Commit Graph

65 Commits

Author SHA1 Message Date
Andrew Kushnir 0723331b2a refactor(forms): move common validators-related logic to the `AbstractControlDirective` class (#38280)
This commit refactors validators-related logic that is common across most of the directives.

A couple notes on this refactoring:
* common logic was moved to the `AbstractControlDirective` class (including `validator` and
`asyncValidator` getters)
* sync/async validators are now composed in `AbstractControlDirective` class eagerly when validators
are set with `_setValidators` and `_setAsyncValidators` calls and the result is stored in directive
instance (thus getters return cached versions of validator fn). This is needed to make sure composed
validator function remains the same (retains its identity) for a given directive instance, so that
this function can be added and later removed from an instance of an AbstractControl-based class
(like `FormControl`). Preserving validator function is required to perform proper cleanup (in followup
PRs) of the AbstractControl-based classes when a directive is destroyed.

PR Close #38280
2020-10-28 09:48:20 -07:00
Misko Hevery ca11ef2376 fix(core): Store ICU state in `LView` rather than in `TView` (#39233)
Before this refactoring/fix the ICU would store the current selected
index in `TView`. This is incorrect, since if ICU is in `ngFor` it will
cause issues in some circumstances. This refactoring properly moves the
state to `LView`.

closes #37021
closes #38144
closes #38073

PR Close #39233
2020-10-21 18:33:00 -07:00
Keen Yee Liau 7768aeb62f fix(platform-server): Resolve absolute URL from baseUrl (#39334)
This commit fixes a bug when `useAbsoluteUrl` is set to true and
`ServerPlatformLocation` infers the base url from the supplied
`url`. User should explicitly set the `baseUrl` when they turn on
`useAbsoluteUrl`.

Breaking change:
If you use `useAbsoluteUrl` to setup `platform-server`, you now need to
also specify `baseUrl`.
We are intentionally making this a breaking change in a minor release,
because if `useAbsoluteUrl` is set to `true` then the behavior of the
application could be unpredictable, resulting in issues that are hard to
discover but could be affecting production environments.

PR Close #39334
2020-10-21 09:41:58 -07:00
Kristiyan Kostadinov 9fb7bdea89 fix(router): incorrect signature for createUrlTree (#39347)
The type of the `navigationExtras` param was accidetally changed to the wrong symbol
in 783a5bd7bb.
These changes revert it to the correct one.

PR Close #39347
2020-10-20 13:36:12 -07:00
Adam c4becca0e4 feat(router): add new initialNavigation options to replace legacy (#37480)
As of Angular v4, four of the options for
`ExtraOptions#initialNavigation` have been deprecated. We intend
to remove them in v11. The final state for these options is:
`enabledBlocking`, `enabledNonBlocking`, and `disabled`. We plan
to remove and deprecate the remaining option in the next two
major releases.

New options:
- `enabledNonBlocking`: same as legacy_enabled
- `enabledBlocking`: same as enabled

BREAKING CHANGE:

* The `initialNavigation` property for the options in
  `RouterModule.forRoot` no longer supports `legacy_disabled`,
  `legacy_enabled`, `true`, or `false` as valid values.
  `legacy_enabled` (the old default) is instead `enabledNonBlocking`
* `enabled` is deprecated as a valid value for the
  `RouterModule.forRoot` `initialNavigation` option. `enabledBlocking`
  has been introduced to replace it

PR Close #37480
2020-10-14 11:20:51 -07:00
Joey Perrott 783a5bd7bb fix(router): remove preserveQueryParams symbol (#38762)
Remove preserveQueryParams as it was deprecated for removal in v4, use
queryParamsHandling="preserve" instead.

BREAKING CHANGE: preserveQueryParams has been removed, use
queryParamsHandling="preserve" instead

PR Close #38762
2020-10-14 10:40:55 -07:00
Adrien Crivelli b0b4953fd6 fix(router): Allow undefined inputs on routerLink (#39151)
This make it coherent with typing of Router.createUrlTree to which
those inputs are directly forwarded to. And hence it allow to pass
`undefined`, `null` or a value to the routerLink directive.

BREAKING CHANGE: in most cases this should not break, but if you were
accessing the values of `queryParams`, `fragment` or `queryParamsHandling`
you might need to relax the typing to also accept `undefined` and `null`.

