Cache reference resolution for external references as finding
the declaration of a symbol is expensive and does not change
for a program once created.
This resolves a signficant performance regression in the langauge
service.
PR Close#21359
This helps ensure we use the same tsconfig.json file for all compilations.
Next steps are to make it the same tsconfig.json file used by the editor
PR Close#20964
When the SW fetches URLs listed in a manifest with hashes, it checks
the content hash against the manifest to make sure it has the correct
version of the URL. In the event of a mismatch, the SW is supposed to
consider the manifest invalid, and avoid using it. There are 3 cases
to consider by which this can happen.
Case 1: during the initial SW installation, a manifest is activated
without waiting for every URL to be fully loaded. In the background,
every prefetch URL listed by the manifest is requested and cached.
One such prefetch request could fail the hash test, and cause the
manifest to be treated as invalid. In such a case, the SW should
enter a state of EXISTING_CLIENTS_ONLY, as the latest manifest is
invalid.
This case works today.
Case 2: during the initial SW installation, as in Case 1, a manifest
is activated without waiting for each URL to fully load. However,
it's possible that the application could request a URL with a bad
hash before background initialization tries to load that URL. This
happens if, for example, the application has a broken index.html.
In this case, the SW should enter a state of EXISTING_CLIENTS_ONLY,
and serve the request from the network instead.
What happens today is that the internal error escapes the SW and
is returned as a rejected Promise to respondWith(), causing a
browser-level error that the site cannot be loaded, breaking the
site.
This change allows the SW to detect the error and enter the correct
state, falling back on the network if needed.
Case 3: during checkForUpdate(), the SW will try to fully cache the
new update before making it the latest version. Failure here is
complicated - if the page fails to load due to transient network
conditions (timeouts, 500s, etc), then it makes sense to continue
serving the existing cached version, and attempt to activate the
update on the next cycle.
If the page fails due to non-transient conditions though (400 error,
hash mismatch, etc), then the SW should consider the updated
manifest invalid, and enter a state of EXISTING_CLIENTS_ONLY.
Currently, all errors are treated as transient.
This change causes the SW to treat all errors during updates as
non-transient, which can cause the SW to unnecessarily enter a
safe mode. A future change can allow the SW to remain in normal mode
if the error is provably transient.
PR Close#21288
Chrome 63 can cause the navigationStart event for the first
run to arrive with a different pid than the start of the
benchpress run. This makes the first collected result invalid.
This workaround causes the sampler to ignore runs that have this
condition.
PR Close#21396
Bazel runs on newer version of RxJs than is installed in Yarn. The never version subclasses `EmptyError` in a different way which fails the `instanceof` check. This change makes the `instanceof` check more robust with respect to `EmptyError`.
PR Close#21053
- Add tests target for `test`, `test_node_only` and `test_web` in `core` package.
- Created a `_testing_init` pseudo package where bootstrap code for tests is kept.
- Moved `source_map_util` from `test` to `testing` so to prevent circular dependency.
- Removed `visibility:public` for testing `BUILD` packages.
PR Close#21053
`main()` function used to be needed to support dart, since dart
Does not allow top level statements. Since we no longer use dart
The need for `main()` has been removed.
In preparation for `Basel` and standardized way of running tests
we are removing `main()`
PR Close#21053
This patch fixes animations so that if multiple sub @triggers are used
and are blocked by a parent animation then the engine will not lead
itself into an infinite loop.
PR Close#21119
This PR fixes a circular dependency among those files in Renderer3:
`query` -> `di` -> `instructions` -> `query` -> ...
Looking at the above dependencies the `di` -> `instructions` import is
a problematic one. Previously `di` had an import from `instructions`
since we can known about "current node" only in `instructions`
(and we need "current node" to create node injector instances).
This commit refactors the code in the way that functions in the
`di` file don't depend on any info stored module-global variables
in `instructions`.
PR Close#20855
Closure Compiler renames all properties that are "internal" to the
program. `DOMAnimation` however is external, it is a browser API, so its
fields must not be renamed.
