Signal functions are marked with a `[SIGNAL]` symbol property, which was
previously set to `true`.
This commit changes the `[SIGNAL]` property to have the `ReactiveNode` as
its value instead. This doesn't change the behavior of the signal library
itself, but exposes the signal internals in a way where developer tooling
can look at any signal and interact with its dependency graph.
PR Close#49529
Previously the signals implementation maintained separate interfaces for
`Producer` and `Consumer` nodes, with implementers choosing to implement
one or both interfaces. Operations defined against those interfaces were
exposed as importable functions to be called with the object implementing
the relevant contract as the first argument.
This commit refactors the implementation to merge both abstractions into a
single `ReactiveNode` base class, which represents both producers and
consumers in the graph. Implementers choose to interact with a subset of the
`ReactiveNode` API depending on their role in the graph. Operations are now
available as protected methods on the base class, instead of separate
functions.
PR Close#49529
Previously, effects were queued in the global microtask queue and executed
directly. This had an undesired consequence: whichever effect scheduled
first determined the zone in which all other effects would run, *and* that
zone depended on where the signal happened to be set which triggered that
first effect. This behavior would be extremely unpredictable.
This commit adds zone awareness to the effect API. effects now capture the
current zone when they're created, and this zone is used to run the effect
callback regardless of which zone set the signal.
PR Close#49529
This commit adds a `DeepReadonly` type to the signals API, and makes signal
getters return an immutable version of their value type. This doesn't
prevent all mutation but adds some friction against modifying the values
within signals outside of the proper mutation APIs.
PR Close#49529
The `effect` implemented in the signal library is useful for testing but
does not integrate with Angular. This commit moves that code to the
actual framework package and integrates it with automatic cleanup via
`DestroyRef`. A simpler effect implementation is used in the signal tests to
test the `Watch` primitive.
Further commits will update the scheduling to tie effects together with
change detection.
PR Close#49529
This commit switches the `signal` and `computed` APIs to accept an optional
options argument as their second argument, instead of an equality function
directly. The equality function has moved to an option in the options
argument.
PR Close#49529
This commit renames the `SettableSignal` interface to align with recent
naming changes. `WritableSignal` is considered a more suitable name since
it supports more mutation operations than just directly setting.
PR Close#49529
This commit adds an assertion function to the public API, which allows
authors of functions which rely on `inject` to validate that they're being
called with the right context. This mostly produces a nicer error message
than calling `inject()` and relying on Angular's default error message for
that.
PR Close#49529
This fixes two cases where DOM node navigation during hydration would be broken. It also fixes an infinite loop on clearing DOM during skip hydration blocks
PR Close#49615
This commit implements a simple tracker of the pending tasks during initial rendering. The class allows adding and removing tasks from the set. The class also exposes a promise that gets resolved once the last task is removed.
This tracker is needed to keep track of ongoing processes like Router navigation (and potentially HTTP requests) and acts as a signaling mechanism to SSR and hydration that the application is in the "stable" state and a serialization can be performed.
This class would also act as a future replacement for the `ApplicationRef.isStable` for zoneless applications.
PR Close#49576
`ComponentRef.setInput` currently sets the input on the component regardless
of the previous value the method was called with. This results in
different behavior from bindings in templates, which only set inputs
when the value differs in the `Object.is` check from its previous value.
BREAKING CHANGE: ComponentRef.setInput will only set the input on the
component if it is different from the previous value (based on `Object.is`
equality). If code relies on the input always being set, it should be
updated to copy objects or wrap primitives in order to ensure the input
value differs from the previous call to `setInput`.
PR Close#49607
Currently when the value of a styling property that has a unit is empty string a invalid value is generated.
Example:
`[style.width.px] = ""` will generate a value of `"px"`, when instead it should be `""`.
This causes browser to reset the value to an empty string. This is however not the case in Domino with changes in bfc9114d1e.
This commit fixes the issues and generate correct values.
