Add a test to ensure that optimized image loaders can be configured at any
level of the DI hierarchy, including component's node injector.
PR Close#47565
Previously the built-in image loaders for the optimized image directive
were tightly coupled to the preconnect checks infrastructure. This was
creating a problem when developers were trying to provide a loader on
a component level (or, more generally, deeper in the DI hierarchy):
- PreconnectLinkChecker is the application-level service, provided in root;
- it makes sense to provide loaders in different parts of the DI hierarchy;
This refactoring removes the PreconnectLinkChecker configuration from
the loaders infrastructure and makes it application-wide config. The
PRECONNECT_CHECK_BLOCKLIST is also a simple provider (it was a
multi-provider previously) which should make the overall configuration
easier.
PR Close#47565
Ensure that keyframes rules, defined within components with emulated
view encapsulation, are scoped to avoid collisions with keyframes in
other components.
This is achieved by renaming these keyframes to add a prefix that makes
them unique across the application.
In order to enable the handling of keyframes names defined as strings
the previous strategy of replacing quoted css content with `%QUOTED%`
(introduced in commit 7f689a2) has been removed and in its place now
only specific characters inside quotes are being replaced with
placeholder text (those are `;`, `:` and `,`, more can be added in
the future if the need arises).
Closes#33885
BREAKING CHANGE:
Keyframes names are now prefixed with the component's "scope name".
For example, the following keyframes rule in a component definition,
whose "scope name" is host-my-cmp:
@keyframes foo { ... }
will become:
@keyframes host-my-cmp_foo { ... }
Any TypeScript/JavaScript code which relied on the names of keyframes rules
will no longer match.
The recommended solutions in this case are to either:
- change the component's view encapsulation to the `None` or `ShadowDom`
- define keyframes rules in global stylesheets (e.g styles.css)
- define keyframes rules programmatically in code.
PR Close#42608
Adds the logic that will filter out unexposed inputs/outputs and apply the aliases that the author specified when writing the host directives.
PR Close#47536
Currently, the `RouterLink` and `RouterLinkWithHref` classes share a lot of common code (with some special logic around handling `<a>`-related scenarios). This commit unifies the logic of the mentioned directives by moving the necessary handling to the `RouterLink` directive and making it a parent one for the `RouterLinkWithHref` class (i.e. class RouterLinkWithHref extends RouterLink).
This is the first step in upcoming unification to merge both directives and just keep `RouterLink` one (it'd be done in followup PRs).
PR Close#47500
ActivatedRouteSnapshot data gets mutated in the resolve phase of the Router. The title is assigned as part of this.
As a result, the title must be a getter in order to pick up the value that was note available during the class creation.
fixes#47459
BREAKING CHANGE: The title property is now required on ActivatedRouteSnapshot
PR Close#47481
Currently the code that creates a root component assumes that it's always going to deal with a single component definition which won't work with host directives. These changes rework the code so that it's able to apply multiple directives, allowing us to eventually add support for host directives.
I also tried to make the root component creation easier to follow by breaking it up into smaller functions.
PR Close#47530
The Angular compiler will report the invalid banana in box, this code fixes
will try to fix the error and all the same errors in the selected file.
Fixes#44941
PR Close#47393
This commit applies the changes similar to the ones performed for the `inject()` function in df246bb235.
The `TestBed.inject` function is updated to use previously added object-based API for options: now the flags argument supports passing an object which configures injection flags.
DEPRECATED:
The bit field signature of `TestBed.inject()` has been deprecated, in favor of the new options object.
PR Close#46761
This commit applies the changes similar to the ones performed for the `inject()` function in df246bb235.
The `Injector.get` function is updated to use previously added object-based API for options: now the flags argument supports passing an object which configures injection flags.
DEPRECATED:
The bit field signature of `Injector.get()` has been deprecated, in favor of the new options object.
PR Close#46761
1. Remove `zone-async-tagging` implementation from zone.js and move the
implementation to `@angular/core`, so `@angular/core` can import this
package more easily for better treeshaking.
2. Add `async tagging zone` implemenation into `@angular/core` package.
So we don't need to get the `AsyncStackTaggingZoneSpec` from `global`
instance, we can import the `class` directly for better treeshaking.
3. Only load this ZoneSpec when `ngDevMode` is `true`.
PR Close#47416
This commits update `isDevMode` to rely on the `ngDevMode` which in the CLI is set by the bundler.
We also update `@angular/platform-dynamic-browser` and `@angular/compiler` to remove usage of `jitDevMode`, with this change we remove all internal usages of `isDevMode`.
PR Close#47475
Exposes the host directives to the host and its descendants through DI. This can be useful, because it allows the host to further configure the host directives.
