Skip to main content

Using the API

In this guide we will create a web interface that consumes and displays data from the Pegasys Subgraph. The goal is to provide a quick overview of a setup that you can extend to create your own UIs and analytics around Pegasys data.

Many different libraries can be used to create an interface and a connection to the subgraph graphql endpoint, but in this guide we will use React for the interface, and Apollo Client for sending queries. We'll also be using yarn for dependency management.

Setup and Installs#

We'll need to create the basic skeleton for the application. We'll use create-react-app for this. We'll also add the dependencies we need. Navigate to your root location in your command line and run:

yarn create react-app pegasys-demo
cd pegasys-demo
yarn add apollo-client apollo-cache-inmemory apollo-link-http graphql graphql-tag @apollo/react-hooks
yarn start

In your browser you should see the default React app running. In a text editor open App.js within src and replace the contents with this stripped down boilerplate. We'll add to this as we go.

import React from 'react'
import './App.css'
function App() {
return <div></div>
}
export default App

Graphql Client#

We need to set up some middleware in order to make requests to the Pegasys subgraph and receive data. To do this we'll use Apollo and create a graphql client to handle this.

  1. Add the imports shown below and instantiate a new client instance. Notice how we use the link to the Pegasys subgraph here.
import React from "react"
import "./App.css"
import { ApolloClient } from "apollo-client"
import { InMemoryCache } from "apollo-cache-inmemory"
import { HttpLink } from "apollo-link-http"
export const client = new ApolloClient({
link: new HttpLink({
uri: 'https://api.graph.pegasys.exchange/subgraphs/name/pollum-io/pegasys/'
}),
cache: new InMemoryCache(),
})
function App() {
return <div></div>
}
export default App
  1. We also need to add a context so that Apollo can handle requests properly. In your index.js file import the proper provider and wrap the root in it like this:
import React from 'react'
import ReactDOM from 'react-dom'
import App from './App'
import registerServiceWorker from './registerServiceWorker'
import './index.css'
import { ApolloProvider } from 'react-apollo'
import { client } from './App'
ReactDOM.render(
<ApolloProvider client={client}>
<App />
</ApolloProvider>,
document.getElementById('root')
)
registerServiceWorker()

Writing the queries#

Next we'll construct our query and fetch data. For this example we will fetch some data about the Dai token on Pegasys. We'll get the current price, and total liquidity across all pairs. We'll be using the Dai address as an id in this query. We'll also fetch the USD price of SYS to help create USD conversion for Dai data.

  1. First we need to define the query itself. We'll use gql to parse a query string into the GraphQL AST standard. Import the gql helper into the app and use it to create the query. Add the following to your App.js file:
import gql from 'graphql-tag'
const DAI_QUERY = gql`
query tokens($tokenAddress: Bytes!) {
tokens(where: { id: $tokenAddress }) {
derivedSYS
totalLiquidity
}
}
`
const SYS_PRICE_QUERY = gql`
query sysPrice {
bundle(id: "1") {
sysPrice
}
}
`

We use an id of 1 for the bundle because there is only one hardcoded bundle in the subgraph.

Fetch data#

Now we're ready to use these queries to fetch data from the Pegasys subgraph. To do this we can use the useQuery hook which uses our client instance to fetch data, and gives us live info about the status of the request. To do this add the following to your App.js file:

import { useQuery } from '@apollo/react-hooks'
const { loading, error, data: sysPriceData } = useQuery(SYS_PRICE_QUERY)
const { loading: daiLoading, error: daiError, data: daiData } = useQuery(DAI_QUERY, {
variables: {
tokenAddress: '0xeFAeeE334F0Fd1712f9a8cc375f427D9Cdd40d73'
}
})

Notice we're using the Dai token address to fetch data about Dai.

Formatting Response#

Now that we have our data we can format it and display it in the UI. First, we parse the return data to get the actual data that we want. Then we'll use it to get the USD price of Dai. Lastly we'll insert this data into the UI itself.