PR Close#49460
This commit updates the serialization logic to avoid serializing a path for a node that was content projected (based on template information), but is not present in the DOM. This can happen if a <ng-content> is used with the *ngIf="false", for example: `<ng-content *ngIf="false" />`.
PR Close#49590
This commit updates the `LView` in Angular to be a `Consumer` of
signals. If a signal is read when executing a template, it marks the
view dirty. In addition, if a signal is read when executing host
bindings, it also marks views dirty.
One interesting thing about signal reads in host bindings
is that they perform a bit better than what we can do with today's
APIs. In order to re-execute host bindings for an `OnPush` component that
might have changed, you would probably inject `ChangeDetectorRef` and call
`markForCheck`. This will mark the _current component_ and parents
dirty. However, host bindings are executed as part of refreshing the
_parent_ so there is really no need to re-execute the current component
if the only thing that changed is the host bindings. When a signal is
read in host bindings, it marks the parent dirty and not the component
that defined the host binding.
Additionally, this commit avoids allocating a full consumer for each
`LView` by re-using a consumer until template execution results in a
signal read. At this point, we assign that consumer to the `LView` and
create a new consumer to "tentatively" use for the future `LView`
template executions.
Co-authored-by: Dylan Hunn <github@dylanhunn.com>
PR Close#49153
The `moduleId` directive field does not have any effect as of Ivy. In
View Engine it was used for resolving template and styles relative
to the component source location- but ultimately this is not needed
as the Angular compiler knows the source file location at build time,
and at runtime never even consulted `moduleId`. An XHR is always issue
using the extact specified URL.
For Angular CLI users, relative URLs in JIT are still possible because
the CLI has a TS transform that will replace the references with actual
Webpack imports.
`moduleId` does not seem worth keeping in the future, as it's not used
currently, and even if we would consider supporting relative JIT resource
URLs through it, it would be deeply coupled to CommonJS `module.id`.
DEPRECATED: The `@Directive`/`@Component` `moduleId` property is now
deprecated. It did not have any effect for multiple major versions and
will be removed in v17.
PR Close#49496
Previously, we've annotated all disconnected DOM nodes, even if they are not used in content projection. However, this situation is only possible in content projection and if it happens in other cases (for example, when a node was removed using direct DOM manipulations), this should be a mismatch error.
PR Close#49549
`entryComponents` have been deprecated since version 9, because with Ivy they weren't necessary. These changes remove any remaining references.
BREAKING CHANGE:
* `entryComponents` has been deleted from the `@NgModule` and `@Component` public APIs. Any usages can be removed since they weren't doing anyting.
* `ANALYZE_FOR_ENTRY_COMPONENTS` injection token has been deleted. Any references can be removed.
PR Close#49484
This adds user friendly error messages outlining exactly where hydration
mismatches occur and what was expected with easy to read stringified
DOM. It also includes a pointer to exactly where the mismatch happened
and lets the user know what was expected.
PR Close#49502
This commit updates the hydration logic to handle situations where DOM nodes might end up
being disconnected from the DOM tree. We serialize ids of those nodes into the hydration
state transfer data and use the information to switch from hydration to the regular "creation
mode" at runtime.
This situation may happen during the content projection, when some nodes don't make it
into one of the content projection slots (for example, when there is no default
<ng-content /> slot in projector component's template).
Note: we leverage the fact that we have this information available in the DOM emulation
layer (in Domino) for now. Longer-term solution should not rely on the DOM emulation and
only use internal data structures and state to compute this information.
PR Close#49471
This commit removes the `renderApplication` overload that accepts a root component as the first parameter.
BREAKING CHANGE:
`renderApplication` method no longer accepts a root component as first argument. Instead, provide a bootstrapping function that returns a `Promise<ApplicationRef>`.
Before
```ts
const output: string = await renderApplication(RootComponent, options);
```
Now
```ts
const bootstrap = () => bootstrapApplication(RootComponent, appConfig);
const output: string = await renderApplication(bootstrap, options);
```
PR Close#49463
This commit updates the serialization logic to detect a situation when i18n was used for a template and throw an error to indicate that hydration for i18n blocks is not yet supported.
