The NgOptimizedImage directive was previously trying to use an
already destroyed injector in the ngOnDestroy callback. This fix
pre-injects necessery tokens so no injector calls are done
in the destroy process.
PR Close#47243
Yarn 1.x has a bug where it keeps nested unused node modules and doesn't
delete them automatically. This throws off Bazel in some scenarios when
the lock file is updated. This commit invalidates the cache to get a
fresh clean node modules cache without any unused nested directories.
PR Close#47240
Valid density descriptors used in the NgOptimizedImage can contain multiple
digits (ex. 1.25x, 25x). This change fixes the issue where density descriptors
with multiple digits (ex. 25x) where considered invalid.
Please note that a valid density descriptor might still be rejected by the
directive's validation logic if the supplied value is too big.
PR Close#47230
This commit fixes a bug where `rawSrcset` in the image
directive would allow density descriptors like `1.5x`
but not like `1.25x`. Now descriptors with 2+ digits
after the decimal point should work.
PR Close#47197
PR Close#47232
This is a tiny commit to add newlines in the image
directive's "oversized image" error. Currently, the
rendered and intrinsic image sizes are printed
mid-line, which make them a little hard to read.
This commit puts them each on their own line.
PR Close#47188
PR Close#47232
Back when we fixed the pullapprove verify command and our config to
properly include dot-prefixed files, we accidentally started matching
the pullapprove file in the dev-infra group.
This commit fixes this so that dev-infra is not triggered for basic
code owner config changes.
PR Close#47216
The peek-a-boo example image does not exactly reflect what happens. This updates the hook table and the image of the peek-a-boo output. Closes#40138
PR Close#47219
This helper accepts a class for an Angular trait, and returns the NgModule which owns that trait. This will be useful for the language service import project, which needs to edit import arrays on the module.
PR Close#47166
prior to this change the stagger timing was being built during the
ast building instead of dynamically when visiting the stagger animation,
thus it could not handle params correctly, this change makes it so that
during ast building a timing ast is built instead and that ast is used
dynammically to build animations which can handle params correctly
resolves#19786
PR Close#47208
Fix some TypeScript build warnings related to things like redundant `?.`
and `??` operators and uninitialized/non-nullable `@Input` properties.
You can see an example build with the warnings [on CI][1].
[1]: https://circleci.com/gh/angular/angular/1215057#step-104-232
PR Close#47201
The `mergeOverridenImplementation` processor tries to ensure that any
doc used with the `@overrideImplementation` annotation is not itself
public (and explicitly marks is as internal if it is).
Previously, it determined the public/private status of a doc by only
checking the value of the doc's `internal` property (which is mainly set
via the `@internal` annotation). This failed to account for docs marked
as "private exports" (such as those prefixed with `ɵ`), which are
essentially also treated as internal.
This would result in incorrect warnings. [Example warning][1]:
> Constructor doc forms/ɵFormControlCtor was not marked '@internal';
> adding this annotation.
This commit prevents the incorrect warning by also checking the value of
the doc's `privateExport` property to determine its public/private
status.
[1]: https://circleci.com/gh/angular/angular/1215057#step-104-164
PR Close#47201
Previously, error messages related to an overriden constructor doc would
stringify the doc as `[object Object]`, which is not helpful for
debugging the issue. For example:
> Constructor doc [object Object] was not marked '@internal'
This commit improves the error message by including the doc's ID
instead. For example, the above error message would become:
> Constructor doc forms/ɵFormControlCtor was not marked '@internal'
PR Close#47201
Create three new helper methods: `addElementToArrayLiteral`, `objectPropertyAssignmentForKey`, and `updateObjectValueForKey`. These methods make interacting with array and object literals easier.
These will be useful for the standalone imports feature, which will need to add new terms to import arrays in Component and NgModule decorators.
PR Close#47181
This helper accepts a class, and returns the primary Angular Decorator associated with that trait (e.g. the Component, Pipe, Directive, or NgModule decorator). This will be useful for the language service import project, which needs to edit import arrays inside the decorator.
PR Close#47180
Fix compilation issue with TypeScript nightly. TypeScript added a field
`0` to `RegExpMatchArray` in
https://github.com/microsoft/TypeScript/commit/3b80ddca212959ae5dcf2f48704be627283c6468.
Before that, `RegExpMatchArray` and `RegExpExecArray` were identical.
In one place Angular incorrectly expects `RegExpMatchArray` as the
result of a RegExp#exec() call. This assignment fails on TypeScript
nightly with the error:
Property '0' is missing in type 'RegExpExecArray' but required in type 'RegExpMatchArray'.
Fix the issue by using the type `RegExpExecArray` instead. This also
works for older TypeScript versions.
PR Close#47169
Type inference in cases involving `ControlConfig` was previously not working as desired. This was because the compiler was enforcing that `ControlConfig` is a *tuple* -- which is not always that easy to prove! By relaxing this constraint a bit, and just inferring from `ControlConfig` as an array, the type inference catches many more cases, and is generally more correct.
PR Close#47034
In #39004 some logic was introduced that tries to recover invalid expressions by treating the `=` token as a recovery point. It works by skipping ahead to the next recovery point inside the `skip` method which is called whenever an error is reported. This can lead to an infinite loop inside the `parseChain` method which assumes that reporting an error would've skipped over the token, but that won't happen since the `=` token is a recovery point. These changes resolve the infinite loop by breaking the loop if `error` didn't skip to a different token after the error was reported.
Fixes#47131.
PR Close#47151
This commit adds a console warning in development mode
if the ultimate rendered size of the image is much
smaller than the dimensions of the requested image.
In this case, the warning recommends adjusting the
size of the source image or using the `rawSrcset`
attribute to implement responsive sizing.
PR Close#47082
This commit cleans up some stylistic nits and
moves functions out of util.ts into files that
are more specifically named (url.ts and error.ts).
PR Close#47082
This commit fixes a bug in NgOptimizedImage where
if you bound the width or height attribute (e.g.
`[width]=width`), the attribute would only be
reflected as "ng-reflect-x". The actual "width" or
"height" attribute would not be set on the host
element. This is a problem because the exact named
attribute must be set on the element for the
browser to detect it and use it to reserve space
for the image (and thus prevent CLS).
PR Close#47082
This commit adds the docs for the NgOptimizedImage directive.
As a part of this commit, we also remove an export of directive-related symbols previously exposed as public APIs (i.e. APIs can not be used directly).
PR Close#47082
With this commit, the NgOptimizedImage directive will throw
a runtime error if it detects that one of the density descriptors
in rawSrcset is higher than 3x. It's generally not recommended
to use density descriptors higher than ~2, as it causes image to
download at very large sizes on mobile screens (thus slowing down
LCP). The density max is set conservatively to 3 in case apps
expect users to zoom in.
In future commits, we may want to throw even at densities >
than 2 and provide a configuration override for the zoom case.
PR Close#47082
Alex Rickabaugh