26
Creating a GraphQL Based Habit Tracker with Hasura and React (GraphQL/Hasura 101)
GraphQL is an alternative to Rest API created by Facebook:
Rest API require you to make request to many URLs while all GraphQL queries are actually post requests to a single url
Rest API by default require manually writing documentation unless you configure a tool like Swagger, GraphQL API are self-documenting by default
RestAPI typically give large amounts of information whether you need it or not, while GraphQL allows you to specify which data you need.
Although, the downside of creating GraphQL API is having define Types and Resolvers...
Exactly like in typescript or database schemas, typing means defining what datatypes the properties of your data consist of. This can mean typing everything a third time (assuming your database require a definition of schema and your using typescript or a typed langauge to write your API).
Instead of different endpoints that trigger different route handlers, GraphQL has several pre-defined queries (get information) and mutations (create, update, delete information) in the APIs type definitions. Each query and mutation needs a corresponding function referred to as a resolver.
Bottom line, manually building out GraphQL API can result in extra boilerplate in coding all the types and resolvers needed. The benefit is the self-documenting, but still tedious.
Although, what if I said you could have it all.
Now there are several ways to get a GraphQL api pre-made for you like using a Headless CMS like GraphCMS but one platform offers a high level of flexibility and some pretty cool unique features, and that is Hasura.
- Auto generated GraphQL api based on your existing databse schemas
- ability to create custom queries and mutations
- ability to create events and web hooks to automate tasks
- hosted and self-hosted options
- REST API available too if you prefer
Head over to Hasura.io and create a new account and create a new project
Once the project is created launch the console
We need to attach a database to our project (under data), we can do so easily for free using our heroku account (get one if you don't have one).
Once the database is connected click on manage the database then click on create table.
- table name: habits
property | type | ------- |
---|---|---|
id | integer (auto increment) | primary key |
habit | text | |
count | integer | default: 0 |
- Once the table has been added head over to the API tab where you will see GraphiQL a tool for testing GraphQL APIs (think postman).
- On the far right is the documentation explorer to read the documentation that has been created for your api
- On the far left you can see a listing of the queries that have been created
I recommend spend like 30 minutes trying to see if you can figure out how to add, retrieve, update and delete data using graphQL syntax and using the API documentation. I'll summarize below when your done.
This query will get us all the Habits
{
habits {
id
habit
count
}
}
This mutation adds a habit and then gets the list of habits in return
mutation {
insert_habits(objects: {
habit: "Exercise",
count: 3
}){
affected_rows
returning {
id
habit
count
}
}
}
This is a mutation that will update a Habit with the proper id
mutation {
update_habits_by_pk(pk_columns:{id: 3} _set: {count: 4}){
id
habit
count
}
}
This mutation deletes a habit with the proper id
mutation {
delete_habits_by_pk(id:3){
id
habit
count
}
}
So our API is essentially deployed and tested! That was super easy!
You have a few primary options for how to make GraphQL calls from your frontend javascript.
You can use the tried and true fetch or axios to make the call the you'd normally make. Just keep in mind you'll need your Hasura admin secret to make the request. While we can hide this from github with a .env a knowledgable dev can still use dev tools to get your secret. So for production apps you want to make sure to adjust the CORS environmental variable on your hasura project so ONLY the url of your frontend can make requests to your API.
FETCH
fetch('https://your-app-name-here.hasura.app/v1/graphql', {
method: 'POST',
headers: {
'Content-Type': 'application/json',
"x-hasura-admin-secret": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
},
body: JSON.stringify({ query: '{
habits {
id
habit
count
}
}' }),
})
.then(res => res.json())
.then(res => console.log(res));
Axios
axios({
url: "https://your-app-name-here.hasura.app/v1/graphql"
method: 'POST',
headers: {
'Content-Type': 'application/json',
"x-hasura-admin-secret": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
},
data: JSON.stringify({ query: '{
habits {
id
habit
count
}
}' }),
})
.then(res => console.log(res.data));
If making a mutation, the string would just be the mutation instead. Remember, mutations do require the word mutation in the string like the examples we did in GraphiQL.
To configure Apollo client for a React Project
npm install @apollo/client graphql
create a .env file with your url and hasura secret
REACT_APP_HASURA_SECRET=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
REACT_APP_HASURA_URL=https://xxxxxxxxxxxx.hasura.app/v1/graphql
This inside your index.js (assuming your using create-react-app):
import React from "react";
import ReactDOM from "react-dom";
import "./index.css";
import App from "./App";
import reportWebVitals from "./reportWebVitals";
import { ApolloClient, InMemoryCache, ApolloProvider } from "@apollo/client";
// New Apollo Client with Settings
const client = new ApolloClient({
// URL to the GRAPHQL Endpoint
uri: process.env.REACT_APP_HASURA_URL,
// cache strategy, in this case, store in memory
cache: new InMemoryCache(),
// any custom headers that should go out with each request
headers: {
"x-hasura-admin-secret": process.env.REACT_APP_HASURA_SECRET,
},
});
ReactDOM.render(
<ApolloProvider client={client}>
<React.StrictMode>
<App />
</React.StrictMode>
</ApolloProvider>,
document.getElementById("root")
);
// If you want to start measuring performance in your app, pass a function
// to log results (for example: reportWebVitals(console.log))
// or send to an analytics endpoint. Learn more: https://bit.ly/CRA-vitals
reportWebVitals();
Now you can use the useQuery
and useMutation
hooks where needed!
import {useQuery, useMutation, gql} from "@apollo/client"
function App() {
// GraphQL Query String
const QUERY_STRING = gql`{
habits {
id
habit
count
}
}`
// run query using the useQuery Hook
// refetch is a function to repeat the request when needed
const {data, loading, refetch, error} = useQuery(QUERY_STRING)
// return value if the request errors
if (error){
return <h1>There is an Error</h1>
}
// return value if the request is pending
if (loading) {
return <h1>The Data is Loading</h1>
}
// return value if the request is completed
if (data){
return <div>
{data.habits.map(h => <h1 key={h.id}>{h.habit} {h.count}</h1>)}
</div>
}
}
export default App;
make-graphql-query
is a small lightweight library I made for making graphQL queries easy and simple in a framework agnostic way. It's just a tiny abstraction to eliminate a lot of boilerplate in using fetch/axios. Here is how you would use it in React.
