closes: https://github.com/TryGhost/Toolbox/issues/193
- Our Casper fixture was several years out of date.
- I'd already updated the ghost-api to point to the latest API version, which was the main difference
- This makes sure the full fixture is up to date and correct, and therefore that we're truly testing if Ghost right now works
- It also adds instructions for how to update it in future
GOTCHA: the mock-express-style tests are failing if the API difference between test-theme and casper are different
- I've tried to look into why this is - it's something to do with the overridden route settings not working properly if the API version changes
- Given that we may not keep this style of testing AND we are definitely not keeping API versions, I'm ignoreing this for now
- To get around it, I'm setting both themes to be v4 API, so that Casper is exactly as in main
refs: https://github.com/TryGhost/Toolbox/issues/193
- The Casper fixture in the codebase is copied to a tmp folder structure every single time we start Ghost in an e2e test
- We shouldn't keep files here that don't actually get used!
- The files I'm removing are src files, used for building, but we already have the built files
refs: https://github.com/TryGhost/Toolbox/issues/168
- Upgraded the remaining themes to be pinned to the canary API
- This required one minor fix because the edit URL has changed for authors in v4/canary
- I also ripped out everywhere that the theme ghost-api version was being pinned to canary, as this was unreliable and only happening in a few places
- It's also going to be unnecessary code as soon as we finish changing to only having one API version
refs: https://github.com/TryGhost/Toolbox/issues/168
- The fixture represented Casper at Ghost version 1.0 which pre-dated the introduction of the posts_per_page config in package.json
- When Casper was upgraded to 2.0 in the fixtures, the lack of pagination broke the e2e tests for pagination
- This change introduces proper code to stub and override posts_per_page rather than keeping and using the old casper-1.4 fixture
- It also required me to remove a handful of CSS-based checks which are no longer true in the new theme version, but also didn't really add anything to the tests
refs: https://github.com/TryGhost/Toolbox/issues/168
- Having large theme fixtures makes tests run slower, so we're working to reduce them
- This fixture was a full copy of a very old version of Casper, but all we needed was a handful of files so that the tests can run
- This theme is also used for testing members, not just price data so I've renamed it for clarity
- The remaining files is the bare minimum we need to test these features
refs: https://github.com/TryGhost/Toolbox/issues/168
- This file was pinning various tests to v3
We're going to be dropping the idea of having multiple versions of the API in each Ghost version.
Because this has not achieved the goal of making it easier to make breaking changes, but it has
created an ordinate amount of technical debt and maintenance overhead.
As we know this is going away in the next major, there is no benefit to us constantly running tests
that check if those versions still work, especially given how long they take.
Instead we're starting work to ensure that all of our test work on canary, and that canary has
excellent test coverage so that we can be sure that our one API version works really well and that
any changes, no matter how subtle are deliberate, tracked and understood.
refs https://github.com/TryGhost/Team/issues/1252
We need a way to signal whether or not a Tier is active or archived, and
we'll be using the active flag in the same way we do for Offers.
Since we now have 2 products by default for all ghost sites, free and default paid, the usage of default product which so far was using first product needs to be updated to use the first paid product.
- updates default product usage to use first paid tier
- updates tests
refs: https://github.com/TryGhost/Toolbox/issues/150
Instead of loading all themes for each set of tests in the e2e suite, only load the frontend at all for frontend tests, and only load themes for the theme tests.
