Skip to content

Latest commit

 

History

History
 
 

08-hardhat-graph

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
abis
abis
 
 
bin
bin
 
 
contracts
contracts
 
 
scripts
scripts
 
 
test
test
 
 
.env.example
.env.example
 
 
README-CN.md
README-CN.md
 
 
README.md
README.md
 
 
graphql.md
graphql.md
 
 
hardhat.config.js
hardhat.config.js
 
 
package.json
package.json
 
 

README.md

中文 / English

Introduction of Graph

When coding smart contract, the change of status is usually expressed by triggering a event. The Graph is a GraphQL interface for capturing the chain events and provide a query event, it can track changes in data. Actually, many kinds of defi is based on The Graph to query data.

Overview of steps

  • deploy the contract on Goerli and call trigger event.
  • create Subgraph that defines the data index.
  • deploy Subgraph to TheGraph for data indexing
  • query and index data in DApp

If you has a private chain, you can clone Graph node code https://github.com/graphprotocol/graph-node/ and finish data indexing by maintain your own Graph node.

TheGraph defines that how to create data index, which is called Subgraph, including three components:

  • MainFest (subgraph.yaml) - define config
  • Schema mode(schema.graphq1) - define data, reference https://graphql.cn/learn/
  • Mapping (mapping.ts) - define the convertion between event and data

Steps

  1. Install related dependencies

    yarn install

  2. Configure the private key

    You should add private key in .env that formatted with "PRIVATE_KEY=xxxx", which read from the code. You also need to set your infura node id, add private key in .env that formatted with "INFURA_ID=xxxx"

  3. Deploy contracts(testing graph's simple contracts)

    npx hardhat run ./scripts/deploy.js --network goerli

    print info as follows:

    Deploying contracts with the account: xxxxxxxxxxxxxx
Account balance: 10000000000000000000000
Token address: 0x5FbDB2315678afecb367f032d93F642f64180aa3
Transfer 50 to receiver  0xf39Fd6e51aad88F6F4ce6aB8827279cffFb92266
Account balance of receiver is:  50

  4. TheGraph create a space of Subgraph

    We have to index data by the node of TheGraph, so we need to create a Subgraph on TheGraph Studio. If lacking of a account of TheGraph, you can connect to wallet to register. The account name is address of wallet, the following is called <THEGRAPH_USERNAME>.

    After approving the signature of wallet, it will jump to the "My Subgraphs" panel and click Create a Subgraph button.

    Input your project name(e.x. TEST01), it will be called <SUBGRAPH_NAME> as follows. Clicking continue button, then it will jump to subgraph which is the main page of project.

    Note: latest Graph CLI only support to deploy mainnet and goerli. If you want to use it on the others networks, you need to login by Github account, then create and deploy it on Hosted Service.

  5. Develop and deploy subgraph

    use yarn to install Graph CLI globally

    yarn global add @graphprotocol/graph-cli

  6. Initialize configuration:

    graph init --studio <SUBGRAPH_NAME>

    If using Hosted Service, the initializing command is as follows:

    graph init --product hosted-service <GITHUB_USER>/<SUBGRAPH NAME>
    • choose "Subgraph name" and "Direction to create the subgraph", then enter it.
    • choose goerli in Ethereum network
    • input your contract address generated when the contract was deployed in Step 3 in "Contract address"
    • When it came to "fetch ABI from Etherscan", it will fail, then show "ABI file (path)" which prompt that input the path of abi, we should input the path of SimpleToken.json(./abis/SimpleToken.json)
    • If yarn install fails (network error), you can enter in the new directory and install npm dependencies manually.

  7. Modify the mode of defining

    • modifying template in ./scripts/schema.graphql and ./scripts/mapping.ts

    • <SUBGRAPH_NAME>/schema.graphql the modified content as follows:

      type TransferEntity @entity {
  id: ID!
  from: Bytes! # address
  to: Bytes! # address
  value: BigInt!
}

type ApprovalEntity @entity {
  id: ID!
  owner: Bytes! # address
  spender: Bytes! # address
  value: BigInt!
}

    • <SUBGRAPH_NAME>/src/mapping.ts the modified content as follows:

      import { BigInt } from '@graphprotocol/graph-ts';
import { SimpleToken, Transfer, Approval } from '../generated/SimpleToken/SimpleToken';
import { TransferEntity, ApprovalEntity } from '../generated/schema';

export function handleTransfer(event: Transfer): void {
  // Entities can be loaded from the store using a string ID; this ID
  // needs to be unique across all entities of the same type
  let entity = TransferEntity.load(event.transaction.from.toHex());

  // Entities only exist after they have been saved to the store;
  // `null` checks allow to create entities on demand
  if (entity == null) {
    entity = new TransferEntity(event.transaction.from.toHex());
  }

  // BigInt and BigDecimal math are supported
  entity.value = event.params.value;

  // Entity fields can be set based on event parameters
  entity.from = event.params.from;
  entity.to = event.params.to;

  // Entities can be written to the store with `.save()`
  entity.save();

  // Note: If a handler doesn't require existing field values, it is faster
  // _not_ to load the entity from the store. Instead, create it fresh with
  // `new Entity(...)`, set the fields that should be updated and save the
  // entity back to the store. Fields that were not set or unset remain
  // unchanged, allowing for partial updates to be applied.

