You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In this epic, we will make minimal changes in the Rust node to adapt to the Output Unification. In particular, we should avoid making significant changes to the Postgres Database schema used by the Rust Rollups Node. For this issue, we also won't change the GraphQL API.
The Cartesi-machine Linux driver will handle outputs as binary data. So, the rollups-http-server will encode these outputs before sending them to the driver. The Emulator Unit will do most of these changes, but they might need some help from us to do modifications on the Rust part of the Rollups HTTP server. The outputs will be encoded using the Solidity ABI encoding using the following contract interface.
Outside the machine, the server manager will receive the outputs as binary data and decode them back to structured data. The server-manager API might change slightly, so we must modify the advance-runner and the host-runner.
I believe it won't be necessary for the host-runner to encode and decode the outputs to binary. We can store the structured outputs directly in the database.
📚 Context
In this epic, we will make minimal changes in the Rust node to adapt to the Output Unification. In particular, we should avoid making significant changes to the Postgres Database schema used by the Rust Rollups Node. For this issue, we also won't change the GraphQL API.
The Cartesi-machine Linux driver will handle outputs as binary data. So, the
rollups-http-server
will encode these outputs before sending them to the driver. The Emulator Unit will do most of these changes, but they might need some help from us to do modifications on the Rust part of the Rollups HTTP server. The outputs will be encoded using the Solidity ABI encoding using the following contract interface.Outside the machine, the server manager will receive the outputs as binary data and decode them back to structured data. The server-manager API might change slightly, so we must modify the advance-runner and the host-runner.
I believe it won't be necessary for the host-runner to encode and decode the outputs to binary. We can store the structured outputs directly in the database.
📈 Subtasks
The text was updated successfully, but these errors were encountered: