Trait tvix_castore::proto::blob_service_server::BlobService

pub trait BlobService:
    Send
    + Sync
    + 'static {
    type ReadStream: Stream<Item = Result<BlobChunk, Status>> + Send + 'static;

    // Required methods
    fn stat<'life0, 'async_trait>(
        &'life0 self,
        request: Request<StatBlobRequest>,
    ) -> Pin<Box<dyn Future<Output = Result<Response<StatBlobResponse>, Status>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait;
    fn read<'life0, 'async_trait>(
        &'life0 self,
        request: Request<ReadBlobRequest>,
    ) -> Pin<Box<dyn Future<Output = Result<Response<Self::ReadStream>, Status>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait;
    fn put<'life0, 'async_trait>(
        &'life0 self,
        request: Request<Streaming<BlobChunk>>,
    ) -> Pin<Box<dyn Future<Output = Result<Response<PutBlobResponse>, Status>> + Send + 'async_trait>>
       where Self: 'async_trait,
             'life0: 'async_trait;
}

Generated trait containing gRPC methods that should be implemented for use with BlobServiceServer.

Required Associated Types

type ReadStream: Stream<Item = Result<BlobChunk, Status>> + Send + 'static

Server streaming response type for the Read method.

Required Methods

fn stat<'life0, 'async_trait>( &'life0 self, request: Request<StatBlobRequest>, ) -> Pin<Box<dyn Future<Output = Result<Response<StatBlobResponse>, Status>> + Send + 'async_trait>>

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

Stat can be used to check for the existence of a blob, as well as gathering more data about it, like more granular chunking information or baos. Server implementations are not required to provide more granular chunking information, especially if the digest specified in StatBlobRequest is already a chunk of a blob.

fn read<'life0, 'async_trait>( &'life0 self, request: Request<ReadBlobRequest>, ) -> Pin<Box<dyn Future<Output = Result<Response<Self::ReadStream>, Status>> + Send + 'async_trait>>

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

Read allows reading (all) data of a blob/chunk by the BLAKE3 digest of its contents. If the backend communicated more granular chunks in the Stat request, this can also be used to read chunks. This request returns a stream of BlobChunk, which is just a container for a stream of bytes. The server may decide on whatever chunking it may seem fit as a size for the individual BlobChunk sent in the response stream, this is mostly to keep individual messages at a manageable size.

fn put<'life0, 'async_trait>( &'life0 self, request: Request<Streaming<BlobChunk>>, ) -> Pin<Box<dyn Future<Output = Result<Response<PutBlobResponse>, Status>> + Send + 'async_trait>>

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

Put uploads a Blob, by reading a stream of bytes.

The way the data is chunked up in individual BlobChunk messages sent in the stream has no effect on how the server ends up chunking blobs up, if it does at all.

Implementors

impl<T> BlobService for GRPCBlobServiceWrapper<T>

where T: Deref<Target = dyn BlobService> + Send + Sync + 'static,

type ReadStream = Pin<Box<dyn Stream<Item = Result<BlobChunk, Status>> + Send>>