Struct VertexAiRagConfig 

pub struct VertexAiRagConfig {
    pub project: String,
    pub location: String,
    pub rag_corpora: Vec<String>,
    pub similarity_top_k: Option<u32>,
    pub vector_distance_threshold: Option<f64>,
}

Configuration for Vertex AI RAG retrieval.

Mirrors ADK-Python’s VertexRagStore retrieval parameters. A retrieval may target one or more rag corpora (resource names) and constrain the result set via similarity_top_k and vector_distance_threshold.

Fields

project: String

Google Cloud project ID.

location: String

Google Cloud location (e.g. us-central1).

rag_corpora: Vec<String>

The RAG corpus resource name(s). Format: projects/{project}/locations/{location}/ragCorpora/{corpus_id}

similarity_top_k: Option<u32>

Number of contexts to retrieve (maps to similarity_top_k).

vector_distance_threshold: Option<f64>

Only return contexts with a vector distance smaller than this threshold.

Implementations

impl VertexAiRagConfig

pub fn from_corpus(corpus: impl Into<String>) -> Self

Create a new config from a single corpus resource name.

The project and location are parsed from the corpus resource name when it is fully-qualified (projects/{project}/locations/{location}/...).

pub fn with_project_location( self, project: impl Into<String>, location: impl Into<String>, ) -> Self

Set the explicit project / location used to build the endpoint URL.

pub fn with_similarity_top_k(self, top_k: u32) -> Self

Set the number of contexts to retrieve.

pub fn with_vector_distance_threshold(self, threshold: f64) -> Self

Set the vector distance threshold.

Trait Implementations

impl Clone for VertexAiRagConfig

fn clone(&self) -> VertexAiRagConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for VertexAiRagConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.