Sample Hydra Project

It is a sample Hydra v3 project for the Kusama network. It showcase the two main features of Hydra v3:

Dealing with runtime upgrades via spec filters
Using ad-hoc data loaders via block hooks

1. Bootstrap

Run

yarn && yarn bootstrap

It will generate the model files as defined in schema.graphql, create the database and run all the necessary migrations in one shot.

NB! Use with caution in production, as it will delete all the existing records.

2. Generate Types for events and extrinsics

A separate tool Hydra Typegen can be used for generating Typescript classes for the event handlers (the mappings).
Run

yarn typegen

to generated types for the event balances.Transfer. Under the hood, it will run hydra-typegen twice against different blocks and produce typed classes in mappings/generated/types-V1 and mappings/generated/types-V2 respectively.

3. Build Mappings

Mappings is a separated TypeScript module created in the mappings folder. The handlers exported by the module should match the ones defined in manifest.yml in the mappings section. Once the necessary files are generated, build it with

yarn mappings:build

Note that we have to versions of the balancesTransfer mapping: balancesTransferV1 and balancesTransferV2. Depending on the specVersion filter the processor will pick up the right one. The former is going to be applied for specVersion up to 1045 and the latter for specVersion from 1045 onwards.

Another mapping, genesisLoader is a pre-block hook, which means it runs before any other mapping in the block. By setting it's filter range to [0, 0] it is restricted only to the genesis block and is convenient place to seed some data if needed.

4. Run local Hydra Indexer

A stack of Hydra Indexer services is configured in docker-compose-indexer.yml. Run

$ docker-compose -f docker-compose-indexer.yml up -d

and check the status at localhost:4001 by quering

query {
  indexerStatus {
    chainHeight
    head
    hydraVersion
  }
}

Make sure that the last indexer block (head) and the last known finalized block (chainHeight) are both positive and hydraVersion is 3.x.

5. Run the processor and the GraphQL server

Then run the processor:

yarn processor:start

Afterwards, start the GraphQL server in a separate terminal (opens a GraphQL playground at localhost by default):

yarn query-node:start:dev

Running in Docker

Docker files are located in ./docker. First, build the builder image:

$ docker build . -f docker/Dockerfile.builder -t builder

Images for the GraphQL query node and the processor depend on the builder image. Build with

$ docker build . -f docker/Dockerfile.query-node -t query-node:latest
$ docker build . -f docker/Dockerfile.processor -t processor:latest

In order to run the docker-compose stack, we need to create the schema and run the database migrations.

$ docker-compose up -d db 
$ yarn docker:db:migrate

The last command runs yarn db:bootstrap in the builder image. A similar setup strategy may be used for Kubernetes (with builder as a starter container).

Now everything is ready:

$ docker-compose up