  // It is also possible to access smart contracts from mappings. For
  // example, the contract that has emitted the event can be connected to
  // with:
  //
  // let contract = Contract.bind(event.address)
  //
  // The following functions can then be called on this contract to access
  // state variables and other data:
  //
  // - contract.approve(...)
  // - contract.totalSupply(...)
  // - contract.transferFrom(...)
  // - contract.increaseAllowance(...)
  // - contract.balanceOf(...)
  // - contract.decreaseAllowance(...)
  // - contract.transfer(...)
  // - contract.allowance(...)
}

export function handleApproval(event: Approval): void {
  // Entities can be loaded from the store using a string ID; this ID
  // needs to be unique across all entities of the same type
  let entity = ApprovalEntity.load(event.transaction.from.toHex());

  // Entities only exist after they have been saved to the store;
  // `null` checks allow to create entities on demand
  if (entity == null) {
    entity = new ApprovalEntity(event.transaction.from.toHex());
  }

  // BigInt and BigDecimal math are supported
  entity.value = event.params.value;

  // Entity fields can be set based on event parameters
  entity.owner = event.params.owner;
  entity.spender = event.params.spender;

  // Entities can be written to the store with `.save()`
  entity.save();
}

  8. Modify the name of entity

    • enter graphtest directory
    • modify entities in subgraph.yaml:
    ---
entities:
  - TransferEntity
  - ApprovalEntity

  9. Authorize and deploy Subgraph

    Get your <DEPLOY KEY> first, find it in your home page of subgraph:

    • Authorize

      graph auth --studio <DEPLOY KEY>

      If use Hosted Service, the initing command is as follows:

      graph auth --product hosted-service <ACCESS_TOKEN>

    • enter the directory of subgraph

      cd ./<SUBGRAPH_NAME>

    • BUILD SUBGRAPH

      graph codegen && graph build

    • DEPLOY SUBGRAPH

      graph deploy --studio <SUBGRAPH_NAME>

      If use Hosted Service, the initing command is as follows:

      graph deploy --product hosted-service <GITHUB_USER>/<SUBGRAPH NAME>
      • You have to input Version Label, such as 0.0.1, or it will give an error prompt You must provide a version label.

Check if subgraph deploy

Enter your home page of subgraph from subgraph panel, check index progress, it could be called when the progress is finished.

An sample request has been pre-produced here, click to button to request data. So far we have succeeded

Graph Node build locally

  1. Build graph-node For convenience's sake, we use the docker compose provided by official to deploy node, database and IPFS.

Note: if yours are latest mac (big sur system), do not use brew cask install docker when install docker, please refer to https://www.jianshu.com/p/50037be9c00d for details.

graph-node:
  image: graphprotocol/graph-node
  ports:
    - '8000:8000'
    - '8001:8001'
    - '8020:8020'
    - '8030:8030'
    - '8040:8040'
  depends_on:
    - ipfs
    - postgres
  environment:
    postgres_host: postgres
    postgres_user: graph-node
    postgres_pass: let-me-in
    postgres_db: graph-node
    ipfs: 'ipfs:5001'
    ethereum: 'mainnet:http://127.0.0.1:8545' # the mainnet need to follow to the network of subgraph.yml
    # ethereum: 'dev:https://goerli.infura.io/v3/INFURA_ID' # could connect to test networks
    RUST_LOG: info

Note: the node in graph-node need start archive mode(when start the node, add flag --syncmode full --gcmode archive). 2. graph-node startup

  1. Startup with docker compose directly
docker-compose -f docker-compose.yml up -d

  1. Compile subgraph

    Enter the directory of subgraph to run the command

    Because the command was executed in the previous step: npx hardhat run ./scripts/deploy.js --network goerli

    subgraph.yml need to be changed as follows:

dataSources:
  - kind: ethereum/contract
    name: SimpleToken
    network: goerli

graph codegen --output-dir src/types/
graph build
  1. Deploy subgraph
graph create davekaj/SimpleToken --node http://127.0.0.1:8020

graph deploy davekaj/SimpleToken --debug --ipfs http://localhost:5001 --node http://127.0.0.1:8020
  1. Use GraphQL to query data。

subgraph

subgraph provide the data, data sources and the mode of querying with GraphQL API. Developers can use the deployed subgraph by others[17], or they can customize it and deploy

  1. GraphQL Schema
    GraphQL Schema define the type or entities of data which you want to save and query. Or you can define the config about relationship or full-text search.
  2. subgraph manifest( yaml ) manifest define the smart contract indexed by subgraph, ABI of contracts, events that subscribe to these contracts, and how to map the event datas to the node of Graph to store and allowed to query.
  3. AssemblyScript mapping AssemblyScripts mappings allows you to use the type of entities defined in schema to store indexed data. Graph CLI also uses the combination of schema and ABI of smart contract to generate the type of AssemblyScript
  4. establish relationship with @derivedFrom Define reverse queries on entities with the @derivedfrom field to create a virtual field in entities which coule be queried. But you can not configure it manually by mapping API. Actually, it is grew from the relationship defined on another entity. The relation do not make sence to the relation of store, if you only store one and grow another, the performance of indexing and querying will be better.

Reference

official documentation:

reference document for this project:

other reference documents: