Struct axum_extra::extract::Cached

pub struct Cached<T>(pub T);

Cache results of other extractors.

Cached wraps another extractor and caches its result in request extensions.

This is useful if you have a tree of extractors that share common sub-extractors that you only want to run once, perhaps because they’re expensive.

The cache purely type based so you can only cache one value of each type. The cache is also local to the current request and not reused across requests.

Example

use axum_extra::extract::Cached;
use axum::{
    async_trait,
    extract::FromRequestParts,
    response::{IntoResponse, Response},
    http::{StatusCode, request::Parts},
};

#[derive(Clone)]
struct Session { /* ... */ }

#[async_trait]
impl<S> FromRequestParts<S> for Session
where
    S: Send + Sync,
{
    type Rejection = (StatusCode, String);

    async fn from_request_parts(parts: &mut Parts, state: &S) -> Result<Self, Self::Rejection> {
        // load session...
    }
}

struct CurrentUser { /* ... */ }

#[async_trait]
impl<S> FromRequestParts<S> for CurrentUser
where
    S: Send + Sync,
{
    type Rejection = Response;

    async fn from_request_parts(parts: &mut Parts, state: &S) -> Result<Self, Self::Rejection> {
        // loading a `CurrentUser` requires first loading the `Session`
        //
        // by using `Cached<Session>` we avoid extracting the session more than
        // once, in case other extractors for the same request also loads the session
        let session: Session = Cached::<Session>::from_request_parts(parts, state)
            .await
            .map_err(|err| err.into_response())?
            .0;

        // load user from session...
    }
}

// handler that extracts the current user and the session
//
// the session will only be loaded once, even though `CurrentUser`
// also loads it
async fn handler(
    current_user: CurrentUser,
    // we have to use `Cached<Session>` here otherwise the
    // cached session would not be used
    Cached(session): Cached<Session>,
) {
    // ...
}

Tuple Fields

0: T

Trait Implementations

impl<T: Clone> Clone for Cached<T>

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

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<T: Debug> Debug for Cached<T>

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

Formats the value using the given formatter.

impl<T: Default> Default for Cached<T>

fn default() -> Cached<T>

Returns the “default value” for a type.

impl<T> Deref for Cached<T>

type Target = T

The resulting type after dereferencing.

fn deref(&self) -> &Self::Target

Dereferences the value.

impl<T> DerefMut for Cached<T>

fn deref_mut(&mut self) -> &mut Self::Target

Mutably dereferences the value.

impl<S, T> FromRequestParts<S> for Cached<T>

where S: Send + Sync, T: FromRequestParts<S> + Clone + Send + Sync + 'static,

type Rejection = <T as FromRequestParts<S>>::Rejection

If the extractor fails it’ll use this “rejection” type. A rejection is a kind of error that can be converted into a response.

fn from_request_parts<'life0, 'life1, 'async_trait>( parts: &'life0 mut Parts, state: &'life1 S, ) -> Pin<Box<dyn Future<Output = Result<Self, Self::Rejection>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Perform the extraction.