refs https://github.com/TryGhost/Team/issues/2370
Due to a possible bug in either `@sentry/node` (mainly the Express
middlewares and the usage of deprectated Domain) and Node v16+,
unhandled promise rejections are transformed into uncaught exceptions
and cause Ghost to crash in unexpected situations.
Reverting to `v7.11.1` fixes this (but definitely not ideal at all)
because errors are caught in the Express middleware.
Reproduction repo:
https://github.com/SimonBackx/sentry-node-unhandled-rejection-crash
fixes https://github.com/TryGhost/Team/issues/2377
When there is an error thrown that is not a Ghost error, there is no
status code in the error. Calling res.writeHead with an undefined status
code, throws an error and crashes Ghost.
This change fixes that and adds logging for those errors.
refs https://github.com/TryGhost/Team/issues/2371
- playwright tests were broken due to state changes based on prev tests that were not accounted for
- in case of multiple newsletters, portal tests expected another step between stripe checkout for newsletter selection
- site settings test was disabling members, but not re-enabling it back
refs TryGhost/Team#2371
- check that members can unsubscribe from newsletters by toggling
preferences in their account settings
- check that member can log out
refs
f5aae1e2c5
refs
0f9ed54a6f
- changing playwright portal tests to work for single tier setup caused failure for comped upgrade tests as they were relying on button text that changed
fixes https://github.com/TryGhost/Team/issues/2368
- Removed the usage of the `isLocalContentImage` Koenig util for the
email header and feature image url generation.
- While we were trying to set the width to 1200px, we didn't have that
size hardcoded. So that url would redirect back to the original location
instead of serving a smaller image. So I added a new internal size to
the `imageOptimization` config.
- This is fixed in both the new and old email flow and includes some
extra tests for the new flow.
refs. https://github.com/TryGhost/Team/issues/2371
- two extra assertion was needed for discount and free-trials to check
if the offers are listed in ‘Active’ offers and the URLs load portal
refs https://github.com/TryGhost/Team/issues/2371
- Test enabling private site and checking access with a password
- The test flow is lacking a check for site access through password
due to a Playwirght bug. This should be cleaned up in the future
refs. https://github.com/TryGhost/Team/issues/2371
- deleteAllMembers was an unnecessary step
- since there's a generated code appended to the name of the archived offer, it had to be shorter to avoid potential naming conflicts
refs https://github.com/TryGhost/Team/issues/2371
- Adds a test for publishing and sending
- Adds a test for email only sending
- Updated some util methods in the publishing spec to remove the dependency on the post bookmark (which is not present for email only posts)
refs https://github.com/TryGhost/Team/issues/2371
Note that the "Choose" button is "Continue" when running this test
standalone so currently it needs to run with the full suite.
refs https://github.com/TryGhost/Team/issues/2371
- extracting the re-used actions to utils allows tests to be self-descriptive rather than relying on comments and keeps the selectors and related actions in one place to help refactoring if/when they change
refs. https://github.com/TryGhost/Team/issues/2371
- Test for archived offers should be moved to ‘Archived’ view of the offer list in Admin, and the offer URL should redirect to the site's homepage for logged out visitors
refs https://github.com/TryGhost/Team/issues/2371
- bumped timeout between saving and refreshing to account for slower
speeds in CI
- increased specificity for the frontend text comparisons so the output
when failing is smaller and easier to parse
refs https://github.com/TryGhost/Team/issues/2371
- the 100ms timeout was enough for local tests to pass but was still failing on CI
- bumped to 200ms and skipped the creation of a new paragraph to reduce what the editor is doing
refs https://github.com/TryGhost/Team/issues/2371
- added timeout between clicking the editor and starting to type otherwise some of the typing events could be missed causing a mismatch in actual vs expected output
ref https://github.com/TryGhost/Team/issues/2371
- updated Member exports with csv validation
- added member fixtures to be loaded into Ghost to ensure filtering
works correctly when downloading / exporting members csv.
refs https://github.com/TryGhost/Team/issues/2371
- The "data-test-*" selectors in playwright did not work with publishing channel selectors. This is a quick hack to enable working around it
refs https://github.com/TryGhost/Team/issues/2367
This ensures that a Member is not considered subscribed to any emails, so that
counts for newsletter recipients are correct. Eventually we will filter members
on their email suppression status but this is not implemented yet.
Refs https://github.com/TryGhost/Team/issues/2371
- Tests whether the post access selection of public, members, or paid-members matches the expected post visibility on the frontend.
refs https://github.com/TryGhost/Team/issues/2371
- Adds a test that schedules a post 5 seconds in the future and waits
for it to be published
- Reduced the time restrictions for scheduling:
- The minimum time in the frontend is now 5 seconds in the future (came
from 5 minutes in the future)
- The time picker now suggests 10 minutes in the future instead of the
minimum scheduling time (came from 5 minutes)
- In the backend, a post will be allowed to be scheduled if it is at
least 2 minutes in the past (came from 2 minutes in the future)
- The scheduler will publish a post if it is at least 5 minutes in the
past, and maximum 5 minutes in the future (came from 2 minutes)
refs https://github.com/TryGhost/Team/issues/2371
- tests that a free member can upgrade to a paid tier via stripe checkout and the payment details are reflected in portal and member detail page on admin
refs https://github.com/TryGhost/Team/issues/2371
- tests modifying the content of a published post
- extracted publish flow into a `publishPost` function that returns a new browser page object with the newly created post loaded
refs https://github.com/TryGhost/Team/issues/2369
- this checks whether the Offer redemption count is set to 1, which
would be indicative that the Offer was successfully counted as
redeemed
refs https://github.com/TryGhost/Toolbox/issues/479
- this includes a handful of improvements to get Playwright working on a
local environment including:
- adding `testing-browser` environment so we don't nuke `development`
environments, and makes all the necessary changes to get Ghost to
behave when this is running
- stopped running one global instance of Ghost as this doesn't provide
a clean environment
- copies a few default fixtures that are needed for the new
environment
- we should start to keep tests grouped by their area, so first we split
by Admin tests and then Portal tests, and within that we split into
setup/Tiers/Offers etc
refs https://github.com/TryGhost/Toolbox/issues/479
- e2e and integration test suites are running on port 2369. Playwright was not following this convention, without good reason.
- Port 2368 is the default port for development and production processes, so using it for test environment is not ideal
refs: https://github.com/TryGhost/Toolbox/issues/481
* Correctly setup environment variable to run both local & staging browser-based tests
* Use non-production Ghost Admin build, since production builds require HTTPS to use Stripe Connect
refs https://github.com/TryGhost/Toolbox/issues/476
- The email verification trigger and host settings related bugs have been a cause of bugs in past releases. The admin client verification source did not have any test coverage in the past.
- The members test suite size is getting out of hand. This test is quite verbose, because of the state it's trying to check.
- In the future we should consider splitting up Member API (and probably other) test suites into smaller pieces.
no issue
- The sleep method has been used in 8 modules reimplementing the same thing over and over again. It's usually a sign of async event processing outside of the request/response loop. It's good to have a single point of implementation for a "hack" like this, so we could track it easier and address the even processing delay in a more optimal way centrally if it ever becomes a bottleneck
