- the integrationTesting utils are specific to the express mock style of testing
- all other tests can use the url-service-utils to check the url service is finished
- 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However it's really old and out of date because of the boot refactor and several changes since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
- this test file uses a different pattern to the other test files
- not yet sure if the pattern is terrible or genius, need to assess before moving it into a folder full of what are meant to be exemplary tests
closes https://github.com/TryGhost/Team/issues/1125
refs 3c822e0457
- Email-only is not considered a general availability feature and can be used without special flags.
- It allows to publish a new post type "email only" that only goes out as an email newletter and is available through an undescoverable URL (does not appear anywhere publicly similarly to preview posts) on the site.
- e2e tests are tests that cover critical functionality by booting ghost
- integration tests are more like unit tests, but need to initialise and use a db
- so settings shouldn't start Ghost, url service is critical and should be in integration, and preview is critical and should be in e2e
- We have a bunch of important server-related e2e tests
- Make these clear in their own folder
- "server" is everything that isn't the api or the frontend - kind of a catch-all concept
refs https://linear.app/tryghost/issue/CORE-35/refactor-route-and-redirect-settings
- Frontend is not meant to know about the underlying source of the "routes" configuration, so any reads/edits/validations are being moved into a backend service. This should also simplify the coupling of the backend with the frontend where the latter will get a JSON blob with all needed configuration during the boot
- Nother problem the "get" method had was hiding an underlying function it was doing - reading the file from the filesystem SYNCRONOUSLY. It might be a thing we need to do during the "web" app initialization, but there's no clear need to do this in a sync fassion during the bootup for example. Also having a more explicit name should help :)
refs https://github.com/TryGhost/Team/issues/990
- Relying on uuid instead of slug makes the posts less discoverable and partially soves discoverability through overriden robots.txt files
refs https://github.com/TryGhost/Team/issues/953
- Emails posts should be not explorable by the rest of the frontend similarly to the draft or scheduled posts. Email posts should also keep the content gating, so that specific parts of content can still be gated based on the post's visibility setup
- A separate frontend router was chosen to implement this part of the system instead of a moutable express app due to increased complexity to introduce the latter approach.
- All "sent" email-only posts will be accessible through the `/email/:slug/` route
refs https://github.com/TryGhost/Team/issues/948
- When working on related problem noticed these declarations were scattered all over the place in the test without special purpose. Moved them all up into a central "once per suite" stub. Not sure if this subing is working at all, but at least it will be easier to change it once per suite in the future once someone spots a problem with this approach
refs https://github.com/TryGhost/Team/issues/948
- While working on referenced issue came around adding tests to new top level route - '/email/' and it was pretty hard to navigate through single blob of tests. Broke them down by the area/route into separate suites for better mainteinability
This commit achieves a few things:
- ☑️ No longer having to remember whether a command is yarn something or grunt something
- ☑️ Simplification of tools hopefully making them easier to remember and use
- ☑️ Complete removal of the need for grunt from our test tooling
Several of the tools still use grunt under the hood, but the **entrypoint** should aways be `yarn xxx`.
- `grunt main` -> `yarn main`
- `grunt dev` -> `yarn dev`
- `grunt build` -> `yarn build`
- `grunt test:file-or-folder` -> `yarn test file-or-folder`
- `grunt test-unit` -> `yarn test:unit`
- `grunt test-acceptance` -> `yarn test:acceptance`
- `grunt test-regression` -> `yarn test:regression`
- `grunt validate` -> removed due to lack of use
There is now also `yarn test:all` to run all 3 classes of tests
This PR also reorders & restructures the Gruntfile extensively so that:
- The remaining useful commands are all at the top of the file
- Config and other blah happens after all the useful commands
- All release-only config happens in the release task at the very end of the file
---
DONE:
* Removed all references to npm/bower
* Removed all references to lint / deprecated command
* Moved debug to yarn dev:debug
* Removed all references to travis
* Removed broken help task + useless comment
* Removed unused knex-migrator and clean:test setup tasks
* Added new test commands, removed grunt validate
* Moved stubClientFiles to test utility and use in the few tests that need it
* Used mocha in yarn directly except grunt test:x
* Swapped grunt test for yarn test
* extensive cleanup and reshuffling
- This is part of the quest to separate the frontend and server & get rid of all the places where there are cross-requires