This change marks `DOMAnimation` as external using `declare interface`,
which will cause Closure Compiler to back off and prevent renaming of
any of its fields.
PR Close#21125
Previously, the router would merge path and matrix params, as well as
data/resolve, with special rules (only merging down when the route has
an empty path, or is component-less). This change adds an extra option
"paramsInheritanceStrategy" which, when set to 'always', makes child
routes unconditionally inherit params from parent routes.
Closes#20572.
The compiler host would force any file that is in node_modules
into the list of files that needed to be type checked which
captures .js files if `allowJs` is set to `true`. This should
have only forced .d.ts files into the project to enable
generation of factories.
Fixes: #19757
This allows examples to be found during aio's `yarn serve-and-sync`, which only
looks for examples in `packages/examples/<packageName>/**/*`, where
`packageName` is the name of the package that the modified file belonged to;
e.g. `core`, `common`, etc.).
Allows a directive to use the expression passed directly to a property
as a guard instead of filtering the type through a type expression.
This more accurately matches the intent of the ngIf usage of its template
enabling better type inference.
Moved NgIf to using this type of guard instead of a function guard.
Closes: #20967
Currently the Service Worker checks for updates only on SW startup,
an event which happens frequently but also nondeterministically. This
makes it hard for developers to observe the update process or reason
about how updates will be delivered to users. This problem is
exacerbated by the DevTools behavior of keeping the SW alive
indefinitely while opened, effectively preventing the page from
updating at all.
This change causes the SW to additionally check for updates on
navigation requests (app page reloads). This creates deterministic
update behavior, and is much easier for developers to reason about.
It does leave the old update-on-SW-startup behavior in place, as
removing that would be a breaking change.
Fixes#20877
Closure Compiler cannot infer that the swtich statement is exhaustive,
which causes it to complain that the method does not always return a
value.
Work around the problem by throwing an exception in the default case,
and using the `: never` type to ensure the code is unreachable.
Due to an overly agressive assert the compiler would generate
an internal error when referencing an enum declared in
namspace.
Fixes#18170
PR Close#20947
Previously, this code would unconditionally add a @fileoverview
comment to generated files, and only if the contained any code at all.
However often existing fileoverview comments should be copied from the
file the generated file was originally based off of. This allows users
to e.g. include Closure Compiler directives in their original
`component.ts` file, which will then automaticallly also apply to code
generated from it.
This special cases `@license` comments, as Closure disregards directives
in comments containing `@license`.
PR Close#20870
Since our version of Chromium is also pinned, a new ChromeDriver (that
drops support for our Chromium version) can cause random (and unrelated
to the corresponding changes) errors on CI.
This commit pins the version of ChromeDriver and it should now be
manually upgraded to a vrsion that is compatible with th currently used
Chromium version.
PR Close#20940
`$any()` can now be used in a binding expression to disable type
checking for the rest of the expression. This similar to `as any` in
TypeScript and allows expression that work at runtime but do not
type-check.
PR Close#20876
Structural directives can now specify a type guard that describes
what types can be inferred for an input expression inside the
directive's template.
NgIf was modified to declare an input guard on ngIf.
After this change, `fullTemplateTypeCheck` will infer that
usage of `ngIf` expression inside it's template is truthy.
For example, if a component has a property `person?: Person`
and a template of `<div *ngIf="person"> {{person.name}} </div>`
the compiler will no longer report that `person` might be null or
undefined.
The template compiler will generate code similar to,
```
if (NgIf.ngIfTypeGuard(instance.person)) {
instance.person.name
}
```
to validate the template's use of the interpolation expression.
Calling the type guard in this fashion allows TypeScript to infer
that `person` is non-null.
Fixes: #19756?
PR Close#20702
For some reason, prior to this fix, the boolean set matching
code (within `animation_transition_expr.ts`) failed to remain
the same when compiled with closure. This refactor makes sure
that the code stays in tact.
