": string,
  "documentMetadata": {
    object (DocumentMetadata)
  },
  "derivedStructData": {
    object
  },
  "pageSpan": {
    object (PageSpan)
  },
  "chunkMetadata": {
    object (ChunkMetadata)
  },
  "dataUrls": [
    string
  ],
  "annotationContents": [
    string
  ],
  "annotationMetadata": [
    {
      object (AnnotationMetadata)
    }
  ],
  "relevanceScore": number
}
Fields
name

string

The full resource name of the chunk. Format: projects/{project}/locations/{location}/collections/{collection}/dataStores/{dataStore}/branches/{branch}/documents/{documentId}/chunks/{chunk_id}.

This field must be a UTF-8 encoded string with a length limit of 1024 characters.

id

string

Unique chunk ID of the current chunk.

content

string

Content is a string from a document (parsed content).

documentMetadata

object (DocumentMetadata)

Metadata of the document from the current chunk.

derivedStructData

object (Struct format)

Output only. This field is OUTPUT_ONLY. It contains derived data that are not in the original input document.

pageSpan

object (PageSpan)

Page span of the chunk.

chunkMetadata

object (ChunkMetadata)

Output only. Metadata of the current chunk.

dataUrls[]

string

Output only. Image Data URLs if the current chunk contains images. Data URLs are composed of four parts: a prefix (data:), a MIME type indicating the type of data, an optional base64 token if non-textual, and the data itself: data:[][;base64],

annotationContents[]

string

Output only. Annotation contents if the current chunk contains annotations.

annotationMetadata[]

object (AnnotationMetadata)

Output only. The annotation metadata includes structured content in the current chunk.

relevanceScore

number

Output only. Represents the relevance score based on similarity. Higher score indicates higher chunk relevance. The score is in range [-1.0, 1.0]. Only populated on SearchResponse.

Methods

get

Gets a Document.

list

Gets a list of Chunks.