- install
npm install make-graphql-query
create a .env file with your url and hasura secret
REACT_APP_HASURA_SECRET=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
REACT_APP_HASURA_URL=https://xxxxxxxxxxxx.hasura.app/v1/graphql
- create a gqlFunc.js file in /src, this file exports a function that knows your graphql URL and automatically has any neccessary headers.
import makeGraphQLQuery from "make-graphql-query";
export default makeGraphQLQuery({
url: process.env.REACT_APP_HASURA_URL,
headers: {
"x-hasura-admin-secret": process.env.REACT_APP_HASURA_SECRET,
},
});
Then we can just import it and use it as needed!
import graphQLQuery from "./gqlFunc";
import { useState, useEffect } from "react";
function App() {
// state to hold query results
const [query, setQuery] = useState(null);
// useState to fetch data on load
useEffect(() => {
//making graphql query
graphQLQuery({
query: `{
habits {
id
habit
count
}
}`,
}).then((response) => setQuery(response));
}, []);
// pre-query completion jsx
if (!query){
return <h1>Loading</h1>
};
// post-query completion jsx
return <div>
{query.habits.map((h) => <h2 key={h.id}>{h.habit} - {h.count}</h2>)}
</div>
}
export default App;
Let's modify our Apollo and MGQ versions of our component to also create a new habit. GraphQL queries can take variables if declared, below is an example of the create mutation with variables.
mutation add_habit ($objects: [habits_insert_input!]!){
insert_habits(objects: $objects){
affected_rows
}
}
- Note the type of the variable must be an exact match as to where you use it, use GraphiQL to determine the necessary types when making your own queries.
App.js
import {useQuery, useMutation, gql} from "@apollo/client"
import { useState } from "react"
function App() {
// GraphQL Query String
const QUERY_STRING = gql`{
habits {
id
habit
count
}
}`
const MUTATION_STRING = gql`mutation add_habit ($objects: [habits_insert_input!]!){
insert_habits(objects: $objects){
affected_rows
}
}`
// run query using the useQuery Hook
// refetch is a function to repeat the request when needed
const {data, loading, refetch, error} = useQuery(QUERY_STRING)
// create function to run mutation
const [add_habit, response] = useMutation(MUTATION_STRING)
// state to hold form data
const [form, setForm] = useState({habit: "", count: 0})
// handleChange function for form
const handleChange = (event) => setForm({...form, [event.target.name]: event.target.value})
// handleSubmit function for when form is submitted
const handleSubmit = async (event) => {
// prevent refresh
event.preventDefault()
// add habit, pass in variables
await add_habit({variables: {objects: [form]}})
// refetch query to get new data
refetch()
}
// check if mutation failed
if(response.error){
<h1>Failed to Add Habit</h1>
}
// return value if the request errors
if (error){
return <h1>There is an Error</h1>
}
// return value if the request is pending
if (loading) {
return <h1>The Data is Loading</h1>
}
// return value if the request is completed
if (data){
return <div>
<form onSubmit={handleSubmit}>
<input type="text" name="habit" value={form.habit} onChange={handleChange}/>
<input type="number" name="count" value={form.count} onChange={handleChange}/>
<input type="submit" value="track habit"/>
</form>
{data.habits.map(h => <h1 key={h.id}>{h.habit} {h.count}</h1>)}
</div>
}
}
export default App;
App.js
import graphQLQuery from "./gqlFunc";
import { useState, useEffect } from "react";
function App() {
// state to hold query results
const [query, setQuery] = useState(null);
// state to hold form data
const [form, setForm] = useState({habit: "", count: 0})
// function to get habits
const getHabits = async () => {
//making graphql query
const response = await graphQLQuery({
query: `{
habits {
id
habit
count
}
}`,
});
// assigning response to state
setQuery(response);
};
// function to add a habit
const addHabit = async (variables) => {
//define the query
const q = `mutation add_habit ($objects: [habits_insert_input!]!){
insert_habits(objects: $objects){
affected_rows
}
}`
// run query with variables
await graphQLQuery({query: q, variables})
// get updated list of habits
getHabits()
}
// useState to fetch data on load
useEffect(() => {
getHabits();
}, []);
// handleChange function for form
const handleChange = (event) => setForm({...form, [event.target.name]: event.target.value})
// handleSubmit function for when form is submitted
const handleSubmit = (event) => {
// prevent refresh
event.preventDefault()
// add habit, pass in variables
addHabit({objects: [form]})
}
// pre-query completion jsx
if (!query) {
return <h1>Loading</h1>;
}
// post-query completion jsx
return (
<div>
<form onSubmit={handleSubmit}>
<input type="text" name="habit" value={form.habit} onChange={handleChange}/>
<input type="number" name="count" value={form.count} onChange={handleChange}/>
<input type="submit" value="track habit"/>
</form>
{query.habits.map((h) => (
<h2 key={h.id}>
{h.habit} - {h.count}
</h2>
))}
</div>
);
}
export default App;
Hopefully this gives you some more insight into how to use GraphQL and how easy it can be to spin up a GraphQL API using Hasura.
26