Hydration for i18n blocks will be added in follow up versions.
PR Close#49497
Angular Hydration uses Components as a hydration boundary, i.e. you can enable/disable hydration on per-component basis. This commit enforces that the `ngSkipHydration` can only be applied on component host nodes (an error if thrown otherwise).
PR Close#49500
This commit updates the serialization logic to recognized similar repeated views and instead of including the same info over and over again, a special field is added to the serialized view object with a number of repetitions. The hydration logic also recognizes the flag and creates the necessary number of dehydrated views in a container.
This optimization should help minimize the amount of extra annotations required for cases like *ngFor with large number of items.
PR Close#49475
This commit adds a logic to remove all views that were not cleaimed during the hydration. The process is started once the ApplicationRef becomes stable on the client (which matches the timing of serialization on the server).
PR Close#49455
This commit updates the logic to store references to the comment (anchor) node as a segment head, when a container is empty (no views). Previously, the logic verified whether a container is non-empty and only stored segment head in this case. The problem is that the ViewContainerRef later on uses the segment head to start the DOM lookup and as a result, the process can not be started, since we have `null`. Now the segment head reference would point to a comment node (the anchor node) and ViewContainerRef would be able to start the lookup process. For empty views, the segment head would become an anchor of the container (which is exactly what we need).
PR Close#49455
This commit adds serialization and hydration logic for content projection.
While hydration for regular elements relies on their location in the TNode tree, the content projection may move elements around, so in order to hydrate them correcty, the runtime needs some extra information. This commit adds a serialization logic that adds element locations (instructions on how to navigate to a particular element from another known location of other element) into the hydration state for the following cases:
- when a TNode is a first element in projection segment (other nodes are linked from that node)
- when a TNode's next sibling is different before and after projection (we serialize extra info about the template-based sibling)
- when a TNode's previous sibling was a content projection (i.e. `<ng-content>` slot), because we can not rely on the previous element in this case (projection happens at a later point)
PR Close#49454
This change makes it possible to remove a previously registered destroy
callback - to do so it is enough to call the unregistration function
returned from the onDestroy method call.
PR Close#49493
Before this change the effect creation function was returning
an instance of the Effect interface with explicit destroy and
schedule operations. This commit makes the following changes
to the return type of effect:
- the interface is renamed from Effect to EffectRef;
- the only available method on this interface is destroy.
We want to tighten the interface in question so only a minimal
set of useful operations is initially exposed.
PR Close#49490
This commit adds serialization and hydration logic for content projection.
While hydration for regular elements relies on their location in the TNode tree, the content projection may move elements around, so in order to hydrate them correcty, the runtime needs some extra information. This commit adds a serialization logic that adds element locations (instructions on how to navigate to a particular element from another known location of other element) into the hydration state for the following cases:
- when a TNode is a first element in projection segment (other nodes are linked from that node)
- when a TNode's next sibling is different before and after projection (we serialize extra info about the template-based sibling)
- when a TNode's previous sibling was a content projection (i.e. `<ng-content>` slot), because we can not rely on the previous element in this case (projection happens at a later point)
PR Close#49454
This commit adds the ability to override the `WeakRef` implementation used
by signals as a private API. This can be used in certain environments to
allow tests which use older browsers to still run, albeit without weak
references.
PR Close#49466
Showing a minimum app to reproduce the bug.