refs https://github.com/TryGhost/Toolbox/issues/158
- To make the development quicker the copy-paste code was left in the codebase but gave immediate smell. Removed it to stop bothering me every time I look at the related code :)
refs https://github.com/TryGhost/Toolbox/issues/129
- The "Origin" header is require in Admin API requests so it makes sense to bake it into a default request made by most tests. Reduces unnecesary fluf around test request setup and removes a "config" dependency in each tests suite using the "e2e framework"
refs https://github.com/TryGhost/Toolbox/issues/158
- Allows for much smaller amount of code to configure a test to work with chai-jest-snapshots. They now work automatically for all regression tests and could be enabled for other suites by adding the "--require=./test/utils/snapshots.js" parameter in respective test:* package script
- Regenerated snapshot for authentication test as the naming structure
changed a little with the snapshot metadata being taken on a higher
level in the test (uses the suite name instead of a specific describe it
used to be called from)
refs https://github.com/TryGhost/Toolbox/issues/158
- Moving common mocking/stubbing/spying logic into an outside utils module allows test suites to keep agnostic towards which framework powers mocking etc. Should also substitute email service stubbing used in multiple places
- Bonus, got rid of should dependency, which is deprecated
no issue
- The response from the server might not always contain an array with errors (for example when there's a generic 503 error). This changes prevents tests from hanging and adds at least some relevant output
refs https://github.com/TryGhost/Toolbox/issues/129
- Adds content folder creation based on fixtures for e2e tests using the e2e framework
- Mostly a copy-paste for now from existing e2e utils with some simplifications, so we can "complicate things" later only if needed
refs 24505db918
- The code is heavily borrowed from the referenced commit by @ErisDS. It starts a fresh abstraction layer for all e2e tests to use, provides simle interface for request agent initialization and state management (read the e2e-framework top file description for more)
- Main methods the framework exposes are:
- getAgent - to initialize request agent with a fresh Ghost instance
- initFixtures - to initializer a fresh db state
- resetDb - cleans up the db state if there were state manipulations during the test suite run (POST, PUT requests or non-default fixtures)
refs https://github.com/TryGhost/Toolbox/issues/129
- Having a fresh module should allow building new testing utility concepts from the ground up without being tied by the mystery baggage of the legacy localUtils/testUtils
- The outline of the concepts the framework will be handling is in the commont on the top of the e2e-framework file
refs https://github.com/TryGhost/Toolbox/issues/152
- Being able to set up test suites without blocking a port opens a door to new ways to run tests - for example this has been one of the blockers for running mocha tests in parallel
- Additional benefit is lighter statrup, which reducec the test execution time slightly. Doesn't seem like much but these things stack up!
no issue
- we've seen some instances of SQLite saying "database disk image is malformed"
- I think this happens because we copy the file whilst we are still connected to
the old DB
- this commit destroys the connection before copying the clean file to the live
DB file
refs https://github.com/TryGhost/Toolbox/issues/136
- we nuke and reinitialize the DB many times between tests
- this forms a good portion of the time taken and we shouldn't be spending
so much time on just resetting the DB back to a known state
- this commit switches out the knex-migrator reset + init calls for SQLite to
a function which keeps a clean copy of the DB and copies the file back
when we "reset"
- for MySQL, existing functionality is kept
- this massively speeds up tests because it saves ~700ms+ for every reset
- whilst this change seems to work, it's just the start so there's a
lot more refactoring needed. this change is currently gated to CI until
it's deemed safe/sane enough to run on local machines without blowing
up
refs: TryGhost/Toolbox#147
* Replaces all references to isIgnitionError with isGhostError
* Switches use of GhostError to InternalServerError - as GhostError is no longer public
There are places where InternalServerError is not the valid error, and new errors should be added to the @tryghost/errors package to ensure that we can use semantically correct errors in those cases.
https://github.com/TryGhost/Toolbox/issues/140
- Allows to fully start an instance with a custom routes.yaml file without a need to do workarounds - e.g. we used to call an internal Admin API from a regression test suite, which is not a good practice
- Providing "routesFilePath" to startGhost method copies that file to the default settings location and start the instance with that configuration. No need to do API calls or check if the routing service "isFinished"
no issue
- this commit adds a counter for the number of boots we do in tests
- which therefore allows us to calculate the average boot time we
experience
- only useful for debugging test performance
refs https://github.com/TryGhost/Toolbox/issues/135
- This optimization is expected to play a role in more consistent "backend-only" boot where the previous test state might have left over a different theme version which might cause in unwanted URL Services reainitializations.
