The `lib/startup`-addon is used to create a custom
index.html page to boot up the app. Because this is
a custom approach and we aren't relying on setting
rootURL to a different value when running the app
in `production` we need to set `publicAssetURL`
in the configuration we pass to `ember-auto-import`.
Reference:
9a2887f1d0/docs/upgrade-guide-2.0.md (quick-summary)
* Use postcss instead of ember-cli-sass
This will make it possible to work with tailwindcss.
* configure postcss to compile sass
* add "sub-app" css into app/styles tree
* pin node@14 via volta
Only used by people that use volta
* Install tailwind and autoprefixer
* Create tailwind config
* Use tailwind via postcss
* Fix: tailwind changes current styling
When adding tailwind to the bottom of app.scss we apparently
change the way the application looks. We will import
it first to make sure we don't change the current styling
of the application right now.
* Automatic import of HDS colors in tailwind
* Install @hashicorp/design-system-components
* install add-on
* setup postcss scss pipeline to include tokens css
* import add-on css
* Install ember-auto-import v2
HDS depends on v2 of ember-auto-import so we need to upgrade.
* Upgrade ember-cli-yadda
v0.6.0 of ember-cli-yadda adds configuration for webpack.
This configuration is incompatible with webpack v5
which ember-auto-import v2 is using.
We need to upgrade ember-cli-yadda to the latest
version that fixes this incompatability with auto-import v2
* Install ember-flight-icons
HDS components are using the addon internally.
* Document HDS usage in engineering docs
* Upgrade ember-cli-api-double
* fix new linting errors
* v3.20.2...v3.24.0
* Fix handle undefined outlet in route component
* Don't use template helper for optional modal.open
Using the optional-helper here will trigger a computation
in the same runloop error. This is because we are setting
the `modal`-property when the `<Ref>` component gets
rendered which will update the `this.modal`-property which
will then recompute the `optional`-helper leading to this
error.
Instead we will create an action that will call the `open`-method
on the modal when it is defined. This gets rid of the double
computation error as we will not access the modal property
twice in the same runloop when `modal` is getting set.
* Fix - fn needs to be passed function tab-nav
We create functions in the component file instead
so that fn-helper stops complaining about the
need to pass a function.
* Update ember-exam to 6.1 version
"Makes it compatible" with ember-qunit v5
* scheduleOnce setMaxHeight paged-collection
We need to schedule to get around double-computation error.
* Fix - model.data is removed from ember-data
This has been private API all along - we need to
work around the removal.
Reference: https://github.com/emberjs/data/pull/7338/files#diff-9a8746fc5c86fd57e6122f00fef3155f76f0f3003a24b53fb7c4621d95dcd9bfL1310
* Fix `propContains` instead of `deepEqual` policy
Recent model.data works differently than iterating attributes.
We use `propContains` instead of `deepEqual`. We are only
interested in the properties we assert against and match
the previous behavior with this change.
* Fix `propContains` instead of `deepEqual` token
* Better handling single-records repo test-helper
`model.data` has been removed we need to handle proxies and
model instances differently.
* Fix remaining repository tests with propContains
We don't want to match entire objects - we don't care
about properties we haven't defined in the assertion.
* Don't use template helper for optional modal.open
Using a template helper will give us a recomputation error -
we work around it by creating an explicit action on
the component instead.
* Await `I $verb the $pageObject object` step
* Fix no more customization ember-can
No need to customize, the helper handles destruction
fine on its own.
* Fix - don't pass `optional` functions to fn
We will declare the functions on the component instead.
This gives us the same behavior but no error from
`fn`, which expects a function to be passed.
* Fix - handle `undefined` state on validate modifier
StateChart can yield out an undefined `state` we need
to handle that in the validate modifier
* Fix linting errors tests directory
* Warn / turn off new ember linting issues
We will tackle them one by one and don't want to
autofix issues that could be dangerous to auto-fix.
