These tests were previously not running on CI so they have always been broken,
or got broken just recently :-(.
test(ivy): mark failing test targets with fixme-ivy-jit and fixme-ivy-local tags
PR Close#26735
Comment nodes that are child nodes of unsafe elements are identified as text nodes. This results in the comment node being returned as an encoded string.
Add a check to ignore such comment nodes.
PR Close#25879
We are close enough to blacklist a few test targets, rather than whitelist targets to run...
Because bazel rules can be composed of other rules that don't inherit tags automatically,
I had to explicitly mark all of our ts_library and ng_module targes with "ivy-local" and
"ivy-jit" tags so that we can create a query that excludes all fixme- tagged targets even
if those targets are composed of other targets that don't inherit this tag.
This is the updated overview of ivy related bazel tags:
- ivy-only: target that builds or runs only under ivy
- fixme-ivy-jit: target that doesn't yet build or run under ivy with --compile=jit
- fixme-ivy-local: target that doesn't yet build or run under ivy with --compile=local
- no-ivy-jit: target that is not intended to build or run under ivy with --compile=jit
- no-ivy-local: target that is not intended to build or run under ivy with --compile=local
PR Close#26471
Originally, the ivy_switch mechanism used Bazel genrules to conditionally
compile one TS file or another depending on whether ngc or ngtsc was the
selected compiler. This was done because we wanted to avoid importing
certain modules (and thus pulling them into the build) if Ivy was on or
off. This mechanism had a major drawback: ivy_switch became a bottleneck
in the import graph, as it both imports from many places in the codebase
and is imported by many modules in the codebase. This frequently resulted
in cyclic imports which caused issues both with TS and Closure compilation.
It turns out ngcc needs both code paths in the bundle to perform the switch
during its operation anyway, so import switching was later abandoned. This
means that there's no real reason why the ivy_switch mechanism needed to
operate at the Bazel level, and for the ivy_switch file to be a bottleneck.
This commit removes the Bazel-level ivy_switch mechanism, and introduces
an additional TypeScript transform in ngtsc (and the pass-through tsc
compiler used for testing JIT) to perform the same operation that ngcc
does, and flip the switch during ngtsc compilation. This allows the
ivy_switch file to be removed, and the individual switches to be located
directly next to their consumers in the codebase, greatly mitigating the
circular import issues and making the mechanism much easier to use.
As part of this commit, the tag for marking switched variables was changed
from __PRE_NGCC__ to __PRE_R3__, since it's no longer just ngcc which
flips these tags. Most variables were renamed from R3_* to SWITCH_* as well,
since they're referenced mostly in render2 code.
Test strategy: existing test coverage is more than sufficient - if this
didn't work correctly it would break the hello world and todo apps.
PR Close#26550
This commit adds an ngTemplateGuard_ngIf static method to the NgIf
directive and an ngTemplateContextGuard static method to NgFor. The
function of these two static methods is to enable type narrowing
within generated type checking code for consumers of the directives.
PR Close#26203
Previously in Ivy, metadata for directives/components/modules/etc was
carried in .d.ts files inside type information encoded on the
DirectiveDef, ComponentDef, NgModuleDef, etc types of Ivy definition
fields. This works well, but has the side effect of complicating Ivy's
runtime code as these extra generic type parameters had to be specified
as <any> throughout the codebase. *DefInternal types were introduced
previously to mitigate this issue, but that's the wrong way to solve
the problem.
This commit returns *Def types to their original form, with no metadata
attached. Instead, new *DefWithMeta types are introduced that alias the
plain definition types and add extra generic parameters. This way the
only code that needs to deal with the extra metadata parameters is the
compiler code that reads and writes them - the existence of this metadata
is transparent to the runtime, as it should be.
PR Close#26203
Create getter methods `getXXXDef` for each definition which
uses `hasOwnProperty` to verify that we don't accidently read form the
parent class.
Fixes: #24011Fixes: #25026
PR Close#25736
Various user code uses 'instanceof' to check whether a particular instance
is a TemplateRef, ElementRef, etc. Ivy needs to work with these checks.
PR Close#25775
defineComponent() and friends can return a flyweight EMPTY object for
specific fields when they contain no data. InheritDefinitionFeature
was attempting to write into these flyweight objects, which have been
protected with Object.freeze().