- What has been happening here is the themes.test.js suite was uploading a theme with a v4 api and when the users api test suite loaded up it switched back to a default v2 theme, which caused routing reinitialization
- The root problem here is the themese suite is leaving a mess behind so a "restartModeGhostStart" is not really possible anymore - this should be cleaned up separately
refs https://github.com/TryGhost/Toolbox/issues/135
- The global default should stay the same as it used to be and we can introduce an override for "withFrontend:false" on casa-by-case or area-by-area bases
refs https://github.com/TryGhost/Toolbox/issues/135
- When running e2e-API test in most cases there's no need to boot Ghost instance with full frontend. This should improve the boot time which should reflect on the speed of running test suites
- The tests where the "forceStart" and "withFrontend" are used together indicate that there's still some work to do to fully separate frontend/backend boot line. The force start is also unnecessary, but was needed to reinitialize all services properly - should be investigated!
refs https://github.com/TryGhost/Toolbox/issues/135
- These fixture had to be updated due to hardcoded ids introduced in the fixtures.json for test
- This allows to boot up a Ghost instance without recalculating urls for most common database states
refs https://github.com/TryGhost/Toolbox/issues/135
- These fixtures needs id's to be able to form a stable urls and resources cache that would be used across all e2e tests when booting without a frontend
- These two things are meant to improve performance at the cost of reliability.
- Perfect for testing, however I think they make a minimal impact on modern SSDs :(
- Still worth a shot to see if it helps with CI
refs https://github.com/TryGhost/Toolbox/issues/116
- Allows to have deterministic routing on each start of the instance - can use cache when starting tests
- To make use of the new fixtures need to place them in the content/data
folder when the test suite is initialized, then the cache detection will
kick in and boot up the UrlService in a constant time intead of having
to check "urlServiceUtils.isFinished"
refs: https://github.com/TryGhost/Toolbox/issues/133
- instead of just a collection of utils, we now have a class that manages fixtures
- this should allow us to change the path to fixtures, e.g. between prod/dev and test, so that different fixtures can be loaded by default
- also makes it easier to test the fixture manager code itself
refs https://github.com/TryGhost/Members/commit/9e59f5a9
Since we have a DynamicRedirectManager for handling adding/removing
redirects at runtime, we no longer need the custom-redirects middleware.
The redirects service does however need an init method now to add the
custom redirects at Ghost boot, so it's been refactored into our Class &
DI pattern.
refs https://github.com/TryGhost/Team/issues/1175
We found the ETag header sent when serving the Admin template for /ghost/ was not changing between versions which after an upgrade could result in out of date cached content being served containing links to JS/CSS files that no longer existed.
The culprit is weak etags served by Node's `send` package, coupled with Admin template filesize not changing between versions and `npm pack` setting a fixed modification date for every file. See https://github.com/pillarjs/send/issues/176 for more details.
- updated the Admin app's controller to read the template and generate an md5 hash of the contents so we can serve a strong ETag header value when serving the `/ghost/` html
- done a fastest-possible overhaul on this style of tests to try to get them to work independently again
This is a pattern that was introduced a while ago to try to speed up our e2e tests and I'm not sure if it's staying or going
It uses a minimal frontend-only version of the boot process and a custom-built express testing tool
However it's really old and out of date because of the boot refactor and several changes since
This highlights the key problem with it - it doesn't rely on any of our "core" boot process, it makes it up, and therefore how reliable are these tests?
Ideally we need to get these tests working with the real boot process in some capacity
We would then need to make sure we have all the tests in e2e-frontend written in this style
refs https://github.com/TryGhost/Team/issues/1104
- bumped `@tryghost/custom-theme-settings-service` so it throws a more appropriate `ValidationError` when setting keys don't exist or a select value is not known
- changed the custom theme settings service to have a `.init()` method which creates an instance of the service under `.api` so that we're able to create the instance at a particular point in the boot process when we know the models have been initialised
- there were problems in tests because the service was being initialised through the require chain before models were initialised through the boot process
- fixed incorrect `camelCase` of resource name in API responses