Reproduction Details:
Passes without `ng build --prod`: https://burger.stackblitz.io/
Fails with `ng build --prod`: http://burger.fxck.cz/Closes#20374
PR Close#20725
Closure Compiler by default will report diagnostics from type checks in
any JavaScript code, including code emitted by the Angular compiler.
Disabling `checkTypes` substantially reduces warning spam for users, and
allows them to run with stricter compiler flags (e.g. treating actual
diagnostics from user code as errors).
Closure Compiler will still type check the code and use types (where
found and correct) for optimizations.
PR Close#20828
Add enough BUILD files to make it possible to
`bazel build packages/core/test`
Also re-format BUILD.bazel files with Buildifier.
Add a CI lint check that they stay formatted.
PR Close#20768
The package.json esm2015 points to the wrong path.
"esm15" should be "esm2015"
Service Worker can't be compiled with use of Closure Compiler
PR Close#20800
Not every application is served from the domain root. The Service
Worker made a bad assumption that it would be, and so requested
/ngsw.json from the domain root.
This change corrects this assumption, and requests ngsw.json without
the leading slash. This causes the request to be interpreted
relative to the SW origin, which will be the application root.
The Service Worker contains a mechanism by which it will postMessage
itself a signal to initialize its caches. Through this mechanism,
initialization happens asynchronously while keeping the SW process
alive.
Unfortunately in Firefox, the SW does not have the ability to
postMessage itself during the activation event. This prevents the
above mechanism from working, and the SW initializes on the next
fetch event, which is often too late.
Therefore, this change has the application wait for SW changes and
tells each new SW to initialize itself. This happens in addition to
the self-signal that the SW attempts to send (as self-signaling is
more reliable). That way even on browsers such as Firefox,
initialization happens eagerly.
Currently a bug exists where attempting to inject SwPush crashes the
application if Service Workers are unsupported. This happens because
SwPush doesn't properly detect that navigator.serviceWorker isn't
set.
This change ensures that all passive observation of SwPush and
SwUpdate doesn't cause crashes, and that calling methods to perform
actions on them results in rejected Promises. It's up to applications
to detect when those services are not available, and refrain from
attempting to use them.
To that end, this change also adds an `isSupported` getter to both
services, so users don't have to rely on feature detection directly
with browser APIs. Currently this simply detects whether the SW API
is present, but in the future it will be expanded to detect whether
a particular browser supports specific APIs (such as push
notifications, for example).
Currently, the way to not use the SW is to not install its module.
However, this means that you can't inject any of its services.
This change adds a ServiceWorkerModule.disabled() MWP, that still
registers all of the right providers but acts as if the browser does
not support Service Workers.
Saving `oldProgram` in `AngularCompilerProgram` instances is causing a memory leak for unemitted programs.
It's not actually used so simply not saving it fixes the memory leak.
Fix#20691
PR Close#20692
This changes XhrBackend to not strip the XSSI prefix from error text
if such a prefix is present but the remaining body does not parse as
JSON.
PR Close#19958
Previously, XhrBackend would call JSON.parse('') if the response body was
empty (a 200 status code with content-length 0). This changes the XhrBackend
to attempt the JSON parse only if the response body is non-empty. Otherwise,
the body is left as null.
Fixes#18680.
Fixes#19413.
Fixes#19502.
Fixes#19555.
PR Close#19958
Previously, HttpClient used the overly clever test "body || null"
to determine when a body parameter was provided. This breaks when
the valid bodies '0' or 'false' are provided.
This change tests directly against 'undefined' to detect the presence
of the body parameter, and thus correctly allows falsy values through.
Fixes#19825.
Fixes#19195.
PR Close#19958
The errors produced when error were encountered while interpreting the
content of a directive was often incomprehencible. With this change
these kind of error messages should be easier to understand and diagnose.
PR Close#20459
The type-check block generated with `"fullTemplateTypeCheck"` was
invalid if the it contained a template ref as would be generated
using the `else` micro-syntax of `NgIf`.
Fixes: #19485
PR Close#20463