Struct LlmResponse 

pub struct LlmResponse<T> { /* private fields */ }

LLM-optimized response wrapper with token counting.

This wrapper automatically:

1. Serializes to the requested format (JSON or TOON)
2. Calculates estimated token counts for both formats
3. Adds informative headers about token usage

Token Estimation

Token counts are estimated using a simple heuristic:

• ~4 characters per token (GPT-3/4 average)

For more accurate counts, use a proper tokenizer.

Example

use rustapi_rs::prelude::*;
use rustapi_rs::toon::{LlmResponse, AcceptHeader, OutputFormat};

#[derive(Serialize)]
struct ApiData {
    items: Vec<Item>,
}

// With content negotiation
async fn get_items(accept: AcceptHeader) -> LlmResponse<ApiData> {
    let data = ApiData { items: vec![...] };
    LlmResponse::new(data, accept.preferred)
}

// Always TOON format
async fn get_items_toon() -> LlmResponse<ApiData> {
    let data = ApiData { items: vec![...] };
    LlmResponse::toon(data)
}

Implementations

impl<T> LlmResponse<T>

pub fn new(data: T, format: OutputFormat) -> Self

Create a new LLM response with the specified format.

pub fn json(data: T) -> Self

Create a JSON-formatted LLM response.

pub fn toon(data: T) -> Self

Create a TOON-formatted LLM response.

pub fn without_token_headers(self) -> Self

Disable token counting headers.

pub fn with_token_headers(self) -> Self

Enable token counting headers (default).

Trait Implementations

impl<T: Clone> Clone for LlmResponse<T>

fn clone(&self) -> LlmResponse<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug> Debug for LlmResponse<T>

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

Formats the value using the given formatter.

impl<T: Serialize> IntoResponse for LlmResponse<T>

fn into_response(self) -> Response

Convert self into a Response

impl<T: Send> OperationModifier for LlmResponse<T>

fn update_operation(_op: &mut Operation)

Update the operation

impl<T: Serialize> ResponseModifier for LlmResponse<T>

fn update_response(op: &mut Operation)

Update the operation with response information