Signed-off-by: Adrien Crivelli <adrien.crivelli@gmail.com>

PR Close #39151
2020-10-12 12:52:37 -07:00
Kristiyan Kostadinov 4a1c12c773 feat(core): remove ViewEncapsulation.Native (#38882)
Removes `ViewEncapsulation.Native` which has been deprecated for several major versions.

BREAKING CHANGES:
* `ViewEncapsulation.Native` has been removed. Use `ViewEncapsulation.ShadowDom` instead. Existing
usages will be updated automatically by `ng update`.

PR Close #38882
2020-10-08 11:56:03 -07:00
Andrew Kushnir 246de9aaad fix(forms): improve types of directive constructor arguments (#38944)
Prior to this change, the `validators` and `asyncValidators` fields of a few Forms directives
were typed as `any[]`. This commit updates the types and makes them consistent for all directives
in the Forms package.

BREAKING CHANGE:

Directives in the `@angular/forms` package used to have `any[]` as a type of `validators` and
`asyncValidators` arguments in constructors. Now these arguments are properly typed, so if your
code relies on directive constructor types it may require some updates to improve type safety.

PR Close #38944
2020-10-06 10:14:21 -07:00
Ajit Singh 3c474ecf56 fix(common): add boolean to valid json for testing (#37893)
boolean is a valid json but at present we cannot test Http request using boolean added support for boolean requests

Fixes #20690

PR Close #37893
2020-10-05 17:07:41 -07:00
lazarljubenovic f4f1bcc997 fix(forms): include null in .parent of abstract control (#32671)
It's perfectly valid for an abstract control not to have a defined parent; yet previously the
types were asserting that AbstractControl#parent is not a null value. This changes correctly
reflects the run-time behavior through the types.

BREAKING CHANGE: Type of AbstractFormControl.parent now includes null

`null` is now included in the types of .parent. If you don't already have a check for this case,
the TypeScript compiler might compain. A v11 migration exists which adds the not-null assertion
operator where necessary.

In an unlikely case your code was testing the parnet against undefined with sitrct equality,
you'll need to change this to `=== null` instead, since the parent is not explicily initialized
with `null` instead of being left `undefined`.

Fixes #16999

PR Close #32671
2020-10-05 09:30:44 -07:00
Andrea Canciani 4744c229db fix(common): correct typing and implementation of `SlicePipe` (#37447)
Even in the overloads, state that it can accept `null` and
`undefined`, in order to ensure easy composition with `async`.

Additionally, change the implementation to return `null` on an
`undefined` input, for consistency with other pipes.

BREAKING CHANGE:
The `slice` pipe now returns `null` for the `undefined` input value,
which is consistent with the behavior of most pipes. If you rely on
`undefined` being the result in that case, you now need to check for it
explicitly.

PR Close #37447
2020-09-28 12:23:33 -04:00
Andrea Canciani 4dfe0fa068 fix(common): correct and simplify typing of `KeyValuePipe` (#37447)
As shown in the tests, `KeyValuePipe.transform` can accept
`undefined`, in which case it always returns `null`.

Additionally, the typing for `string` keys can be made generic, so the
comparison function is only required to accept the relevant cases.

Finally, the typing for `number` records now shows that the comparison
function and the result entries will actually receive the string version
of the numeric keys, just as shown in the tests.

BREAKING CHANGE:
The typing of the `keyvalue` pipe has been fixed to report that for
input objects that have `number` keys, the result will contain the
string representation of the keys. This was already the case and the
code has simply been updated to reflect this. Please update the
consumers of the pipe output if they were relying on the incorrect
types. Note that this does not affect use cases where the input values
are `Map`s, so if you need to preserve `number`s, this is an effective
way.

PR Close #37447
2020-09-28 12:23:33 -04:00
Andrea Canciani 3b919ef10f fix(common): correct and simplify typing of I18nPluralPipe (#37447)
I18nPluralPipe can actually accept `null` and `undefined` (which are
convenient for composing it with the async pipe), but it is currently
typed to only accept `number`.

PR Close #37447
2020-09-28 12:23:32 -04:00
Andrea Canciani 7b2aac97df feat(common): stricter types for number pipes (#37447)
Make typing of number pipes stricter to catch some misuses (such as
passing an Observable or an array) at compile time.

