galacticcouncil/Basilisk-api-old

Implementation of the subsquid processor / indexer for Basilisk

Basilisk API

This project serves as the API for Basilisk's UIs, Dashboards and analytics.

Prerequisites

• Node v16.x
• Docker
• Built Basilisk node in ./../Basilisk-node
• Built Polkadot node in ./../polkadot

⚠️ Make sure to have both the Basilisk-node and polkadot accessible and built using cargo build --release in the aforementioned directories. Otherwise the testnet scripts available in package.json won't work.

Quick start

# Start the local basilisk testnet
yarn run testnet:start

# Start the local indexer targeting the basilisk node from the previous step
# If you need to debug the indexer, omit the `-d` option
yarn run indexer:start -- -d

# Prepare the dev environment for the processor
yarn run processor:clean-and-setup

# Start processing the indexed data
yarn run processor:start

# Start graphql server for processed data
yarn run processor:query-node:start

Generate custom types

1. define custom types in ./chain/definitions.ts
2. run processor:gen-chain-types

Configuration

Project's configuration is driven by environment variables, defined in .env, and manifest.yml. For more details see https://docs.subsquid.io.

Local testnet

Run yarn run testnet:start and visit the Polkadot apps UI to observe new blocks being produced after a while.

Development indexer

Run yarn run indexer:start to start a local indexer, which will index the locally running Basilisk node. Open Graphiql Online and use http://localhost:4010/v1/graphql as the GraphQL endpoint to interact with the indexer.

You can run the following query to get started:

query MyQuery {
  indexerStatus {
    inSync
    lastComplete
    chainHeight
  }
}

Apple M1 support

For Apple M1 support you need to build the indexer docker image locally:

git clone https://github.com/subsquid/hydra
cd hydra
git checkout v5
./scripts/docker-build.sh --target indexer-gateway -t subsquid/hydra-indexer-gateway:5
./scripts/docker-build.sh --target indexer -t subsquid/hydra-indexer:5
./scripts/docker-build.sh --target indexer-status-service -t subsquid/hydra-indexer-status-service:5

Development processor

Run yarn run processor:clean-and-setup to setup the development environment. Then run yarn run processor:start to start the processor.

Entity over time resolver

Resolvers have to be placed in the same module called server-extension ( src/server-extension.ts )

A generic factory is provided in order to simplify creation of 'Entity Over Time' resolvers. ( src/resolvers/factory.ts )

Model used in this resolver has to have at least one timestamp field.

Creating new resolver is simple as creating new class in the server-extension module:

@Resolver()
export class TestBlockOverTimeResolver extends entityOverTimeResolverFactory<TestBlock>(
    EntityType,
    EntityModel,
    'resolverName',
    'db_name',
    'time_fieldname'
) {
}

Parameters:

• EntityType is ObjectType class which will be returned from the resolver,eg:

@ObjectType()
export class EntityType{

    @Field({nullable: false})
    block_height!: bigint

    @Field({nullable: false})
    created_at!: Date

    constructor(props: any) {
        Object.assign(this, props);
    }
}

• EntityModel is Entity type class representing model in the DB. This should be auto-generated from the given graphql schema. Eg:

@Entity_()
export class EntityModel{
  constructor(props?: Partial<TestBlock>) {
    Object.assign(this, props)
  }

  @PrimaryColumn_()
  id!: string

  @Column_("numeric", {transformer: marshal.bigintTransformer, nullable: false})
  blockHeight!: bigint

  @Column_("timestamp with time zone", {nullable: false})
  createdAt!: Date
}

• db_name is exact table name in the database.
• time_fieldname is exact field name of timestamp field in the table.

Testing

Make sure to have Basilisk-node with XYK pools enabled. Use make build to build releases/testing-basilisk binary and launch in rococo-local folder polkadot-launch testing-config.json.

Do not run yarn run processor:query-node:start, because we need to keep port 4000 free.

# migrate test data
yarn run migrate:lbp
yarn run migrate:xyk

# start indexer and processor
yarn run processor:clean-and-setup
yarn run indexer:start
yarn run processor:start

# run all tests
yarn run test

# alternative
yarn run test:unit
yarn run test:e2e