These queries will return an response object for each query. Within each one we're interested in the root field we defined in the query definition. For the daiData response we defined this as tokens, and for the sysPriceData query we defined this as sysPrice. Within each one we'll get an array of results. Because we're only querying for single entities we'll reference the 0 index in the data array.

Add the following lines to your App.js file to parse the responses:

const daiPriceInSys = daiData && daiData.tokens[0].derivedSYS
const daiTotalLiquidity = daiData && daiData.tokens[0].totalLiquidity
const sysPriceInUSD = sysPriceData && sysPriceData.bundles[0].sysPrice

Displaying in the UI#

Finally we can use our parsed response data to hydrate the UI. We'll do this in two steps.

  1. First we'll create loading states. To detect if a query is still pending a response we can reference the loading variables we've already defined. We'll add two loading states, one for the Dai price, and one for the Dai total liquidity. These may flicker fast because the time to query is fast.

  2. Populate with loaded data. Once we detect that the queries have finished loading we can populate the UI with the real data.

To do this add the following lines in the return function of your App.js file:

return (
<div>
<div>
Dai price:{' '}
{sysLoading || daiLoading
? 'Loading token data...'
: '$' +
// parse responses as floats and fix to 2 decimals
(parseFloat(daiPriceInSys) * parseFloat(sysPriceInUSD)).toFixed(2)}
</div>
<div>
Dai total liquidity:{' '}
{daiLoading
? 'Loading token data...'
: // display the total amount of DAI spread across all pools
parseFloat(daiTotalLiquidity).toFixed(0)}
</div>
</div>
)

Next steps#

This should render a very basic page with these two stats about the Dai token within Pegasys. This is a very basic example of what you can do with the Pegasys subgraph and we encourage you to build out more complex and interesting tools!

You can visit our analytics site to see a more advanced analytics page, or visit the github for more detailed examples of using the Pegasys subgraph to create UIs.

Review#

In the end your App.js file should look like this:

import React, { useEffect } from 'react'
import './App.css'
import { ApolloClient } from 'apollo-client'
import { InMemoryCache } from 'apollo-cache-inmemory'
import { HttpLink } from 'apollo-link-http'
import { useQuery } from '@apollo/react-hooks'
import gql from 'graphql-tag'
export const client = new ApolloClient({
link: new HttpLink({
uri: 'https://api.graph.pegasys.exchange/subgraphs/name/pollum-io/pegasys/'
}),
fetchOptions: {
mode: 'no-cors'
},
cache: new InMemoryCache()
})
const DAI_QUERY = gql`
query tokens($tokenAddress: Bytes!) {
tokens(where: { id: $tokenAddress }) {
derivedSYS
totalLiquidity
}
}
`
const SYS_PRICE_QUERY = gql`
query bundles {
bundles(where: { id: "1" }) {
sysPrice
}
}
`
function App() {
const { loading: sysLoading, data: sysPriceData } = useQuery(SYS_PRICE_QUERY)
const { loading: daiLoading, data: daiData } = useQuery(DAI_QUERY, {
variables: {
tokenAddress: '0xeFAeeE334F0Fd1712f9a8cc375f427D9Cdd40d73'
}
})
const daiPriceInSys = daiData && daiData.tokens[0].derivedSYS
const daiTotalLiquidity = daiData && daiData.tokens[0].totalLiquidity
const sysPriceInUSD = sysPriceData && sysPriceData.bundles[0].sysPrice
return (
<div>
<div>
Dai price:{' '}
{sysLoading || daiLoading
? 'Loading token data...'
: '$' +
// parse responses as floats and fix to 2 decimals
(parseFloat(daiPriceInSys) * parseFloat(sysPriceInUSD)).toFixed(2)}
</div>
<div>
Dai total liquidity:{' '}
{daiLoading
? 'Loading token data...'
: // display the total amount of DAI spread across all pools
parseFloat(daiTotalLiquidity).toFixed(0)}
</div>
</div>
)
}
export default App