This commit adds detection to InheritDefinitionFeature to identify the
frozen objects.
PR Close#25755
By pulling in `compiler` into `core` the `compiler` was not
100% tree-shakable and about 8KB of code was retained
when tree-shaken with closure.
PR Close#25531
Provides a runtime and compile time switch for ivy including
`ApplicationRef.bootstrapModule`.
This is done by naming the symbols such that `ngcc` (angular
Compatibility compiler) can rename symbols in such a way that running
`ngcc` command will switch the `@angular/core` module from `legacy` to
`ivy` mode.
This is done as follows:
```
const someToken__PRE_NGCC__ = ‘legacy mode’;
const someToken__POST_NGCC__ = ‘ivy mode’;
export someSymbol = someToken__PRE_NGCC__;
```
The `ngcc` will search for any token which ends with `__PRE_NGCC__`
and replace it with `__POST_NGCC__`. This allows the `@angular/core`
package to be rewritten to ivy mode post `ngcc` execution.
PR Close#25238
When @angular/core is compiled by ngtsc, a factory file is generated
for ApplicationModule, that is currently invalid because r3_symbols
does not export NgModuleFactory. This change fixes that issue and
ensures the generated ngfactory file for @angular/core is valid.
PR Close#25392
This commit creates an API for factory functions which allows them
to be inherited from one another. To do so, it differentiates between
the factory function as a wrapper for a constructor and the factory
function in ngInjectableDefs which is determined by a default
provider.
The new form is:
factory: (t?) => new (t || SomeType)(inject(Dep1), inject(Dep2))
The 't' parameter allows for constructor inheritance. A subclass with
no declared constructor inherits its constructor from the superclass.
With the 't' parameter, a subclass can call the superclass' factory
function and use it to create an instance of the subclass.
For @Injectables with configured providers, the factory function is
of the form:
factory: (t?) => t ? constructorInject(t) : provider();
where constructorInject(t) creates an instance of 't' using the
naturally declared constructor of the type, and where provider()
creates an instance of the base type using the special declared
provider on @Injectable.
PR Close#25392
data about tasks.
When building a list of pending tasks for callers of whenStable(),
Testability will copy data about the task into a new object, in order to
avoid leaking references to tasks.
This change copies more properties from Tasks into the list of pending
tasks, as well as a reference to Task.data to give callers more
information about the tasks that are pending.
Specifically, this also copies runCount and task ID, which are needed in
order for callers to know when a given task is repeating.
PR Close#25010
To match the View Engine behavior.
We should make this configurable so that the node injector is tree shaken when
directives do not need to be published.
PR Close#25291
This commit adds basic support for <ng-container> - most of the
functionality should work as long as <ng-container> is a child of
a regular element.
PR Close#25227
Update XMB placeholders(<ph>) to include the original value on top of an
example. Placeholders can by definition have one example(<ex>) tag and a
text node. The text node is used by TC as the "original" value from the
placeholder, while the example should represent a dummy value.
For example: <ph name="PET"><ex>Gopher</ex>{{ petName }}</ph>.
This change makes sure that we have the original text, but it *DOES NOT*
make sure that the example is correct. The example has the same wrong
behavior of showing the interpolation text rather than a useful
example.
No breaking changes, but tools that depend on the previous behavior and
don't consider the full XMB definition may fail to parse the XMB.
Fixes b/72565847
PR Close#25079
- `directiveInjector()` is used to inject anything in the directive / component
/ pipe factories so adding `InjectionToken<T>` as a supported token type.
- `getOrCreateInjectable()` should search first in the node injector tree and
then in the module injector tree (was either or before the PR).
PR Close#25166
Fixes#25018.
Instantiating a NgModuleRef from NgModuleFactory reuses the NgModuleDefinition if it is already present. However the NgModuleDefinition has a providers array which modified when tree shakable providers are instantiated. This corrupts the provider definitions the next time the same factory is used to create a new NgModuleRef - Two provider definitions can end up with the same index anf the injector could potentially return a completely wrong object for a provider token.
This scenario is more likely on the server where the same NgModuleFactory is reused across requests.
The fix clones the cached NgModuleDefinition so that any tree shakable providers added later do not affect the cached copy.