* Auto-fix linting issues
* More linting configuration
* Fix remaining linting issues
* Fix linting issues new files after rebase
* ui: Remove ember-cli-uglify config now we are using terser (#14574)
Co-authored-by: John Cowen <johncowen@users.noreply.github.com>
This commit moves our in-app LockSessions code into an external 'app', which can theoretically be side-loaded but for now it just makes for good isolation/code hygiene.
Functionally, there is kind of one change here, and that is we only show the 'Lock Session' tab if you have permissions to see them. Currently as our UI authorization endpoint needs to be changed slightly to suit our usecase, you will always have permissions to see Lock Sessions as we hardcode the session:read to true (obvs this is a frontend thing, not a backend thing), so it doesn't really change anything from a user perspective.
Also added very bare docs while I was here.
Small note here, ideally we need to add the each individual tab depending on whether an 'app' is enabled or not instead of just permissions, ideally it would be done totally from The Outside rather than a can based conditional on the inside, just something else to be thinking about for the future.
We need a way to load certain CSS based on the environment you are viewing, i.e. we have debug CSS that we use for our Eng Documentation and various other DX utilities that shouldn't be compiled into our production or test builds.
Previously we would compile two entirely different CSS files (app and debug) and the load one or the other depending on which environment you were in.
This approach just empties out the debug.css file in certain environments (prod/test) which means we can just import that file from app. When in staging/development this imports the contents of debug.css (quite a bit of CSS) whereas when building for production/test this debug.css is emptied out during the build process.
There is a slight little hack in order to have this work, we import _debug.scss which imports the debug.scss file. I couldn't for the life of me figure out how to have broccoli empty out a file during the build process, so instead we essentially copy over debug.scss during dev and create an empty file during prod to _debug.scss.
When using make build to build an artifact for production CSS remains at ~58kb (during dev its a lot bigger than this)
- Simplifies how we 'import' our configuration files a little in order to make them more grokable.
- Starts to exclude files based on explicit configuration rather than convention.
- Adds the first instance of us being able to select an implementation (of multiple) of a component at runtime.
This sounds a bit 'backwards' as the end goal here is to add an improved UX to partitions, not namespaces. The reason for doing it this way is that Namespaces already has a type of 'improved UX' CRUD in that it has one to many relationship in the form when saving your namespaces (the end goal for partitions). In moving Namespaces to use the same approach as partitions we:
- Ensure the new approach works with one-to-many forms.
- Test the new approach without writing a single test (we already have a bunch of tests for namespaces which are now testing the approach used by both namespaces and partitions)
Additionally:
- Fixes issue with missing default nspace in the nspace selector
- In doing when checking to see that things where consistent between the two, I found a few little minor problems with the Admin Partition CRUD so fixed those up here also.
- Removed the old style Nspace notifications
This commit tries to make the development experience of working on our OIDC support a little more realistic, essentially by creating our own OIDC provider in our application (only during development builds). You can still provide a real OIDC provider to work with via our dev time environment/cookie variables as before, just now we default to the behaviour in this commit. Overall this makes it much easier to verify our OIDC support in the UI, and also opens up avenues for us to be able to test more scenarios that we couldn't before (for example not only successful logins, but also erroneous, potentially with multiple error reasons).
* Add `is` and `test` helpers in a similar vein to `can`
Adds 2 new helpers in a similar vein to ember-cans can:
- `is` allows you to use vocab/phrases such as (is "something model") which calls isSomething() on the models ability.
- `test` allows you to use vocab/phrases such as (test "is something model") or (test "can something model")which calls isSomething() / canSomething() on the models ability. Mostly using the is helper and the can helper. It's basically the is/can helper combined.
* Adds TextInput component + related modifiers/helpers/machines/services (#11189)
Adds a few new components/modifiers/helpers to aid building forms.
- state-chart helper, used in lieu of a more generic approach for requiring our statecharts.
- A few modifications to our existing disabled modifier.
- A new 'validation' modifier, a super small form validation approach built to make use of state charts (optionally). Eventually we should be able to replace our current validation approach (ember-changeset-validations + extra deps) with this.