BREAKING CHANGE:
The signatures of the number pipes now explicitly state which types are
accepted. This should only cause issues in corner cases, as any other
values would result in runtime exceptions.

PR Close #37447
2020-09-28 12:23:32 -04:00
Andrea Canciani daf8b7f100 feat(common): stricter types for DatePipe (#37447)
Make typing of DatePipe stricter to catch some misuses (such as passing
an Observable or an array) at compile time.

BREAKING CHANGE:
The signature of the `date` pipe now explicitly states which types are
accepted. This should only cause issues in corner cases, as any other
values would result in runtime exceptions.

PR Close #37447
2020-09-28 12:23:32 -04:00
Andrea Canciani 5f815c0565 fix(common): correct and simplify typing of AsyncPipe (#37447)
`AsyncPipe.transform` will never return `undefined`, even when passed
`undefined` in input, in contrast with what was declared in the
overloads.

Additionally the "actual" method signature can be updated to match the
most generic case, since the implementation does not rely on wrappers
anymore.

BREAKING CHANGE:
The async pipe no longer claims to return `undefined` for an input that
was typed as `undefined`. Note that the code actually returned `null` on
`undefined` inputs. In the unlikely case you were relying on this,
please fix the typing of the consumers of the pipe output.

PR Close #37447
2020-09-28 12:23:32 -04:00
Andrea Canciani c7d5555dfb fix(common): let case conversion pipes accept type unions with `null` (#36259) (#37447)
The old implementation of case conversion types can handle several
values which are not strings, but the signature did not reflect this.

The new one reports errors when falsy non-string inputs are given to
the pipe (such as `false` or `0`) and has a new signature which
instead reflects the behaviour on `null` and `undefined`.

Fixes #36259

BREAKING CHANGE:
The case conversion pipes no longer let falsy values through. They now
map both `null` and `undefined` to `null` and raise an exception on
invalid input (`0`, `false`, `NaN`) just like most "common pipes". If
your code required falsy values to pass through, you need to handle them
explicitly.

PR Close #37447
2020-09-28 12:23:32 -04:00
Andrew Scott e4f4d18e7e refactor(router): Adjust type of parameter in navigateByUrl and createUrlTree to be more accurate (#38227)
`router.navigateByUrl` and `router.createUrlTree` only use a subset of the `NavigationExtras`. This commit
changes the parameter type to use new interfaces that only specify the properties used by
those function implementations. `NavigationExtras` extends both of those interfaces.

Fixes #18798

BREAKING CHANGE: While the new parameter types allow a variable of type
`NavigationExtras` to be passed in, they will not allow object literals,
as they may only specify known properties. They will also not accept
types that do not have properties in common with the ones in the `Pick`.
To fix this error, only specify properties from the `NavigationExtras` which are
actually used in the respective function calls or use a type assertion
on the object or variable: `as NavigationExtras`.

PR Close #38227
2020-09-25 10:36:13 -04:00
klemenoslaj a2068523fd feat(service-worker): add the option to prefer network for navigation requests (#38565)
This commit introduces a new option for the service worker, called
`navigationRequestStrategy`, which adds the possibility to force the service worker
to always create a network request for navigation requests.
This enables the server redirects while retaining the offline behavior.

Fixes #38194

PR Close #38565
2020-09-22 09:29:20 -07:00
Ingo Bürk 2b1b7180db fix(forms): type NG_VALUE_ACCESSOR injection token as array (#29723)
NG_VALUE_ACCESSOR is a multi injection token, users can and
should expect more than one ControlValueAccessor to be
available (and this is how it is used in @angular/forms).

This is now reflected in the definition of the injection token
by typing it as an array of ControlValueAccessor. The motivating
reason is that using the programmatic Injector api will now
type Injector#get correspondingly.

fixes #29351

BREAKING CHANGES

NG_VALUE_ACCESSOR is now typed as a readonly array rather than
a mutable scalar. It is used as a multi injection token and as
such it should always be expected that more than one accessor
may be returned.

PR Close #29723
2020-09-21 12:26:07 -07:00
Ajit Singh dd8d8c8289 fix(common): add `params` and `reportProgress` options to `HttpClient.put()` overload (#37873)
When the response type is JSON, the `put()` overload signature did not have `reportProgress`
and  `params` options. This makes it difficult to type-check this overload.