PR Close#25022
Previously ngtsc would use a tuple of class types for listing metadata
in .d.ts files. For example, an @NgModule's declarations might be
represented with the type:
[NgIf, NgForOf, NgClass]
If the module had no declarations, an empty tuple [] would be produced.
This has two problems.
1. If the class type has generic type parameters, TypeScript will
complain that they're not provided.
2. The empty tuple type is not actually legal.
This commit addresses both problems.
1. Class types are now represented using the `typeof` operator, so the
above declarations would be represented as:
[typeof NgIf, typeof NgForOf, typeof NgClass].
Since typeof operates on a value, it doesn't require generic type
arguments.
2. Instead of an empty tuple, `never` is used to indicate no metadata.
PR Close#24862
Previously, some of the *Def symbols were not exported or were exported
as public API. This commit ensures every definition type is in the
private export namespace.
PR Close#24862
ngInjectorDef.imports is generated from @NgModule.imports plus
@NgModule.exports. A problem arises as a result, because @NgModule
exports contain not only other modules (which will have ngInjectorDef
fields), but components, directives, and pipes as well. Because of
locality, it's difficult for the compiler to filter these out at
build time.
It's not impossible, but for now filtering them out at runtime will
allow testing of the compiler against complex applications.
PR Close#24862
On accident a few of the definition types were emitted as public API
symbols. Much of the Ivy API surface is still prefixed with ɵ,
indicating it's a private API. The definition types should be private
for now.
PR Close#24738
Tree shakable providers use the APP_ROOT token to determine where to attach themselves. APP_ROOT gets set on NgModule with BrowserModule irrespective of whether it is actually the root(Ex. in case of SSR app where the shell app is first bootstrapped without BrowserModule being the root module).
This change allows a NgModule with BrowserModule to explicitly mark itself as not the root by setting APP_ROOT token to false. This allows tree shakable providers to be attached to the right rott module.
PR Close#24814
With these changes, the types are a little stricter now and also not
compatible with Protractor's jasmine-like syntax. So, we have to also
use `@types/jasminewd2` for e2e tests (but not for non-e2e tests).
I also had to "augment" `@types/jasminewd2`, because the latest
typings from [DefinitelyTyped][1] do not reflect the fact that the
`jasminewd2` version (v2.1.0) currently used by Protractor supports
passing a `done` callback to a spec.
[1]: 566e039485/types/jasminewd2/index.d.ts (L9-L15)Fixes#23952Closes#24733
PR Close#19904
This updates the r3_pipe_compiler to not depend on global analysis,
and to produce ngPipeDef instructions in the same way that the other
compilers do. It's a precursor to JIT and AOT implementations of
@Pipe compilation.
PR Close#24703
This propagates other custom equality testers added by users. Additionally, if
an Angular project is using jasmine 2.6+, it will allow Jasmine's custom object
differ to print out pretty test error messages.
fixes#22939
PR Close#22983
- Adds InheritanceDefinitionFeature to ivy
- Ensures that lifecycle hooks are inherited from super classes whether they are defined as directives or not
- Directives cannot inherit from Components
- Components can inherit from Directives or Components
- Ensures that Inputs, Outputs, and Host Bindings are inherited
- Ensures that super class Features are run
PR Close#24570
Used to resolve resource URLs on `@Component` when used with JIT compilation.
```
@Component({
selector: 'my-comp',
templateUrl: 'my-comp.html', // This requires asynchronous resolution
})
class MyComponnent{
}
// Calling `renderComponent` will fail because `MyComponent`'s `@Compenent.templateUrl`
// needs to be resolved because `renderComponent` is synchronous process.
// renderComponent(MyComponent);
// Calling `resolveComponentResources` will resolve `@Compenent.templateUrl` into
// `@Compenent.template`, which would allow `renderComponent` to proceed in synchronous manner.
// Use browser's `fetch` function as the default resource resolution strategy.
resolveComponentResources(fetch).then(() => {
// After resolution all URLs have been converted into strings.
renderComponent(MyComponent);
});
```
PR Close#24637
Previously the todo app imported reflect-metadata, since it is a dependency
of JIT and the todo app tests run in both JIT and AOT modes. However, the
code doesn't get tree-shaken away in AOT mode.
This change adds a target //packages/core/test/bundling/util:reflect_metadata
which, depending on whether the compile flag is in JIT or AOT mode, either
includes reflect-metadata or is a no-op.