- A new TextInput component, which is the first of our new components specifically to make it easy to build forms with validations. This is still a WIP, I left some comments in pointing out where this one would be progressed, but as we don't need the planned functionality yet, I left it where it was. All of this will be fleshed out more at a later date.
Documentation is included for all of ^
* ui: Adds initial CRUD for partitions (#11190)
Adds basic CRUD support for partitions. Engineering-wise probably the biggest takeaway here is that we needed to write very little javascript code to add this entire feature, and the little javascript we did need to write was very straightforwards. Everything is pretty much just HTML. Another note to make is that both ember-changeset and ember-data (model layer things) are now completely abstracted away from the view layer of the application.
New components:
- Consul::Partition::Form
- Consul::Partition::List
- Consul::Partition::Notifications
- Consul::Partition::SearchBar
- Consul::Partition::Selector
See additional documentation here for more details
New Route templates:
- index.hbs partition listing/searching/filtering
- edit.hbs partition editing and creation
Additionally:
There is some additional debug work here for better observability and to prevent any errors regarding our href-to usage when a dc is not available in our documentation site.
Our softDelete functionality has been DRYed out a little to be used across two repos.
isLinkable was removed from our ListCollection component for lists like upstream and service listing, and instead use our new is helper from within the ListCollection, meaning we've added a few more lighterweight templateOnly components.
* ui: Exclude all debug-like files from the build (#11211)
This PR adds **/*-debug.* to our test/prod excluded files (realised I needed to add test-support.js also so added that here as its more or less the same thing). Conditionally juggling ES6 static imports (specifically debug ones) for this was also getting a little hairy, so I moved it all to use the same approach as our conditional routes. All in all it brings the vendor build back down to ~430kb gzipped.
* ui: Fix text search for upstream instances
* Clean up predicates for other model types
* Add some docs around DataCollection and searching
* Enable UI Engineering Docs for our preview sites
* Use debug CSS in dev and staging
* Add inline-code CSS component
* Add %inline-code to all the places where we need it
* Inject selected env variables into the translations file
* Add ingress gateway upstream 'host header' intro text
* Make sure we can use actual correct component casing for titles but still have nice consistent menu item casing in the side nav
* Configure ember-auto-import so we can use a stricter CSP
* Create a fake filesystem using JSON to avoid inline scripts in index
We used to have inline scripts in index.html in order to support embers
filepath fingerprinting and our configurable rootURL.
Instead of using inline scripts we use application/json plus a JSON blob
to create a fake filesystem JSON blob/hash/map to hold all of the
rootURL'ed fingerprinted file paths which we can then retrive later in
non-inline scripts.
We move our inlined polyfills script into the init.js external script,
and we move the CodeMirror syntax highlighting configuration inline
script into the main app itself - into the already existing CodeMirror
initializer (this has been moved so we can lookup a service located
document using ember's DI container)
* Set a strict-ish CSP policy during development
This PR removes storybook and adds docfy and uses docfy to render our existing README files.
This now means we can keep adding README documentation without committing any specific format or framework. If we eventually move to storybook then fine, or if we just want to remove docfy for whatever reason then fine - we will still have a full set of README files viewable via GitHub.
* Add templating to inject JSON into an application/json script tag
Plus an external script in order to pick it out and inject the values we
need injecting into ember's environment meta tag.
The UI still uses env style naming (CONSUL_*) but we uses the new style
JSON/golang props behind the scenes.
Co-authored-by: Paul Banks <banks@banksco.de>
* ui: Add the most basic workspace root in /ui
* We already have a LICENSE file in the repository root
* Change directory path in build scripts ui-v2 -> ui
* Make yarn install flags configurable from elsewhere
* Minimal workspace root makefile
* Call the new docker specific target
* Update yarn in the docker build image
* Reconfigure the netlify target and move to the higher makefile
* Move ui-v2 -> ui/packages/consul-ui
* Change repo root to refleect new folder structure
* Temporarily don't hoist consul-api-double
* Fixup CI configuration
* Fixup lint errors
* Fixup Netlify target
Michael Klein
John Cowen
James Tran