This commit adds them to the overload signature.

Fixes #23600

PR Close #37873
2020-09-16 15:28:21 -07:00
Artem Halas 6acea54f62 fix(common): mark locale data arrays as readonly (#30397)
To discourage developers from mutating the arrays returned
from the following methods, their return types have been marked
as readonly.

* `getLocaleDayPeriods()`
* `getLocaleDayNames()`
* `getLocaleMonthNames()`
* `getLocaleEraNames()`

Fixes #27003

BREAKING CHANGE:
The locale data API has been marked as returning readonly arrays, rather
than mutable arrays, since these arrays are shared across calls to the
API. If you were mutating them (e.g. calling `sort()`, `push()`, `splice()`, etc)
then your code will not longer compile. If you need to mutate the array, you
should now take a copy (e.g. by calling `slice()`) and mutate the copy.

PR Close #30397
2020-09-10 15:35:30 -07:00
Yuhei Yasuda c880e393e9 fix(router): If users are using the Alt key when clicking the router links, prioritize browser’s default behavior (#38375)
In most browsers, clicking links with the Alt key has a special behavior, for example, Chrome
downloads the target resource. As with other modifier keys, the router should stop the original
navigation to avoid preventing the browser’s default behavior.

When users click a link while holding the Alt key together, the browsers behave as follows.

Windows 10:

| Browser    | Behavior                                    |
|:-----------|:--------------------------------------------|
| Chrome 84  | Download the target resource                |
| Firefox 79 | Prevent navigation and therefore do nothing |
| Edge 84    | Download the target resource                |
| IE 11      | No impact                                   |

macOS Catalina:

| Browser    | Behavior                                    |
|:-----------|:--------------------------------------------|
| Chrome 84  | Download the target resource                |
| Firefox 79 | Prevent navigation and therefore do nothing |
| Safari 13  | Download the target resource                |

PR Close #38375
2020-09-08 14:07:11 -07:00
Alex Rickabaugh 3e97435f1c refactor(compiler-cli): split out template diagnostics package (#38576)
The template type-checking engine includes utilities for creating
`ts.Diagnostic`s for component templates. Previously only the template type-
checker itself created such diagnostics. However, the template parser also
produces errors which should be represented as template diagnostics.

This commit prepares for that conversion by extracting the machinery for
producing template diagnostics into its own sub-package, so that other parts
of the compiler can depend on it without depending on the entire template
type-checker.

PR Close #38576
2020-09-03 14:02:31 -07:00
Joey Perrott fdea1804d6 fix(core): remove CollectionChangeRecord symbol (#38668)
Remove CollectionChangeRecord as it was deprecated for removal in v4, use
IterableChangeRecord instead.

BREAKING CHANGE: CollectionChangeRecord has been removed, use IterableChangeRecord
instead

PR Close #38668
2020-09-02 16:45:19 -07:00
Sonu Kapoor 036a2faf02 feat(service-worker): add `UnrecoverableStateError` (#36847)
In several occasions it has been observed when the browser has evicted
eagerly cached assets from the cache and which can also not be found on the
server anymore. This can lead to broken state where only parts of the application
will load and others will fail.

This commit fixes this issue by checking for the missing asset in the cache
and on the server. If this condition is true, the broken client will be
notified about the current state through the `UnrecoverableStateError`.

Closes #36539

PR Close #36847
2020-08-31 11:41:11 -07:00
Alan Agius 0fc44e0436 feat(compiler-cli): add support for TypeScript 4.0 (#38076)
With this change we add support for TypeScript 4.0

PR Close #38076
2020-08-24 13:06:59 -07:00
Andrew Scott dbfb50e9f4 fix(router): ensure routerLinkActive updates when associated routerLinks change (#38511)
This commit introduces a new subscription in the `routerLinkActive` directive which triggers an update
when any of its associated routerLinks have changes. `RouterLinkActive` not only needs to know when
links are added or removed, but it also needs to know about if a link it already knows about
changes in some way.

Quick note that `from...mergeAll` is used instead of just a simple
`merge` (or `scheduled...mergeAll`) to avoid introducing new rxjs
operators in order to keep bundle size down.

