cinderblock-core 0.7.0

Resource trait definitions, CRUD operations, runtime context, and in-memory data layer for cinderblock
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

use crate::Resource;

pub mod in_memory;

/// Persistence backend for a specific resource type.
///
/// The trait is parameterized on `R` so that different data layers can
/// impose different bounds on the resources they support. For example,
/// `InMemoryDataLayer` has a blanket impl for all `R: Resource`, while
/// a SQLite data layer can additionally require `R: SqlResource`.
pub trait DataLayer<R: Resource>: std::fmt::Debug + Send + Sync + 'static {
    /// Persist a new resource and return the stored version.
    ///
    /// The returned resource may differ from the input when the data layer
    /// provides server-generated values (e.g. autoincrement primary keys).
    fn create(&self, resource: R) -> impl Future<Output = Result<R, crate::CreateError>> + Send;

    fn read(
        &self,
        primary_key: &R::PrimaryKey,
    ) -> impl Future<Output = Result<R, crate::ReadError>> + Send;

    fn update(&self, resource: R) -> impl Future<Output = Result<(), crate::UpdateError>> + Send;

    /// Remove a resource by primary key, returning the deleted resource.
    fn destroy(
        &self,
        primary_key: &R::PrimaryKey,
    ) -> impl Future<Output = Result<R, crate::DestroyError>> + Send;
}