PR Close#47476
When reporting type-checking diagnostics in external templates we create a
`ts.SourceFile` of the template text, as this is needed to report Angular
template diagnostics using TypeScript's diagnostics infrastructure. Each
reported diagnostic would create its own `ts.SourceFile`, resulting in
repeatedly parsing of the template text and potentially high memory usage
if the template is large and there are many diagnostics reported. This commit
caches the parsed template in the template mapping, such that all reported
diagnostics get to reuse the same `ts.SourceFile`.
Closes#47470
PR Close#47471
`TNode`s have the `directiveStart` and `directiveEnd` properties that indicate the indexes at which directive instances (including components) have been stored. Currently there are several places throughout the codebase which assume that if a component matches a node, its index will always be `directiveStart`.
As far as I can tell, we probably ended up accumulating these assumptions, because we needed a quick way of accessing the component instance and it happened to be conventiently stored at `directiveStart`. The reason why it's always stored at `directiveStart` is likely to match the lifecycle hook execution order from ViewEngine.
With host directives these assumptions won't be valid anymore, because we want the host directives to _always_ execute before the host component that they're on so that the host has a chance to override them. To achieve this we have to insert host directives before the component.
These changes address the issue by introducing a new `TNode.componentOffset` property which indicates the offset after `TNode.directiveStart` at which the component is stored. Furthermore, I've removed the `isComponentHost` flag since it was duplicating the information from `TNode.componentOffset` and I've audited and fixed all the places where we read `directiveStart` to account for the changed data structure.
Reasons for some of the decisions I made along the way:
* In the case of host directives, I decided to go against our current convention of executing the component lifecycle hooks before the directive, because lifeycle hooks are a chance to change the component state (e.g. in `ngOnChanges`) and running the component hooks first would allow the host directives to undo any overrides made by the host.
* I decided to go with a `componentOffset`, instead of a `componentIndex` indicating the exact index the component is at, because as the runtime is set up at the moment, it would be difficult to know what index the component is going to end up at. Another problem is that we appear to have some logic that moves the entire "directive window" by incrementing both `directiveStart` and `directiveEnd`. By using an offset, we don't have to worry about the index remaining correct.
PR Close#47490
related to https://github.com/angular/angular/pull/47438
After jest 28, `jest-environment-node` and `jest-environment-jsdom` need
to be installed by the user themselves, and the API has some breaking
changes, so this PR fix these issues to make the zone/jest integration
test code work as expected.
PR Close#47486
make sure that when an animation is used via the `useAnimation` function
and a delay has been provided then that delay gets correctly applied
(this PR is a follow up for #47285)
PR Close#47468
Expands the runtime to allow for basic host directives to be invoked within a template. This is achieved by making a second pass over the directives that were matched based on their selectors and producing a new array of directives that include host directives. Note that the ordering in the array is important, because it determines which host bindings and DI tokens will be overwritten.
PR Close#47430
We have some logic that generates the `InitialInputs` data structure by taking the static attributes of a node and looking them up in the `DirectiveDef.inputs` map to determine if they correspond to an input.
This works fine at the moment, but it will make it trickier to generate the correct `InitialInputs` for host directives, because the `DirectiveDef.inputs` entries might be aliased under a different name.
These changes rework the existing `generateInitialInputs` function so that it does its lookup on the remapped inputs that have accounted for the aliasing process already.
PR Close#47228
In a previous refactor, the `RootContext` was update to only contain a reference to a component. This commit perform further refactoring to get rid of the `RootContext` altogether, while storing component reference directly on the root view (without the `RootContext` wrapper).
PR Close#47056
make sure that when an animation is defined via the `animation` function
(and used via `useAnimation`) and a delay has been provided then that
delay gets correctly applied
resolves#47283
PR Close#47285
This commit updates the `renderApplication`, `renderModule` and `renderModuleFactory` functions to append a special marker (in a form of an attribute, called `ng-server-context`) to the component host elements. This marker is needed to analyze how a page was rendered.
PR Close#47103
Updated methods' description verbs. They are sometimes used with the assumption of the 'it' pronoun and sometimes not. For instance, the verb 'to construct' is used with 's' in one method description and not others. It is the case for other verbs as well. This is also remarkable in the description of the built-in methods of FormArray.
PR Close#47399
This commit updates runtime and compiler to drop unnecessary `srcset` sanitization. The sanitization was needed previously for old browsers, but all modern browsers can handle `srcset` safely without any additional sanitization.
See prior discussion in https://github.com/angular/angular/pull/45182.
Resolves#45164.
PR Close#47302
The forms `submit` event handlers have a `return false` to prevent form submissions from reloading the page, however this also prevents the browser behavior for forms with `method="dialog"`.
These changes add an exception since the `method="dialog"` doesn't refresh the page.
Fixes#47150.
PR Close#47308
As an ongoing effort to stabilize the NgOptimizedImage API before existing the Developer Preview, this commit renames the `rawSrc` attribute used for the NgOptimizedImage selector matching to `ngSrc`. The `rawSrcset` is also renamed to `ngSrcset` for consistency.
