- Emphasized the importance of using the same InjectionToken instance for both provider and injection call.
- Added examples to illustrate correct usages to prevent NullInjectorError.
PR Close#51386
This change aligns the settings between G3 and P3 as `REMOVE_STYLES_ON_COMPONENT_DESTROY_DEFAULT` is already set to `true` internally.
BREAKING CHANGE: `REMOVE_STYLES_ON_COMPONENT_DESTROY` default value is now `true`. This causes CSS of components to be removed from the DOM when destroyed. You retain the previous behaviour by providing the `REMOVE_STYLES_ON_COMPONENT_DESTROY` injection token.
```ts
import {REMOVE_STYLES_ON_COMPONENT_DESTROY} from '@angular/platform-browser';
...
providers: [{
provide: REMOVE_STYLES_ON_COMPONENT_DESTROY,
useValue: false,
}]
```
PR Close#51571
**Unit Testing Code Does Not Compile**
The compilation error is:
```
The compilation error is
./src/testing/http-client.spec.ts - Error: Module build failed (from ./node_modules/@ngtools/webpack/src/ivy/index.js):
Error: /home/projects/obk3vc--run/src/testing/http-client.spec.ts is missing from the TypeScript compilation. Please make sure it is in your tsconfig via the 'files' or 'include' property.
```
I’m not sure what to say about unit testing HTTP in a full Standalone app. Is it different?
_This is the only known remaining defect in this conversion of HTTP to Standalone._
**Edited content of `http-request-data-from-server.md`**
The current version of this page is confusing. In particular
* It tells readers they **should always unsubscribe** from the HttpClient method calls. This is *not true* and this example doesn't even do it. I replaced this instruction with more nuanced advice and an explanation of why it is OK to not unsubscribe to HttpClient methods.
* There is a "helpful" note about using the RxJS `map` operator to transform the response. This is *not "helpfulf"*. It is *confusing* because the sample doesn't use `map` anywhere. It was unnecessary here, even if it might be helpful elsewhere. I removed this note.
* The "Requesting a typed response" section seemed unclear to me, particularly because the guide begins with a `get` request that already has the `Config` return type specification. My revision attempts to make this more clear.
* The bold "callout" about the `observe` and `responseType` options appears out of nowhere after "Requesting a typed response". It's disconcerting at best. I moved it to the bottom of the page and linked to it from the `options` discussion at the top.
I made a few other revisions that I hope improve the readability of this page.
**Corrected `http-make-jsonp-request.md`**
The JSONP example, handwritten in the guide page, would not have compiled. I added one that does to `heroes.service.ts` and displayed it on this page.
**Corrected `http-handle-request-errors.md`
This page ended with a section called "Sending data to a server" that introduces PUT, POST, and DELETE. These features have nothing to do with error handling and the verbiage here duplicates the opening paragraphs of the next topic which does: "Send data to a server". So I deleted this section from the error handling guide page.
**Archived http-setup-server-communication.md**
`http-setup-server-communication.md` appears to be the original long document that has since been divided over the other pages in this folder.
It shouldn’t be in the reader’s flow. I did update it for Standalone. But I also removed it from left-nav and marked as archived.
PR Close#51400
This commit removes deprecated properties on the Router. These are meant
to be configured through DI and not meant to be changed during runtime.
BREAKING CHANGE: The following Router properties have been removed from
the public API:
- canceledNavigationResolution
- paramsInheritanceStrategy
- titleStrategy
- urlUpdateStrategy
- malformedUriErrorHandler
These should instead be configured through the `provideRouter` or
`RouterModule.forRoot` APIs.
PR Close#51502
The localize package intentionally duplicates some logic from the
compiler to avoid adding a dependency. This is now an error in the
packaging rule to prevent common pitfalls/code duplication. Here it's
an explicit decision though so we mark it as such and ask for the check
to be ignored for the particular import.
PR Close#51500
The upgrade package duplicaes some of code due to relative
imports between entry-points. This caused bundlers to
inline shared functions twice in both FESM outputs.
This is an acceptable limitation and we are not changing this
because the primary entry-point is not synced into G3. It's non-trivial
to remove these cross relative imports right now because the primary
entry-point is not even built in G3 so instead we just ignore the
relative imports using a re-export file.
Note: To simplify this change, we continue using namespace exports
as exporting individual named exports for all these possible usages
is rather cumbersome and also we had existing namespace imports for
e.g. `angular1.ts`. The code of upgrade is rarely edited these days
PR Close#51500
The common packages were duplicating a little bit of code due
to relative imports between entry-points. This caused bundlers to
inline shared functions twice in both FESM outputs.i
PR Close#51500
The animations packages were duplicating a little bit of code due
to relative imports between entry-points. This caused bundlers to
inline shared functions twice in both FESM outputs.