Fixes #18469

PR Close #38511
2020-08-18 10:21:49 -07:00
Andrew Scott bee44b3359 Revert "fix(router): ensure routerLinkActive updates when associated routerLinks change (#38349)" (#38511)
This reverts commit e0e5c9f195.
Failures in Google tests were detected.

PR Close #38511
2020-08-18 10:21:47 -07:00
Andrew Scott e0e5c9f195 fix(router): ensure routerLinkActive updates when associated routerLinks change (#38349)
This commit introduces a new subscription in the `routerLinkActive` directive which triggers an update
when any of its associated routerLinks have changes. `RouterLinkActive` not only needs to know when
links are added or removed, but it also needs to know about if a link it already knows about
changes in some way.

Quick note that `from...mergeAll` is used instead of just a simple
`merge` (or `scheduled...mergeAll`) to avoid introducing new rxjs
operators in order to keep bundle size down.

Fixes #18469

PR Close #38349
2020-08-17 12:33:59 -07:00
Anas Barghoud ca798804b2 fix(router): export DefaultRouteReuseStrategy to Router public_api (#31575)
export DefaultRouteStrategy class that was used internally and exposed, and add documentation for each one of methods

PR Close #31575
2020-08-13 16:02:41 -07:00
Andrew Scott 71138f6004 feat(compiler-cli): Add compiler option to report errors when assigning to restricted input fields (#38249)
The compiler does not currently report errors when there's an `@Input()`
for a `private`, `protected`, or `readonly` directive/component class member.
This change adds an option to enable reporting errors when a template
attempts to bind to one of these restricted input fields.

PR Close #38249
2020-08-11 09:55:48 -07:00
JiaLiPassion 8fbf40bf40 feat(core): update reference and doc to change `async` to `waitAsync`. (#37583)
The last commit change `async` to `waitForAsync`.
This commit update all usages in the code and also update aio doc.

PR Close #37583
2020-08-03 12:54:13 -07:00
Alex Rickabaugh c573d91285 refactor(compiler-cli): allow program strategies to opt out of inlining (#38105)
The template type-checking engine relies on the abstraction interface
`TypeCheckingProgramStrategy` to create updated `ts.Program`s for
template type-checking. The basic API is that the type-checking engine
requests changes to certain files in the program, and the strategy provides
an updated `ts.Program`.

Typically, such changes are made to 'ngtypecheck' shim files, but certain
conditions can cause template type-checking to require "inline" operations,
which change user .ts files instead. The strategy used by 'ngc' (the
`ReusedProgramStrategy`) supports these kinds of updates, but other clients
such as the language service might not always support modifying user files.

To accommodate this, the `TypeCheckingProgramStrategy` interface was
modified to include a `supportsInlineOperations` flag. If an implementation
specifies `false` for inline support, the template type-checking system will
return diagnostics on components which would otherwise require inline
operations.

Closes #38059

PR Close #38105
2020-07-29 10:31:20 -07:00
Andrea Canciani 9c8bc4a239 fix(common): narrow `NgIf` context variables in template type checker (#36627)
When the `NgIf` directive is used in a template, its context variables
can be used to capture the bound value. This is sometimes used in
complex expressions, where the resulting value is captured in a
context variable. There's two syntax forms available:

1. Binding to `NgIfContext.ngIf` using the `as` syntax:
```html
<span *ngIf="enabled && user as u">{{u.name}}</span>
```

2. Binding to `NgIfContext.$implicit` using the `let` syntax:
```html
<span *ngIf="enabled && user; let u">{{u.name}}</span>
```

Because of the semantics of `ngIf`, it is known that the captured
context variable is truthy, however the template type checker
would not consider them as such and still report errors when
`strict` is enabled.

This commit updates `NgIf`'s context guard to make the types of the
context variables truthy, avoiding the issue.

Based on https://github.com/angular/angular/pull/35125

PR Close #36627
2020-07-29 10:30:44 -07:00
Misko Hevery 2a45b932e2 build: fix broken build (#38274)
```
export const __core_private_testing_placeholder__ = '';
```
This API should be removed. But doing so seems to break `google3` and
so it requires a bit of investigation. A work around is to mark it as
`@codeGenApi` for now and investigate later.