The motivation for this change is to align the attribute name better with other built-in directives, such as `ngFor`, `ngIf`, `ngClass`, `ngStyle`, etc.
Note: this is technically a breaking change, but since the NgOptimizedImage directive is in the Developer Preview mode, we land the change in a patch branch.
PR Close#47362
This commit adds a re-export of the `NgForOf` class as `NgFor` to improve the DX for cases when the directive is used as standalone. Developers can import `NgFor` class, which better matches the `ngFor` attribute used in a template.
PR Close#47309
Unify default value handling across injector and node injector: as long
as a default value is provided it has the same effect as specifying the
@Optional() flag.
Fixes#47109
PR Close#47242
Using raw objects as a lookup structure will inadvertently find methods defined on
`Object`, where strings are expected. This causes errors downstream when string
operations are applied on functions.
This commit switches over to use `Map`s in the DOM element schema registry to fix
this category of issues.
Fixes#46936
PR Close#47220
When the Angular compiler emits a diagnostic in a template file, it
forces TypeScript to parse that template. Templates are not TypeScript,
so this parse finds a bunch of parsing errors, which Angular then
ignores and we show the diagnostic anyways because we have more context.
This can lead to strange behavior in TypeScript because templates are so
weird that it can break the parser and crash the whole compiler.
For example, certain Angular templates can encounter failures fixed by
microsoft/TypeScript#45987, which are not easily debuggable and require
a TS upgrade to fix.
This commit introduces logic to handle the error gracefully, by falling
back to report the template error on the component class itself. The
diagnostic is extended to still reference the template location and
includes the failure's stack trace, to allow the parsing failure to be
reported to TypeScript (as parsing should in theory not cause a crash).
Closes#43970
PR Close#44001
This option has no longer any effect as Ivy is the only rendering engine.
BREAKING CHANGE: Angular compiler option `enableIvy` has been removed as Ivy is the only rendering engine.
PR Close#47346
Previously, when requesting non-cached asset resources from the network,
the ServiceWorker would strip off all request metadata (including
headers). This was done in order to avoid issues with opaque responses,
but it turned out to be overly aggressive, breaking/worsening legit
usecases (such as requesting compressed data).
This commit fixes this by preserving the headers of such requests.
For reference, Workbox passes the original request as is. (See for
example the [NetworkFirst][1] strategy).
> **Note**
> Data requests (i.e. requests for URLs that belong to a data-group) are
not affected by this. They already use the original resource as is.
[1]: 95f97a207f/packages/workbox-strategies/src/NetworkFirst.ts (L90)Fixes#24227
PR Close#47260
Previously, the `MockServerStateBuilder#withRedirect()` method did two
things: (a) define a redirect from one path to another and (b) specify
the contents of the redirect destination. This was confusing, because it
deviated from the regular way of specifying file contents, which is via
a `MockFileSystem` instance.
This commit slightly simplifies the process of defining redirects by
having the `withRedirect()` method only define the redirect and let the
contents of the redirect destination be specified as usual via
`MockFileSystem`. This makes `MockFileSystem` the single source of truth
for file contents used with `MockServerState`.
PR Close#47260
Previously, the default value of `MockRequest#credentials` (used in
testing) was set to `omit`. This was different than the default value
used in browsers, which is `same-origin` (see [Request docs on MDN][1])
and could lead to incorrect test results.
This commit changes the default value for `credentials` to `same-origin`
to match what would happen in the browser.
[1]: https://developer.mozilla.org/en-US/docs/Web/API/Request/credentials
PR Close#47260
Previously, the criteria for determining if a request was a
[navigation request][1] did not account for the request method. This
incorrectly identified HTML form submit POST requests as navigation
requests and served `index.html` instead of passing them through to the
server, thus breaking the form submission.
This commit fixes this by ensuring that only GET requests are considered
navigation requests.
> **Note**
> HTML forms with their method set to `GET` will still be affected by
> the issue. This is not a big concern, because using `GET` for form
> submission is quite uncommon and generally discouraged (due to
> limitations and security considerations).
[1]: https://angular.io/guide/service-worker-config#handling-navigation-requestsFixes#36368
PR Close#47263
This change removed the deprecated `canParse` method from all the TranslationParsers.
BREAKING CHANGE:
- `canParse` method has been removed from all translation parsers in `@angular/localize/tools`. `analyze` should be used instead.
- the `hint` parameter in the`parse` methods is now mandatory.
PR Close#47275
`PreconnectLinkChecker` checks to see if preconnect links have been added to the `<head>` element but uses `document` directly which does not exist when rendering in Angular Universal. This PR switches the `PreconnectLinkChecker` to use the `DOCUMENT` token instead so that the query works when SSR'ing
PR Close#47353
This is a tiny fix to add paragraph breaks in
image distortion warnings to make them a bit
easier to read. With this change, the intrinsic
and rendered image sizes are printed on their
own lines instead of mid-paragraph.
PR Close#47299
