Spacelocust / ga-exercice

Github Action Exercices

Events

(Event List)

For this part, you will have to trigger the workflow under specific conditions. Each item on the list correspond to a particular scenario where the workflow should (or should not) run.

Push

1. Create and publish 3 branches, "dev", "staging", "feat/feature-1"
2. When you have a push on the branch "dev" only
3. When you have a push on any branch except "staging"
4. When you have a push on any branch where the name starts with "feat"
5. When only the files with a ts extension have been modified
6. When a specific folder have been modified

Pull request

1. When someone creates a PR
2. When someone closed a PR

Scheduled

Help

1. Every 5 minutes
2. Every 1h 30m
3. Every Tuesday at 3am

Push & PR

1. Create two jobs inside a new workflow
2. When you have a push, execute both job
3. When you have a PR, execute only the first job.
4. Create a dev branch, and inside it create a dummy change.txt file. Put your name in it. Publish this branch
5. Create a pull request from the dev branch to main. You should see the first job executing directly inside the PR (from the github UI)

Manual & Custom

1. When you click on run workflow button in the UI
2. Create two workflows
3. Trigger the second workflow from the first one, and pass the github event name from the first one to the second with a variable named "firstJobEvent"

Troubleshooting tools & logs

Install the nektos/act tool to run your workflows locally.

1. Once act is installed, run the command from your cli (act) and let it download the docker image it needs.
2. Create a new repository secret called ACTIONS_STEP_DEBUG and set its value to true
3. Re-run your last workflow from the github action UI
4. Get a particular line from the job's output and extract a link that points to that line. Save it in a result.
5. Download the logs for an entire job
6. Create a new repository secret called ACTIONS_RUNNER_DEBUG and set its value to true
7. Remove the ACTIONS_STEP_DEBUG secret
8. Re-run the last workflow
9. Download the logs from the latest job, and compare the two debug mode. What's the major difference ?
10. Run your workflow with act. Don't forget to specify the event type.

Runners

Here, you will have to compose with the runner to achieve the demanded results. A node app has been added, in the node-app directory

Job's default

1. Write a workflow that installs dependancies and runs the test. Specify a default folder for all the steps inside the job.

Matrix

1. Create a matrix strategy for a node js app that will install dependancies for node 12, 14, 16
2. Create a second matrix strategy to test the node app on ubuntu and windows

Docker image runner

1. Now, instead of installing node js with the action, use a node image for the tests of the node-app

Self-hosted runners

1. Download the self-hosted runner script
2. Create a workflow that will install the node dependancies, test and run the app (npm start)
3. Look at the result from the job. Why is this a bad idea for public repositories ?

Environnements

1. In your repo's settings, create two environnements. One for production, one for staging.
2. Write a workflow that deploy the joke app to heroku on a push to the main branch
3. Write a workflow that deploy the joke app to heroku on a push to the dev branch (on a different heroku app)
4. For each workflow, specify the corresponding environnement and the url of the app you created
5. For the production environnement, set the following secret : JOKE_URL=https://v2.jokeapi.dev/joke/Programming
6. For the staging environnement, set the following secret : JOKE_URL=https://v2.jokeapi.dev/joke/Any?blacklistFlags=racist,sexist
7. Use the secret to incorporate different joke url depending on the environnement (Help)