PR Close #38274
2020-07-28 12:30:59 -07:00
Ajit Singh af80bdb470 fix(core): `Attribute` decorator `attributeName` is mandatory (#38131)
`Attribute` decorator has defined `attributeName` as optional but actually its
 mandatory and compiler throws an error if `attributeName` is undefined. Made
`attributeName` mandatory in the `Attribute` decorator to reflect this functionality

Fixes #32658

PR Close #38131
2020-07-28 11:17:24 -07:00
Igor Minar 5d3ba8dec7 test: update ts-api-guardian's strip_export_pattern to exclude Ivy instructions (#38224)
Previously the instructions were included in the golden files to monitor the frequency and rate of
the instruction API changes for the purpose of understanding the stability of this API (as it was
considered for becoming a public API and deployed to npm via generated code).

This experiment has confirmed that the instruction API is not stable enough to be used as public
API. We've since also came up with an alternative plan to compile libraries with the Ivy compiler
for npm deployment and this plan does not rely on making Ivy instructions public.

For these reasons, I'm removing the instructions from the golden files as it's no longer important
to track them.

The are three instructions that are still being included: `ɵɵdefineInjectable`, `ɵɵinject`, and
`ɵɵInjectableDef`.

These instructions are already generated by the VE compiler to support tree-shakable providers, and
code depending on these instructions is already deployed to npm. For this reason we need to treat
them as public api.

This change also reduces the code review overhead, because changes to public api golden files now
require multiple approvals.

PR Close #38224
2020-07-27 14:37:41 -07:00
Andrew Kushnir ad7046b934 feat(forms): AbstractControl to store raw validators in addition to combined validators function (#37881)
This commit refactors the way we store validators in AbstractControl-based classes:
in addition to the combined validators function that we have, we also store the original list of validators.
This is needed to have an ability to clean them up later at destroy time (currently it's problematic since
they are combined in a single function).

The change preserves backwards compatibility by making sure public APIs stay the same.
The only public API update is the change to the `AbstractControl` class constructor to extend the set
of possible types that it can accept and process (which should not be breaking).

PR Close #37881
2020-07-21 10:30:19 -07:00
Andrew Kushnir d72b1e44c6 refactor(core): rename synthetic host property and listener instructions (#37145)
This commit updates synthetic host property and listener instruction names to better align with other instructions.
The `ɵɵupdateSyntheticHostBinding` instruction was renamed to `ɵɵsyntheticHostProperty` (to match the `ɵɵhostProperty`
instruction name) and `ɵɵcomponentHostSyntheticListener` was renamed to `ɵɵsyntheticHostListener` since this
instruction is generated for both Components and Directives (so 'component' is removed from the name).
This PR is a followup after PR #35568.

PR Close #37145
2020-07-21 09:09:24 -07:00
Ajit Singh 3373453736 feat(common): add ReadonlyMap in place of Map in keyValuePipe (#37311)
ReadonlyMap is a superset of Map, in keyValuePipe we do not change the value of the object so ReadonlyPipe Works right in this case and we can accomodate more types. To accomodate more types added ReadonlyMap in Key Value pipe.

Fixes #37308

PR Close #37311
2020-07-20 14:13:15 -07:00
Judy Bogart 6c9401c338 docs: update router api documentation (#37980)
Edit descriptions, usage examples, and add links to be complete and consistent with API reference doc style

PR Close #37980
2020-07-16 13:52:40 -07:00
Misko Hevery 737506e79c fix(core): Allow modification of lifecycle hooks any time before bootstrap (#35464)
Currently we read lifecycle hooks eagerly during `ɵɵdefineComponent`.
The result is that it is not possible to do any sort of meta-programing
such as mixins or adding lifecycle hooks using custom decorators since
any such code executes after `ɵɵdefineComponent` has extracted the
lifecycle hooks from the prototype. Additionally the behavior is
inconsistent between AOT and JIT mode. In JIT mode overriding lifecycle
hooks is possible because the whole `ɵɵdefineComponent` is placed in
getter which is executed lazily. This is because JIT mode must compile a
template which can be specified as `templateURL` and those we are
waiting for its resolution.

- `+` `ɵɵdefineComponent` becomes smaller as it no longer needs to copy
  lifecycle hooks from prototype to `ComponentDef`
- `-` `ɵɵNgOnChangesFeature` feature is now always included with the
  codebase as it is no longer tree shakable.