1. Create the app and add angular material.
```
ng new project
cd project
ng add @angular/material
```
1. Overwrite the src/app/app.component.html with minimal content.
```
<button mat-button *ngIf="true"><span *ngIf="true" class="{{'a'}}"></span></button>
```
1. Run the app. The button is not shown because of an exception.
```
main.ts:6
ERROR TypeError: item.toLowerCase is not a function
at isCssClassMatching (core.mjs:8726:35)
at isNodeMatchingSelector (core.mjs:8814:22)
at isNodeMatchingSelectorList (core.mjs:8931:13)
at matchingProjectionSlotIndex (core.mjs:14179:13)
at Module.ɵɵprojectionDef (core.mjs:14222:49)
at MatButton_Template (button.mjs:113:99)
at executeTemplate (core.mjs:10534:9)
at renderView (core.mjs:10356:13)
at renderComponent (core.mjs:11529:5)
at renderChildComponents (core.mjs:10216:9)
```
Because isCssClassMatching() function does not take care if the value is not string, while attrs[] may contain AttributeMarker which is actually numbers, item.toLowerCase() throws the exception.
Just inserted a check if the item is string.
Created a testcase for the original fix. It causes an exception without the fix.
fix(core): add a check code to avoid an exception inside isCssClassMatching
Showing a minimum app to reproduce the bug.
1. Create the app and add angular material.
```
ng new project
cd project
ng add @angular/material
```
1. Add `import { MatButtonModule } from '@angular/material/button'`,
and also MatButtonModule inside @NgModule imports in src/app/app.module.ts to use MatButtonModule.
1. Overwrite the src/app/app.component.html with minimal content.
```
<button mat-button *ngIf="true"><span *ngIf="true" class="{{'a'}}"></span></button>
```
1. Run the app. The button is not shown because of an exception.
```
main.ts:6
ERROR TypeError: item.toLowerCase is not a function
at isCssClassMatching (core.mjs:8726:35)
at isNodeMatchingSelector (core.mjs:8814:22)
at isNodeMatchingSelectorList (core.mjs:8931:13)
at matchingProjectionSlotIndex (core.mjs:14179:13)
at Module.ɵɵprojectionDef (core.mjs:14222:49)
at MatButton_Template (button.mjs:113:99)
at executeTemplate (core.mjs:10534:9)
at renderView (core.mjs:10356:13)
at renderComponent (core.mjs:11529:5)
at renderChildComponents (core.mjs:10216:9)
```
Because isCssClassMatching() function does not take care if the value is not string, while attrs[] may contain AttributeMarker which is actually numbers, item.toLowerCase() throws the exception.
Just inserted a check if the item is string.
PR Close#48888
Adds support for marking a directive input as required. During template type checking, the compiler will verify that all required inputs have been specified and will raise a diagnostic if one or more are missing. Some specifics:
* Inputs are marked as required by passing an object literal with a `required: true` property to the `Input` decorator or into the `inputs` array.
* Required inputs imply that the directive can't work without them. This is why there's a new check that enforces that all required inputs of a host directive are exposed on the host.
* Required input diagnostics are reported through the `OutOfBandDiagnosticRecorder`, rather than generating a new structure in the TCB, because it allows us to provide a better error message.
* Currently required inputs are only supported during AOT compilation, because knowing which bindings are present during JIT can be tricky and may lead to increased bundle sizes.
Fixes#37706.
PR Close#49468
This reverts commit 13dd614cd1.
This breaks a g3 Typescript compilation tests where diagnostics are
expected for a missing input in the component.
PR Close#49467
Angular uses inline styles to insert the styles associated with a component. This violates the strict styles [Content Security Policy](https://web.dev/strict-csp/) which doesn't allow inline styles by default. One way to allow the styles to be applied is to set a `nonce` attribute on them, but because the code for inserting the stylesheets is deep inside the framework, users weren't able to provide it without accessing private APIs.
These changes add a new `CSP_NONCE` injection token that will allow users to provide a nonce, if their app is using CSP. If the token isn't provided, the framework will look for an `ngCspNonce` attribute on the app's root node instead. The latter approach is provided as a convenience for apps that render the `index.html` through a server, e.g. `<app ngCspNonce="{% randomNonceAddedByTheServer %}"></app>`.
This PR addresses adding the nonce to framework-generated styles. There will be follow-up PRs that add support for it in critical CSS tags in the CLI, and in Angular Material.
