Development Container Reference

Substreams Dev Container Reference

The Substreams Dev Container is a tool to help you build your first project. You can either run it remotely through Github codespaces or locally by cloning the substreams-starter repository. Inside the Dev Container, the substreams init command sets up a code-generated Substreams project, allowing you to easily build a subgraph or an SQL-based solution for data handling.

Prerequisites

• Ensure Docker and VS Code are up-to-date.

First Navigating the Dev Container

Upon entering the Dev Container, you can either build or import your own substreams.yaml and associate modules within the minimal path, or opt for the automatically generated Substreams paths. Then running Substreams Build generates the Protobuf files.

• Minimal: Starts you with the raw block .proto and requires development. This path is intended for experienced users.

• Non-Minimal: Extracts filtered data using network-specific caches and Protobufs taken from corresponding foundational modules. This path generates a working Substreams out of the box.

To publish your work with the broader community, publish your .spkg to Substreams registry using:

• substreams registry login

• substreams registry publish

Tip: If you run into any problems within the Dev Container, use the help command to access trouble shooting tools.

Building a Sink for Your Project

You can configure your Substreams project to query data either through a Subgraph or directly from an SQL database:

• Subgraph: Run substreams codegen subgraph. This generates a project with a basic schema.graphql and mappings.ts file. You can customize these to define entities based on the data extracted by Substreams. For more information on configuring a Subgraph sink, see the Subgraph documentation.

• SQL: Run substreams codegen sql for SQL-based queries. For more information on configuring a SQL sink, refer to the SQL documentation.

Deployment Options

To deploy a Subgraph, you can either run the graph-node locally using the deploy-local command or deploy to Subgraph Studio by using the deploy command found in the package.json file.

Common Errors

• When running locally, make sure to verify that all Docker containers are healthy by running the dev-status command.

• If you put the wrong start-block while generating your project, navigate to the substreams.yaml to change the block number, then re-run substreams build.