Previously we have read lifecycle hooks from prototype in the
`ɵɵdefineComponent` so that lifecycle hook access would be monomorphic.
This decision was made before we had `T*` data structures. By not
reading the lifecycle hooks we are moving the megamorhic read form
`ɵɵdefineComponent` to instructions. However, the reads happen on
`firstTemplatePass` only and are subsequently cached in the `T*` data
structures. The result is that the overall performance should be same
(or slightly better as the intermediate `ComponentDef` has been
removed.)

- [ ] Remove `ɵɵNgOnChangesFeature` from compiler. (It will no longer
      be a feature.)
- [ ] Discuss the future of `Features` as they hinder meta-programing.

Fix #30497

PR Close #35464
2020-07-15 16:22:46 -07:00
Adam d37049a2a2 feat(platform-server): add option for absolute URL HTTP support (#37539)
In version 10.0.0-next.8, we introduced absolute URL support for
server-based HTTP requests, so long as the fully-resolved URL was
provided in the initial config. However, doing so represents a
breaking change for users who already have their own interceptors
to model this functionality, since our logic executes before all
interceptors fire on a request. See original PR #37071.

Therefore, we introduce a flag to make this change consistent with
v9 behavior, allowing users to opt in to this new behavior. This
commit also fixes two issues with the previous implementation:
1. if the server was initiated with a relative URL, the absolute
URL construction would fail because needed components were empty
2. if the user's absolute URL was on a port, the port would not
be included

PR Close #37539
2020-06-25 14:26:09 -07:00
Manduro 80d0067048 fix(router): `RouterLinkActive` should run CD when setting `isActive` (#21411)
When using the routerLinkActive directive inside a component that is using ChangeDetectionStrategy.OnPush and lazy loaded module routes the routerLinkActive directive does not update after clicking a link to a lazy loaded route that has not already been loaded.

Also the OnPush nav component does not set routerLinkActive correctly when the default route loads, the non-OnPush nav component works fine.

regression caused by #15943
closes #19934

PR Close #21411
2020-06-25 11:56:26 -07:00
Paul Gschwendtner 97dc85ba5e feat(core): support injection token as predicate in queries (#37506)
Currently Angular internally already handles `InjectionToken` as
predicates for queries. This commit exposes this as public API as
developers already relied on this functionality but currently use
workarounds to satisfy the type constraints (e.g. `as any`).

We intend to make this public as it's low-effort to support, and
it's a significant key part for the use of light-weight tokens as
described in the upcoming guide: https://github.com/angular/angular/pull/36144.

In concrete, applications might use injection tokens over classes
for both optional DI and queries, because otherwise such references
cause classes to be always retained. This was also an issue in View
Engine, but now with Ivy, this pattern became worse, as factories are
directly attached to retained classes (ultimately ending up in the
production bundle, while being unused).

More details in the light-weight token guide and in: https://github.com/angular/angular-cli/issues/16866.

Closes #21152. Related to #36144.

PR Close #37506
2020-06-11 13:21:11 -07:00
Andrew Scott cc1e0bbdc0 docs(router): add docs for RouterLink inputs (#37018)
The RouterLink and RouterLinkWithHref inputs do not have any docs. This comment adds jsdoc comments to the inputs.

PR Close #37018
2020-05-18 14:55:41 -07:00
Andrew Scott ef9f8df9ed fix(router): update type for routerLink to include null and undefined (#37018)
PR #13380 added support for `null` and `undefined` but the type on the parameter was not updated.
This would result in a compilation error if `fullTemplateTypeCheck` is enabled.
Fixes #36544

PR Close #37018
2020-05-18 14:55:41 -07:00
Alan Agius f97d8a9cbd refactor: `EventEmitter` to retain behaviour of pre TypeScript 3.9 (#36989)
TypeScript 3.9 introduced a breaking change where extends `any` no longer acts as `any`, instead it acts as `unknown`.

With this change we retain the behavior we had with TS 3.8 which is;

When using the `EventEmitter` as a type you must always provide a  type;
```ts
let emitter: EventEmitter<string>
```

and when initializing the `EventEmitter` class you can either provide a  type or or use the fallback type which is `any`

```ts
const emitter = new EventEmitter(); // EventEmitter<any>
const emitter = new EventEmitte<string>(); // EventEmitter<string>
``

PR Close #36989
2020-05-14 10:50:30 -07:00