Capstone: Full Delivery Pipeline¶

You have built a complete delivery pipeline, piece by piece. This capstone takes a real change all the way from a commit to production, through every stage you built, with no step-by-step hand-holding. 💪

The goal¶

Ship a new feature to production the professional way: tested, built, versioned, deployed to staging, approved, and released, all driven by a single tag.

Your task¶

1. Add a small feature to the app: a /version endpoint that returns {"version": "1.3.0"}.
2. Add a test for it.
3. Commit and push, so the pipeline tests and builds on main.
4. Cut the release: tag v1.3.0 and push the tag.
5. Watch the full flow: test, build, publish, deploy to staging, then wait for approval.
6. Approve the production deploy, and confirm the new version is live.
7. For practice, use your Rollback workflow to return to v1.2.0, then roll forward to v1.3.0 again.

You should not need to touch the workflow files. The pipeline you built already does all of this.

Hints¶

• The new route goes in app.py; the test in test_app.py.
• A plain push runs test and build. The tag is what triggers deployment.
• Production waits on the Review deployments button.

@app.route("/version")
def version():
    return {"version": "1.3.0"}

def test_version_endpoint():
    client = app.test_client()
    response = client.get("/version")
    assert response.status_code == 200
    assert response.get_json() == {"version": "1.3.0"}

git add .
git commit -m "Add version endpoint"
git push
git tag v1.3.0
git push origin v1.3.0

✅ Success criteria¶

You have completed the intermediate track if:

• Pushing to main ran test and build, but did not deploy
• Pushing the v1.3.0 tag deployed to staging automatically
• Production waited for your approval before going live
• The /version endpoint is live in production
• You rolled back to v1.2.0 and forward again with the rollback workflow

🎓 What you've built¶

Your pipeline now delivers, not just integrates. On a release tag it:

1. Runs the tests
2. Builds and publishes a versioned image
3. Deploys to staging automatically
4. Waits for a human approval
5. Deploys to production

And if anything goes wrong, you can roll back to a known-good version in seconds. That is a genuine, production-grade delivery pipeline.

Next up: the Advanced track, where delivery levels up to GitOps, and your pipeline deploys to a Kubernetes cluster that keeps itself in sync with Git.