- At the moment the settings cache is one big shared cache used by the frontend and server liberally
- This change doesn't really solve the fundamental problems, as we still depend on events, and requires from inside frontend
- However it allows us to control the misuse slightly better by getting rid of restricted requires and turning on that eslint ruleset
- These tests protect some advanced, yet important behaviour around URLs
- Moved them out of regression tests into acceptance, as we're slowly trying to clean up regression tests :)
- Checking for differences in behaviour when urls are configured with or without slashes should fall to unit testing
- Generally testing that a correctly configured subdirectory site does what's expected is an acceptance test
- Ghost's 301 to trailing slash behaviour is also well tested, and doesn't need extensive acceptance tests when using a subdir
- Instead, test that the non-subdirectory equivalent route is 404ing as expected, to check that routes are ONLY mounted on the subdir, as well as mounted at all
refs: f1be3418d9
- Since I refactored the boot process & subsequently the test tools, I have been hitting an issue where subdirectory tests don't work if the app has already been started without a subdirectory
- Turns out it's due to the rootApp getting cached, and not replaced no matter what you do, even though if you debug it _looks_ like it's been replaced
- This makes sense as the rootApp is in a separate file and therefore subject to the node module cache and we're using const everywhere
- Therefore, I have added a single line to the test utils to destroy this cache for this file, and the tests now work perfectly
- Next: refactor the test utilities some more to make all this much, much more straightforward and easy to understand
- Having these as destructured from the same package is hindering refactoring now
- Events should really only ever be used server-side
- i18n should be a shared module for now so it can be used everywhere until we figure out something better
- Having them seperate also allows us to lint them properly
refs: bf0823c9a2
- continuing the work of splitting up the theme service into logical components
- this is where it starts to get fiddly as the getActive function in themeService index is required across the frontend/backend mostly due to its use in the getApiVersion method
- for now left one usage of the getActive method in place in ghost-locals middleware ready for the next phase of the refactor, which will move some of the themeService index into a shared location
refs 0f59537b96
refs https://github.com/TryGhost/Team/issues/221
refs 303046bc0a
- When the referenced changes were introduced they did not take into account upcoming engine versions and provided little guidance about what other areas might need to be checked and changed
- The last referenced commit shows an approximate scale of changes that might be beened when frontend engine defaults are modified in the future
- The snake case pattern is used in all the rest of test suite file names
- Cleaned up these two files as they were sticking out in the folder I was touching
refs https://github.com/TryGhost/Ghost/commit/abeabb764
- Continued work started in referenced commit in making sites_spec more manageable.
- Extracted out separate v2/v3/canary specs for each site configuration
- Now these files are easily diff'able and introducing new API version would include copying a file over instead of adding up to a gigantic file
refs https://github.com/TryGhost/Ghost/commit/2f20930ff
- The aim is to split site test suite into separate files v2/v3/v4 after this move
- It is hard to grep or diff the file of this size.
- Having separate files per ghost-api engine will allow tracking changes between different API versions in the future
- It's a continuation of work started in refed commit
no issue
- The site_spec is a catch all suite with too many moving parts. While adding a new tests suite for v4 frontend configuration it felt overwhelming to add hundreds of lines to a >5000 line monstrocity
refs dd715b33dc
- this is the last event that is used to trigger part of the standard boot process
- events make the code harder to read/reason about
- the urlservice is one of the most core and critical components in Ghost, possibly the biggest consumer of time and memory
- we want to have the work it is doing front and center so that we can improve it
- We currently have 2 boot systems in the codebase, until now the tests were still running against the old one
- Regression tests using the forceStart + subdir flag no longer work. These need reworking, but this _should_ be easier later
- Note those tests work fine if they were the first tests run, it is only the case of trying to restart the server with a subdirectory after starting without that doesn't work
- There is now just one way to wait until the UrlService is finished across all of our tests!
- db.ready is only called in the one utility + the actual UrlService tests
closes#11085
- Ghost has been using YAML format for other configurations (e.g. routes). The plan is to move to this format for all user-edited settings files. By default JSON format is still used in Ghost Admin API v2/v3, but will be changed to YAML in API v4. Check referenced issue for more context.
