Lighthouse v1.6.5 treats localhost/1.2.7.0.0.1 as secure domains (i.e. as if they where HTTPS), so we need to stop handling the is-on-https audit specially.
In some cases (unclear when), traceviewer-js, used by Lighthouse under the hood,
assumes `atob`/`btoa` are defined in the global scope. This is true for browser
environments, but not on node.
As a result, some aggregations that required access to model-tracing failed to
produce results, dropping the overall PWA score.
This affected #16665 (e.g. commit 0de6eec7a).
This fixes a flicker when transitioning from server rendered page to client rendered page in lazy loaded routes by waiting for the lazy loaded route to finish loading, assuming initialNavigation on the route is set to 'enabled'.
Fixes#15716
If a usage of `{@link ...}` does not provide a title then
compute it based on the `title` and/or `name` properties
or set the link to invalid.
Closes#16811
Sidenav headers had been anchors w/o hrefs. These can’t take focus which makes you can’t navigate through them with keyboard. For a11y purposes, this PR turns them into buttons.
By reflex we began all component tests with an async `beforeEach` that called `compileComponents`.
In at least one case (`live-example.component.spec.ts`) that led to the `it` tests being async as well.
There is no need to call `.compileComponents` because CLI web pack + plugin inlines all templates and styles.
While `.compileComponents` was harmless, it added complexity and distraction which we should not inflict on future readers and testers.
Previously the logic for deciding when to display
the search result was spread between different
parts of the application and used non-intuitive logic
such as sending a blank results set to the searchResults.
This commit moves the management of displaying
the search results (and also setting focus of the
search input box) to the AppComponent. This makes
it easier to understand what happens and why; but
also allows the search UI components to be more
easily reused (such as embedding them in the 404
page).
(Coincidentally), this wasn't an issue before fdfeaaf1f, because
pre-verification was run after `test.sh`, during which `aio-builds-setup` was
built.
Now that `deploy-staging.sh` is being run before `test.sh`, we need to build
the `aio-builds-setup` scripts first.
Previously, `aio/aio-builds-setup/scripts/travis-preverify-pr.sh` was supposed
to exit with 1 if a PR did not meet the preconditions and 2 if an error occurred
during pre-verification.
It relied on the exit codes of the node script that did the actual work, but
didn't account for errors that would be thrown in the `sh` script itself (e.g.
if the node script was not available). This caused such errors to appear as
non-verified PRs, instead of real errors that should fail the build.
This commit swaps the exit codes, so that now a 2 means non-verified PR and 1
designates an error.
Hybrid apps (mix of Angular and AngularJS) might return AngularJS implementation
of Promises that do not play well with the change detection. Wrapping them in
native Promises fix this issue.
This could be the case when a Resolver returns a `$q` promise.
Use bracket notation to access $inject in downgradeInjectable to
support property renaming. Since the return type is any,
Closure compiler renames $inject.
Since abb36e3cb, we no longer rely on the cli to set up ServiceWorker, but do it
manually as part of `yarn build`. When using `ng serve`, registering the
ServiceWorker fails, because we haven't created `ngsw-manifest.json` nor copied
`worker-basic.min.js` into dist.
This commit works around this, by only registering the service worker in
production mode (which is what the cli does too).
Caveat:
It is not possible to enable ServiceWorker with `ng serve`/`yarn start` and
using the `--prod` flag will try to register it, but fail because the necessary
files (`ngsw-manifest.json` and `worker-basic.min.js`) will not be available.
(As a work-around, you can use `yarn build` and serve the files in `dist/` with
`yarn http-server -- dist -p 4200`.)