Feat: automated QA tests

[angela-tran]

No description provided.

[github-actions]

Coverage report

This PR does not seem to contain any modification to coverable code.

[angela-tran]

I was looking over the open pull requests just now and noticed this PR needed to be rebased. It was an easy one, so went ahead and did it.

Hopefully we'll have some time in the next slush time to finish this up.

[angela-tran]

We had some discussion for this PR, around improving how test config is done as well as adding documentation:

Config file to setup data that is currently in environment variables.

• We can start by setting env_file to tests/playwright/.env in the playwright service for this PR
• Commit a tests/playwright/.env.sample file with example variables
• Basic docs page with getting started steps

On the config piece, I think it would be well worth it to go ahead and implement a YAML config file, similar to what we do in cal-itp/littlepay, for the tests rather than the interim .env setup that's described by our workshop notes.

The blockers have been (1) not having time to spend on this and (2) uncertainty around what the schema of the YAML file should be.

This would be great to look at during our in-person workshop day.

[thekaveman]

I had a thought related to our experience with timeouts not really working for some pages...

We're currently using the playwright.sync_api to e.g. import the Page object and expect helper:

from playwright.sync_api import Page, expect


def test_dev_healthcheck(page: Page):
    page.goto("/healthcheck")

    expect(page.get_by_text("Healthy")).to_be_visible()

If we made the small change to use playwright.async_api, that might avoid the need for the .wait_for_timeout() calls:

from playwright.async_api import Page, expect


async def test_dev_healthcheck(page: Page):
    await page.goto("/healthcheck")

    await expect(page.get_by_text("Healthy")).to_be_visible()