cd72dbe4ea
* feat: fix a small-ish nitpick where extensions would reload after login this could potentially try to re-register all private extensions after each auth signal, which is being prevented anyway because of duplicate aliases, but still nice to remove and not have to listen to * feat: align login UI extension load with backoffice, i.e. wait for external load before registering core extensions * build(deps): bump @hey-api to newest and re-generate client * chore: adds extra error logging * feat: adds retry logic to the api interceptor * feat: warn about incomplete actions * fix: the body was already plain text, but we need to ensure the headers say so as well * feat: warns the user when actions could not be completed * build(deps): update @hey-api/client-fetch * chore: generate new api * feat: simplify error handling to just UmbApiError and UmbCancelError * feat: moves error notifications from interceptors into tryExecute, so you more easily can opt out of it and everything is gathered in one place * feat: recreate responses with correct 'status' and 'statusText' * build: stop dotnet processes after debug session * feat: extrapolate common logic into helper method to create responses * feat: returns a UmbProblemDetails like object on interceptors to be handled by tryExecute * chore: deprecates duplicate, outdated UmbProblemDetails interface and type guard * feat: uses the 'title' of the problem details object to convey the main message * chore: 401 and 403 uses their own interceptors * feat: show no notification if 401 * feat: uses the real request method and url (instead of the template placeholders) to tell the user what did not succeed * feat: retry requests with no timeout/race * feat: throttle and delay signals and disallow them from being updated from the outside * chore: adds more logging to timeouts * chore: optimise imports * test: ignores any test files left in node_modules folder * feat: uses auditTime to wait a bit before showing the timeout screen * feat: adds 404 handling to error interceptor * chore: cleans up after response modification * feat: preserve only a few headers this mimicks the v15 behavior * feat: lets the UI handle 404 errors instead of notifying directly * test: uses create action menu option instead to find the correct locator, and skips a seemingly unnecessary timeout
Umbraco Acceptance Tests
You can watch a video following these instructions here and a longer UmbraCollab recording here. Make sure to use the latest recommended main branch rather than v10 that's mentioned in the video. Alternatively, follow along the instructions below.
Prerequisites
- NodeJS 22+
- A running installed Umbraco on url: https://localhost:44339 (Default development port)
- Install using a
SqlServer/LocalDbas the tests execute too fast forSqliteto handle.
- Install using a
Getting started
The tests are located in the project/folder as Umbraco.Tests.AcceptanceTests. Make sure you run npm ci & npx playwright install in that folder, or let your IDE do that.
The script will ask you to enter the username and password for a superadmin user of your Umbraco CMS.
Executing tests
There are two npm scripts that can be used to execute the test:
npm run test- Executes the tests headless.
npm run ui- Executes the tests in a browser handled by a playwright application.
In case of errors it is recommended to use await page.pause() so you can step through your test.
Executing single tests
If you wish to run a single set of tests, which may be helpful when writing tests you can use the following command. As before, you need to run these tests in the 'tests/Umbraco.Tests.AcceptanceTest' folder.
npx playwright test <testname.ts>
For example to run the Login Test:
npx playwright test tests/DefaultConfig/Login/Login.spec.ts
To run a single test (if you have several in a file), you can use this syntax.
npx playwright test -g "<name of test>"
For example:
npx playwright test -g "can create content with the document link"
Executing tests in UI Mode
If you would like to have an overview of all your test, to be able to see all the steps in the tests being executed and you would like to be able to run all of your tests one after another, and maybe only just one test. Then you should use UI Mode. As before, you need to run these commands in the 'tests/Umbraco.Tests.AcceptanceTest' folder.
npx playwright test --ui
You can also specify which tests you want to run
npx playwright test --ui tests/DefaultConfig
When entering UI Mode, you might only able to see the authenticate test. To fix this you will need to click on the 'Projects' in UI mode and select 'Chromium'. After you've done this. You should be able to see all your tests for the location you specified when running the command.
Environment Configuration
The environment configuration is begin setup by the npm installation script.
This results in the creation of this file: .env.
This file is already added to .gitignore and can contain values that are different for each developer machine.
The file has the following content:
UMBRACO_USER_LOGIN=email for superadmin
UMBRACO_USER_PASSWORD=password for superadmin
URL=https://localhost:44339
You can change this if you like or run the config script to reset the values, type "npm run config" in your terminal.
Documentation
For further documentation on Playwright, see the Playwright documentation.