Fixes#6361.
PR Close#49444
Adds support for marking a directive input as required. During template type checking, the compiler will verify that all required inputs have been specified and will raise a diagnostic if one or more are missing. Some specifics:
* Inputs are marked as required by passing an object literal with a `required: true` property to the `Input` decorator or into the `inputs` array.
* Required inputs imply that the directive can't work without them. This is why there's a new check that enforces that all required inputs of a host directive are exposed on the host.
* Required input diagnostics are reported through the `OutOfBandDiagnosticRecorder`, rather than generating a new structure in the TCB, because it allows us to provide a better error message.
* Currently required inputs are only supported during AOT compilation, because knowing which bindings are present during JIT can be tricky and may lead to increased bundle sizes.
Fixes#37706.
PR Close#49453
This reverts commit 1a6ca68154.
This breaks tests in google3 which might be depending on private APIs. We
need to update these tests before we can land this PR.
PR Close#49449
This restores the separation between adjacent text nodes that is lost during server serialization when parsed by the browser. It adds a special comment node just prior to the serialization process that is then restored as a separated node during hydration.
PR Close#49419
During DOM serialization, empty text nodes are lost when the client parses the DOM. To solve this problem comment nodes are added where the empty nodes are located right before serialization. Those comments then get replaced during hydration with the proper empty text nodes.
PR Close#49419
Prior to this change, a random application ID was generated each time which forced users using server rendering to provide an application ID themselves. This was needed to handle rare cases when multiple Angular applications are rendered on the same page.
With this change the application ID is no longer generated randomly and instead it is hard coded.
BREAKING CHANGE:
The `APP_ID` token value is no longer randomly generated. If you are bootstrapping multiple application on the same page you will need to set to provide the `APP_ID` yourself.
```ts
bootstrapApplication(ComponentA, {
providers: [
{ provide: APP_ID, useValue: 'app-a' },
// ... other providers ...
]
});
```
PR Close#49422
This commit deprecated ` BrowserModule.withServerTransition` instead `APP_ID` should be used instead to configure the app id.
DEPRECATED: `BrowserModule.withServerTransition` has been deprecated. `APP_ID` should be used instead to set the application ID.
NB: Unless, you render multiple Angular applications on the same page, setting an application ID is not necessary.
Before:
```ts
imports: [
BrowserModule.withServerTransition({ appId: 'serverApp' }),
...
]
```
After:
```ts
imports: [
BrowserModule,
{ provide: APP_ID, useValue: 'serverApp' },
...
],
```
PR Close#49422
Adds support for marking a directive input as required. During template type checking, the compiler will verify that all required inputs have been specified and will raise a diagnostic if one or more are missing. Some specifics:
* Inputs are marked as required by passing an object literal with a `required: true` property to the `Input` decorator or into the `inputs` array.
* Required inputs imply that the directive can't work without them. This is why there's a new check that enforces that all required inputs of a host directive are exposed on the host.
* Required input diagnostics are reported through the `OutOfBandDiagnosticRecorder`, rather than generating a new structure in the TCB, because it allows us to provide a better error message.
* Currently required inputs are only supported during AOT compilation, because knowing which bindings are present during JIT can be tricky and may lead to increased bundle sizes.
Fixes#37706.
PR Close#49304
This commit implements hydration support for view containers, which should make `*ngIf`, `*ngFor` and other structural directive work with hydration.
The logic also respects the `ngSkipHydration` flag and skips hydration in such cases.
PR Close#49382
This commit updates the serialization logic to include information about views from view containers (based on underlying LContainer info). The information is needed during hydration to pick a correct instance of dehydrated view (this logic will be implemented in a followup commit).
PR Close#49382
This commit updates the serialization logic to include information about templates that were used to create embedded views. The information is needed during hydration to pick a correct instance of dehydrated view (this logic will be implemented in a followup commit).
PR Close#49382
