Persister

automerge_persistent

Trait Persister 

Source 
pub trait Persister {
    type Error: Error + 'static;

    // Required methods
    fn get_changes(&self) -> Result<Vec<Vec<u8>>, Self::Error>;
    fn insert_changes(
        &mut self,
        changes: Vec<(ActorId, u64, Vec<u8>)>,
    ) -> Result<(), Self::Error>;
    fn remove_changes(
        &mut self,
        changes: Vec<(&ActorId, u64)>,
    ) -> Result<(), Self::Error>;
    fn get_document(&self) -> Result<Option<Vec<u8>>, Self::Error>;
    fn set_document(&mut self, data: Vec<u8>) -> Result<(), Self::Error>;
    fn get_sync_state(
        &self,
        peer_id: &[u8],
    ) -> Result<Option<Vec<u8>>, Self::Error>;
    fn set_sync_state(
        &mut self,
        peer_id: Vec<u8>,
        sync_state: Vec<u8>,
    ) -> Result<(), Self::Error>;
    fn remove_sync_states(
        &mut self,
        peer_ids: &[&[u8]],
    ) -> Result<(), Self::Error>;
    fn get_peer_ids(&self) -> Result<Vec<Vec<u8>>, Self::Error>;
    fn sizes(&self) -> StoredSizes;
    fn flush(&mut self) -> Result<usize, Self::Error>;
}
Expand description

A Persister persists both changes and documents to durable storage.

In the event of a power loss changes should still be around for loading after. It is up to the implementation to decide on trade-offs regarding how often to fsync for example.

Changes are identified by a pair of actor_id and sequence_number. This uniquely identifies a change and so is suitable for use as a key in the implementation.

Documents are saved automerge Backends so are more compact than the raw changes they represent.

Required Associated Types§

Source

type Error: Error + 'static

The error type that the operations can produce

Required Methods§

Source

fn get_changes(&self) -> Result<Vec<Vec<u8>>, Self::Error>

Returns all of the changes that have been persisted through this persister. Ordering is not specified as the automerge Backend should handle that.

Source

fn insert_changes( &mut self, changes: Vec<(ActorId, u64, Vec<u8>)>, ) -> Result<(), Self::Error>

Inserts the given change at the unique address specified by the actor_id and sequence_number.

Source

fn remove_changes( &mut self, changes: Vec<(&ActorId, u64)>, ) -> Result<(), Self::Error>

Removes the change at the unique address specified by the actor_id and sequence_number.

If the change does not exist this should not return an error.

Source

fn get_document(&self) -> Result<Option<Vec<u8>>, Self::Error>

Returns the document, if one has been persisted previously.

Source

fn set_document(&mut self, data: Vec<u8>) -> Result<(), Self::Error>

Sets the document to the given data.

Source

fn get_sync_state(&self, peer_id: &[u8]) -> Result<Option<Vec<u8>>, Self::Error>

Returns the sync state for the given peer if one exists.

A peer id corresponds to an instance of a backend and may be serving multiple frontends so we cannot have it work on ActorIds.

Source

fn set_sync_state( &mut self, peer_id: Vec<u8>, sync_state: Vec<u8>, ) -> Result<(), Self::Error>

Sets the sync state for the given peer.

A peer id corresponds to an instance of a backend and may be serving multiple frontends so we cannot have it work on ActorIds.

Source

fn remove_sync_states(&mut self, peer_ids: &[&[u8]]) -> Result<(), Self::Error>

Removes the sync states associated with the given peer_ids.

Source

fn get_peer_ids(&self) -> Result<Vec<Vec<u8>>, Self::Error>

Returns the list of peer ids with stored SyncStates.

This is intended for use by users to see what peer_ids are taking space so that they can be removed during a compaction.

Source

fn sizes(&self) -> StoredSizes

Returns the sizes components being stored consume.

This can be used as an indicator of when to compact the storage.

Source

fn flush(&mut self) -> Result<usize, Self::Error>

Flush the data out to disk.

Implementors§