PR Close#51500
We were collecting all ES2022 files from entry-points (including
transitive files). Those are later on combined and filtered so that
we know which files to copy over to the package. There was no
deduping here. This did not have an effect, but could be a source
of slowness in `ng_package` and also breaks validation checks which
could show same errors multiple times for the same file.
PR Close#51500
Fixes that there was code duplication between the primary entry-point,
the testing entry-point and the rxjs-interop entry-point.
This code duplication resulted in additional code size (really
neglibible here because rxjs-interop did not duplicate large parts of
core, and `testing` is not used in production).
On the other hand though, the duplication resulted in a subtle JIT
dependency tracking issue due to the `depsTracker` no longer being a
singleton. This caused test failures as in:
https://github.com/angular/angular/pull/51415.
PR Close#51500
Introduces a check into `ng_package` that will ensure that there are no
cross entry-point or cross-package relative imports that would end up
contributing to duplicate code. Not only would duplicate code result in
size increases, but also it could cause subtle hard-to-debug bugs,
especially when cross imports rely on e.g. singletons. Like for example
the deps tracker that is used in angular/core but also in
angular/core/testing.
PR Close#51500
**Pipes Sample Code**
Migrated all sample code in the `examples/pipes` folder. Did not touch the pipes in the ToH or Testing folders.
>The existing, complex discussion of the `CurrencyPipe` within `pipes-transform-data.md` cried out for a new `concurrency-formatting.component` example`.
**Extracted "pipe precedence" into its own page**
The topic had been extracted from `pipe.md` and tacked on to the bottom of the `pipes-overview.md` page.
It's an advanced and somewhat obscure topic that doesn't belong in the overview. Rather than throw it away, I created a new `pipe-precedence.md` page and added it to the bottom of the pipes section navigation.
I also tried to improve both the guide text and the companion component, `precedence.component`.
**How to create a pipe is missing**
The readers are told they can create their own pipes in several places throughout the docs. But there are no links and you can't navigate to a page that covers the topic. This is a serious omission!
The topic is introduced in the `pipes-custom-data-trans.md` page (extracted verbatim from `pipes.md`). But you can't navigate to this page and their are no links to it.
TODO: restore this page and add it to the left-nav.
**Change `pipes.md` references to `pipe-overview.md`**
The original, kitchen-sink page, `pipes.md`, was disconnected from navigation long ago, in favor of multiple pages such as `pipe-overview.md`. The page is still in the AIO documentation and can be found by searching or by links from 3rd party documenters. Landing on that page hides the left-nav.
In this commit, we treat `pipes.md` as deprecated (which it seems to be). Therefore, this commit retargets previous `pipes.md` references to `pipe-overview.md`.
>The `change-detection-slow-computations.md` is the exception. It refers to "pure pipes", a subject not covered in the current pipe documentation. That reference is retargeted to `api/core/Pipe#pure`.
Certain code files are only referenced in `pipe.md`. They still work and are displayed in the overall pipes code sample as before. Now they are marked with deprecation comments for future treatment or removal.
For consistency, certain sections of `pipes.md` were replaced by the contents of the corresponding current pages.
PR Close#51333
With the new control flow and defer blocks it'll be common for several template instructions to be declare one after another. These changes add support for chaining to the `template` instruction which will allow us to save some bytes.
PR Close#51546
Adds the initial implementation to generate the instructions for the `for` loop block.
**Note:** the expressions we support in the `track` paramateter are currently limited to tracking by identity or index, or a specific property of the item. Supporting more advanced expression will require additional work that I'll do in a follow-up PR.
PR Close#51514
This commit fixes an issue where serialization of a view container fails in case it uses a component host as an anchor. This fix is similar to the fix from #51247, but for cases when we insert a component (that acts as a host for a view container) deeper in a hierarchy.
Resolves#51318.
PR Close#51456
`switch` blocks are part of the new control flow syntax. This commit adds support for processing them, and emitting the appropriate templates and conditional instructions.
PR Close#51518
This commit adds runtime implementation of a basic preloading mechanism for defer blocks. The base prefetching logic invokes a dependency loading function (generated by the compiler) when a corresponding `prefetch when` condition is triggered. The `prefetch on` triggers would be implemented in followup PRs.
We plan to explore additional prefetching techniques and will followup with more PRs later (based on the research).
PR Close#51529
