The control flow migration was incorrectly removing `ng-template` elements in scenarios where they were referenced by multiple `*ngIf` directives' `else` clauses and also used independently via `ngTemplateOutlet`.
PR Close#63996
When `noUncheckedIndexedAccess` is not enabled, indexed accesses do not include `undefined` in the type.
This makes the OptionalChainNotNullableCheck inconvenient when wanting to safe guard an indexed access (and when the typings isn't helping you).
Ideally project should enable `noUncheckedIndexedAccess` for better typesafety, but this is often inconveninent (we do not enable it by default) and often not the resort of the developer.
So more better coding convinience, `OptionalChainNotNullableCheck` will not raise an error/warning on indexed Acessed followed by an optional chaining when `noUncheckedIndexedAccess` is not enabled.
See also #63809 which will detect more cases
PR Close#64007
This commit changes `Resource.hasValue()` and its derived types to improve narrowing
of resources whose generic type either does not include `undefined` (i.e. when a default
value has been provided) or when the generic type is `unknown`. This fixes the undesirable
behavior where `hasValue()` would cause the `else` branch of an `hasValue()` conditional
to have a narrowed type of `never`, given that the `hasValue()`'s type guard covers the
entire type range already (meaning that the type in the else-branch cannot be inhabited
in the type system, yielding the `never` type).
By making the `hasValue()` method only a type guard when the generic type includes `undefined`
these problems are avoided.
Fixes#60766Fixes#63545Fixes#63982
PR Close#63994
Fixes an error where using an alias in a defer block caused the compiler CLI to fail when parsing. The resolution logic in ComponentDecoratorHandler was updated to correctly handle deferred dependencies with aliased imports.
PR Close#63966
Since those are top level APIs, `ngDevMode` might not be available at runtime if they're invoked before the variable is set.
fixes#62796
PR Close#63875
This adds an optional flag to the renderer on `removeChild` called `requireSynchronousElementRemoval`, which can tell any downstream renderer that elements need to be removed synchronously. This gets passed down to the legacy animation renderer to ensure that any elements that set this flag aren't impacted by that renderers changes to timing.
fixes: #63893
PR Close#63921
An invalid APP_ID could be responsible to generating broken CSS selectors. (eg `:` is an example for a character that breaks a selector by being a separator for pseudo-selectors.)
We now throw an error if the provided value is not alphanumerical
PR Close#63252
https://github.com/angular/angular/pull/62630 made it so that all ARIA
property bindings would write to their corresponding attribute instead.
The primary motivation for this change was to ensure that ARIA
attributes were always rendered correctly on the server, where the
emulated DOM may not correctly reflect ARIA properties as attributes.
Furthermore, this change added support for binding to ARIA attributes
using the property binding syntax (e.g. `[aria-label]`).
Unfortunately, https://github.com/angular/angular/pull/62630 relied on
the incorrect assumptions that an ARIA property name could be converted
to its attribute name (without hardcoding the conversion), and that the
value of an ARIA property matched its corresponding attribute. For
example, the `ariaLabelledByElements` property's value is an array of
DOM elements, while the corresponding `aria-labelledby` attribute's
value is a string containing the IDs of the DOM elements.
This partially reverts https://github.com/angular/angular/pull/62630 so
that only property bindings with ARIA attribute names (begin with
`aria-`) are converted to attribute bindings.
* `[ariaLabel]` will revert to binding to the `ariaLabel` property.
* `[aria-label]` will continue binding to the `aria-label` attribute.
Note the only difference between `[aria-label]` and `[attr.aria-label]`
is that the former will attempt to bind to inputs of the same name while
the latter will not.
PR Close#63925
Fixes a bug in the standalone migration where 2 imported modules have the same class name but 1 is imported with an alias and would not be added to the component imports array when migrating
Fixes#63913
PR Close#63934
Currently if `TestBed.overrideComponent` is used on a class that uses initializer APIs (e.g. `input()`), the initializer metadata will be wiped out, because `overrideComponent` re-compiles the class with the information set by `setClassMetadata`. `setClassMetadata` only captures decorated members at the moment.
These changes introduce some logic to capture the new initializer-based APIs in `setClassMetadata` as well.
Fixes#57944.
PR Close#63904
Reworks the logic that tracks the decorator metadata for members to do so using the output AST, rather than wrapping the TypeScript AST. This makes it easier to programmatically generate new members that weren't part of the TypeScript AST before.
PR Close#63904
This removes the Zone providers from the `BrowserTestingModule`. These
already exist by default in all other entrypoints to Angular
environments (TestBed compiler, bootstrapModule, and createApplication).
PR Close#63872
Previously, HOST_TAG_NAME had its __NG_ELEMENT_ID__ set at the top level. This
prevented tree-shaking, since the bundler had to keep the assignment as a
potential side effect even when the token was never used.
This change moves the token creation and __NG_ELEMENT_ID__ assignment into a
@__PURE__ IIFE. If HOST_TAG_NAME is not injected anywhere, the IIFE result is
unused and can be dropped entirely by the optimizer. If it is used, the token
still behaves the same at runtime.
PR Close#63861
