Testing Redwood in GitHub actions
A good testing strategy is important for any project. Redwood offers a few different types of tests that you can write to make your app more robust—to ship with confidence. In this guide we'll focus on how to run your Redwood tests in GitHub Actions, so you can test your app on every push or pull request.
We'll set up a tiny project with a few tests and a Postgres database that'll be created and used in every test run on GitHub. If you need to set up tests for an existing project, or if you want to write better tests, check out the (amazing) Testing docs.
Background
Let's start by introducing some concepts and products that we'll use in this guide. Then we'll get to the code.
Continuous Integration
Continuous Integration (CI) is the practice of automatically running your tests on every push or pull request. This is a great way to catch bugs before they're merged into your main branch.
Continuous Deployment
Continuous Deployment (CD) is the practice of automatically deploying your app (and database in this case) to a server after every successful test run. This is a great way to make sure your app or database is always up to date.
GitHub Actions and GitHub Secrets
GitHub Actions is a service that allows you to run a series of commands on a virtual machine. You can use it to run tests, deploy your app, or do anything else you may think of. It's free for public repositories and has a free tier for private ones. For more information, check out GitHub Actions' docs.
GitHub Secrets is a way to store sensitive information like API keys or passwords needed by GitHub Actions. They are encrypted and only exposed to the GitHub Actions service. You can use them to pass sensitive information to your tests or deploy script. For more information, check out GitHub Secrets' docs.
How to run tests in GitHub Actions
All right, let's get to the code. In this how to, we'll focus on how to run your tests in GitHub Actions, but not how to write your tests (see the Testing doc for that).
If you already have a project, you can skip to 4. Set up GitHub Actions.
1. Create a Redwood app
Start by creating a Redwood app and cding into it:
yarn create redwood-app rw-testing-gh-actions
cd rw-testing-gh-actions
Then make sure everything is working:
yarn rw test
If it is, you should see something like this:
...
PASS api api/src/directives/requireAuth/requireAuth.test.ts
PASS api api/src/directives/skipAuth/skipAuth.test.ts
Test Suites: 2 passed, 2 total
Tests: 3 passed, 3 total
Snapshots: 0 total
Time: 1.669 s
Ran all test suites.
Watch Usage: Press w to show more.
2. Modify the Prisma schema
For the purpose of this how to, we'll use the UserExample model that comes with the Redwood app.
We'll also change the database to Postgres since that's what we'll be using in our GitHub Actions.
Here's a handy guide for how to set it up locally. We'll need the connection string so our Redwood app knows where to store the data
On to the changes. Modify your schema.prisma file to look like this:
datasource db {
provider = "postgresql"
url = env("DATABASE_URL")
}
generator client {
provider = "prisma-client-js"
binaryTargets = "native"
}
model UserExample {
id Int @id @default(autoincrement())
email String @unique
name String?
}
Then add your connection strings to your .env file:
Make sure you don't commit this file to your repo since it contains sensitive information.
DATABASE_URL=postgres://postgres:postgres@localhost:54322/postgres
TEST_DATABASE_URL=postgres://postgres:postgres@localhost:54322/postgres
You need one connection string for your development database and one for your test database. Read more about it in the testing doc's The Test Database section.
Next, navigate to the scripts/seed.ts file. Uncomment the contents of the array that contains the "fake" users. We'll also use the createMany method for inserting records in the database so we can skip the duplicates (see the Prisma docs for more info). When you're all done, it should look like this:
...
const data: Prisma.UserExampleCreateArgs['data'][] = [
// To try this example data with the UserExample model in schema.prisma,
// uncomment the lines below and run 'yarn rw prisma migrate dev'
//
{ name: 'alice', email: 'alice@example.com' },
{ name: 'mark', email: 'mark@example.com' },
{ name: 'jackie', email: 'jackie@example.com' },
{ name: 'bob', email: 'bob@example.com' },
]
console.log(
"\nUsing the default './scripts/seed.{js,ts}' template\nEdit the file to add seed data\n"
)
// Note: if using PostgreSQL, using `createMany` to insert multiple records is much faster
// @see: https://www.prisma.io/docs/reference/api-reference/prisma-client-reference#createmany
Promise.all(
//
// Change to match your data model and seeding needs
//
data.map(async (data: Prisma.UserExampleCreateArgs['data']) => {
const record = await db.userExample.createMany({
data,
skipDuplicates: true,
})
console.log(record)
})
)
...
Finally, migrate your database:
yarn rw prisma migrate dev --name init
3. Generate the UserExample scaffold
We need some real tests to work with. Scaffolding out the UserExample model gives us everything we need to create "users" in our app, including some of the services tests which interact with our test database:
yarn rw g scaffold UserExample
Make sure everything is still working:
yarn rw test
You should see something like this:
PASS web web/src/lib/formatters.test.tsx
PASS api api/src/services/userExamples/userExamples.test.ts
Test Suites: 2 passed, 2 total
Tests: 21 passed, 21 total
Snapshots: 0 total
Time: 3.587 s
Ran all test suites related to changed files in 2 projects.