- New format supports all the features available before. The main noticeable change is the structure of config file. It is now grouped by redirect HTTP code instead of specifying `"permanent": true | false` attribute for each config property. Example format for YAML config:
```
302:
/from-url/: /to-url/
301:
/category/([a-z0-9\-]+)/i: /tag/$1/
/v([0-9\.]+)/docs/([a-z0-9\-]+)/i: /docs/$2/
```
- Added 2 new endpoints: `POST redirects/upload` and `GET redirects/download`. These serve as an alias to current GET/POST `/redirects/json. "upload/download" naming pattern is introduced to match the convention with other resources that can be uploaded and downloaded (images, themes etc.). `/redirects/json` endpoints will be removed in Admin API v4
- The parsing code from `custom-redirects.js` has been moved to `frontend/services/redirects/settings.js`. This location is more appropriate for this logic and eventually `custom-redirects.js` middlewear might be moved into "frontend" as this middlewear plays a role mostly effecting that area.
closes https://github.com/TryGhost/Ghost/issues/12247
- Internal preview controller was lacking "mapping" call to post object which handled not only missing meta attribute information but lots of other mappings (e.g. users, tags, etc.)
- Have added a regression test to catch issues like this in the future
closes#11999
- When the routes.yaml file changes (manually or through API) we need
to store a checksum to be able to optimize routes reloads in the future
- Added mechanism to detect differences between stored and current routes.yaml hash value
- Added routes.yaml sync on server boot
- Added routes.yaml handling in controllers
- Added routes hash synchronization method in core settings. It lives in core settings
as it needs access to model layer. To avoid coupling with the frontend settings it accepts
a function which has to resolve to a routes hash
- Added note about settings validation side-effect. It mutates input!
- Added async check for currently loaded routes hash
- Extended frontend settings loader with async loader. The default behavior of the loader is
to load settings syncronously for reasons spelled in 0ac19dcf84
To avoid blocking the eventloop added async loading method
- Refactored frontend setting loader for reusability of settings file path
- Added integrity check test for routes.yaml file
closes#11932
- as per the issue, there is no need to output a <link> tag if the favicon is the default /favicon.ico, as all browsers automatically check for this
- instead the favicon <link> is only output if a custom favicon has been set, telling the browser to look somewhere different to its default location
- some of the tests expected 3 links in the html head, but through the favicon change there are often only 2.
Co-authored-by: RenCloud <rencloud@pop-os.localdomain>
no-issue
This issue only occurs when using custom redirects with a subdirectory
setup, and the path to be redirected from is expressed as a regex, and
the url that is being redirected to is not an external url.
The issue has a few components:
- Redirect paths as a regex generally use the ^ to ensure that they
match the beginning of the path.
- The path that the regex is matched against conditionally excludes the
subdirectory, specifically, the subdirectory is excluded for external
urls
These combined means you end up with a regex like /^\/custom-redirect/
and a path like /subdir/custom-redirect, these will not match/replace
correctly, and you'll end in an infinite redirect loop.
The fix here is to *always* remove the subdirectory when testing regex's
and then conditionally adding it back *only* for the redirect, and only
if it is an internal redirect
- Ghost Admin has a redirect system built in, so not using the # does work
- However, the usage of # or no # was inconsistent, which is always wrong
- Using the correct Ghost Admin URLs makes it easier to reason about the behaviour
* moved `server/config` to `shared/config`
* updated config import paths in server to use shared
* updated config import paths in frontend to use shared
* updated config import paths in test to use shared
* updated config import paths in root to use shared
* trigger regression tests
* of course the rebase broke tests
- this redirect is the only one in our tests which is permanent
- also I verified that this new test works the same before and after the recent changes :)
refs #10898
- Execute string replacement on external paths
- Take non-top-level base URLs into consideration (to avoid #10776 dups)
- Added tests for all of the above cases
- All var declarations are now const or let as per ES6
- All comma-separated lists / chained declarations are now one declaration per line
- This is for clarity/readability but also made running the var-to-const/let switch smoother
- ESLint rules updated to match
How this was done:
- npm install -g jscodeshift
- git clone https://github.com/cpojer/js-codemod.git
- git clone git@github.com:TryGhost/Ghost.git shallow-ghost
- cd shallow-ghost
- jscodeshift -t ../js-codemod/transforms/unchain-variables.js . -v=2
- jscodeshift -t ../js-codemod/transforms/no-vars.js . -v=2
- yarn
- yarn test
- yarn lint / fix various lint errors (almost all indent) by opening files and saving in vscode
- grunt test-regression
- sorted!
- Have a consistent structure so we can co-locate an app with its own middleware
- This is another small step in sorting out the giant mess that was web/shared/middleware
- move all test files from core/test to test/
- updated all imports and other references
- all code inside of core/ is then application code
- tests are correctly at the root level
- consistent with other repos/projects
Co-authored-by: Kevin Ansfield <kevin@lookingsideways.co.uk>