4. Set up GitHub Actions
Create a new file in the .github/workflows directory (create those directories if they don't exist) called ci.yml and add the following:
This action only runs when the main branch is updated, but you can configure it to run on any other branch.
name: Redwood CI
on:
push:
branches: ['main']
pull_request:
branches: ['main']
env:
DATABASE_URL: postgres://postgres:postgres@localhost:5432/postgres
TEST_DATABASE_URL: postgres://postgres:postgres@localhost:5432/postgres
jobs:
build:
runs-on: ubuntu-latest
strategy:
matrix:
node-version: [16.x]
# See supported Node.js release schedule at https://nodejs.org/en/about/releases/
services:
# Label used to access the service container
postgres:
# Docker Hub image
image: postgres
# Provide the password for postgres
env:
POSTGRES_PASSWORD: postgres
# Set health checks to wait until postgres has started
options: >-
--health-cmd pg_isready
--health-interval 10s
--health-timeout 5s
--health-retries 5
ports:
# Maps tcp port 5432 on service container to the host
- 5432:5432
steps:
- uses: actions/checkout@v3
- name: Use Node.js ${{ matrix.node-version }}
uses: actions/setup-node@v3
with:
node-version: ${{ matrix.node-version }}
# install all the dependencies
- run: yarn install
# build the redwood app
- run: yarn rw build
# run the api tests
- run: yarn rw test api --no-watch
# run the web tests
- run: yarn rw test web --no-watch
Now push your changes to the main branch on GitHub; the "Redwood CI" action we just made will run like this:
- Set up the job ("build")
- Initialize the containers and create the postgres instance
- Checkout the code
- Set up Node.js
- Install the Redwood app's dependencies
- Build the Redwood app
- Run the api tests
- Run the web tests
- Clean up the environment
At this point, if all is well, you may start feeling the joy of automated tests! You push a commit, the Action runs, your tests pass, and you get a green checkmark. To savor this moment, consider updating one of your unit tests, making it fail. Push again. Watch it fail. Fix it. Push again. Watch it pass. Repeat and enjoy.
5. Set up CI on pull requests only
We want tests to run on every pull request so we can make sure that our code is working as expected.
Update the ci.yml file by removing the push event. The first lines should look like this:
name: Redwood CI for Pull Requests
on:
pull_request:
branches: ['main']
...
Now, if you open or push to a pull request, this action will run and you'll see something like this:
Once the action is done running, you can see the results in the "Conversation" tab:
6. Deploy the database changes to an actual database
Now for the CD—we want to use another action to deploy the database changes to an actual database, so we can automatically deploy the latest and greatest to a real environment. In this action we'll run the tests one more time against the local database, then deploy the database migrations to the external database.
Create a new file in the .github/workflows directory called cd.yml and add the following:
name: Redwood CD for database deployment
on:
push:
branches: ['main']
env:
DATABASE_URL: postgres://postgres:postgres@localhost:5432/postgres
TEST_DATABASE_URL: postgres://postgres:postgres@localhost:5432/postgres
jobs:
build:
runs-on: ubuntu-latest
strategy:
matrix:
node-version: [16.x]
# See supported Node.js release schedule at https://nodejs.org/en/about/releases/
services:
# Label used to access the service container
postgres:
# Docker Hub image
image: postgres
# Provide the password for postgres
env:
POSTGRES_PASSWORD: postgres
# Set health checks to wait until postgres has started
options: >-
--health-cmd pg_isready
--health-interval 10s
--health-timeout 5s
--health-retries 5
ports:
# Maps tcp port 5432 on service container to the host
- 5432:5432
steps:
- uses: actions/checkout@v3
- name: Use Node.js ${{ matrix.node-version }}
uses: actions/setup-node@v3
with:
node-version: ${{ matrix.node-version }}
# install all the dependencies
- run: yarn install
# build the redwood app
- run: yarn rw build
# run the api tests
- run: yarn rw test api --no-watch
# run the web tests
- run: yarn rw test web --no-watch
# run migrations on the actual database
- run: yarn rw prisma migrate deploy
# run seed script in the actual db
- run: yarn rw prisma db seed
The main changes are:
- We only run the action on push events to the
mainbranch - We run the migrations and seed scripts after the tests
7. Set up GitHub Secrets
Because you're using an actual Postgres instance in your action, you need to set up the secrets for the database connection so that the username and password stay secret.
Go to the "Settings" tab in your GitHub repo and click "Secrets", then "Actions", then "New repository secret".
In the name field, type DATABASE_URL. In the value field, put the actual secret—something like this:
postgres://[USER_NAME]:[PASSWORD]@[HOST]:[PORT]/postgres
When you're done, click "Add secret". This creates a new secret that you can use in your GitHub Actions. In this case, it species the connection string for the database we'll deploy changes to.
You can use the secret in your GitHub Actions by using the ${{ secrets.DATABASE_URL }} syntax:
env:
DATABASE_URL: ${{ secrets.DATABASE_URL }}
Now you can merge the PR and the database changes will be tested, then deployed to the actual database.
What's next? It is up to you—refine and streamline!
As you consider automating your project workflows, keep the following wise philosophical observation in mind...
Civilization advances by extending the number of important operations we can perform without thinking.
—Alfred North Whitehead