Not including reflect-metadata gets the compressed todo bundle down to 12.5 kB.
PR Close#24677
ngtsc is sufficiently capable now that it can compile hello_world
and todo and achieve equivalent results to ngc in ivy (global) mode.
Bundle sizes:
hello_world - 3.0 kB
todo - 14.7 kB
PR Close#24677
This change makes @angular/compiler more tree-shakeable by changing
an enum to a const enum and by getting rid of a top-level map that
the tree-shaker was seeing as a reference which caused r3_identifiers
to be retained.
This drops a few hundred bytes of JS from tree-shaken ngtsc compiled
apps.
PR Close#24677
Previously the repo was depending on an old version of build optimizer.
This change updates to the latest (an RC release in the CLI package).
Additionally, this changes the behavior of ng_rollup_bundle to apply
the optimizer to ngtsc compiled code, and configures it to treat the
@angular/compiler package as side-effect-free.
This results in a substantial size reduction of ngtsc compiled code.
PR Close#24677
This commit takes advantage of the @angular/compiler work for ngInjectorDef
in AOT mode in order to generate the same definition in JIT mode.
PR Close#24632
All errors for existing fields have been detected and suppressed with a
`!` assertion.
Issue/24571 is tracking proper clean up of those instances.
One-line change required in ivy/compilation.ts, because it appears that
the new syntax causes tsickle emitted node to no longer track their
original sourceFiles.
PR Close#24572
This change supports compilation of components, directives, and modules
within ngtsc. Support is not complete, but is enough to compile and test
//packages/core/test/bundling/todo in full AOT mode. Code size benefits
are not yet achieved as //packages/core itself does not get compiled, and
some decorators (e.g. @Input) are not stripped, leading to unwanted code
being retained by the tree-shaker. This will be improved in future commits.
PR Close#24427
clang-format (on mac) has taken a disliking to this particular line, and
rewrites one of the ɵ characters to an invalid Unicode sequence.
PR Close#24479
At runtime in JIT mode, when the compiler writes a reference to a symbol that symbol
is resolved through a symbol table named angularCoreEnv in render3/jit/environment.
Previously, this symbol table was not kept up-to-date with the Ivy instruction set
and the names of symbols the compiler could reference.
This change brings the symbol table in sync, and also adds a test that verifies every
symbol the compiler can reference is available at runtime in the symbol table.
PR Close#24479
When creating content queries from a directive on an element we need to take into account
existing view queries. The same element can be reported to both content and view queries
so freshly created content queries must be combined with pre-existing view queries.
PR Close#24507
NOTE: This does NOT add parsing of namespaced attributes
- Adds AttributeMarker for namespaced attributes
- Adds test for namespaced attributes
- Updates AttributeMarker enum to use CamelCase, and not UPPER_CASE names
PR Close#24386
Previously, the transitive scopes of an NgModuleDef were computed
during execution of the @NgModule decorator. This meant that JIT-
compiled modules could only import other JIT-compiled modules, as
the import mechanism relied on the calculation of transitive scopes
to already have happened for the imported module.
This change moves computation of transitive scopes to a function
`transitiveScopesFor` (and makes it lazy). This opens the door for
AOT -> JIT or JIT -> AOT imports, as transitive scopes for AOT
modules can be calculated when needed by JIT, and AOT modules can
also write expressions that call `transitiveScopesFor` when
importing a JIT-compiled module.
PR Close#24334
Two new CircleCI environments are created: test_ivy_jit and test_ivy_aot.
Both run a subset of the tests that have been marked with Bazel tags as
being appropriate for that environment.
Once all the tests pass, builds are published to the *-builds repo both
for the legacy View Engine compiled code as well as for ivy-jit and ivy-aot.
PR Close#24309
This patch ensures that if a list of nodes (that contain
animation triggers) are moved around then they will retain their
trigger-value state when animated again at a later point.
PR Close#24238
`NgForOf` used to implement `OnChanges` and than use
`ngOnChanges` callback to detect when `ngForOf` binding
changed to update the differ. We now do the checking
manually which puts less pressure on the runtime to do
the bookkeeping and should result in minor perf improvement.
PR Close#23378
After #24113 there is 2 `TNode` in those tests:
- 1 for the host,
- 1 for the text node.
The PR #23924 status was green because it branched off master before #24113 was
merged in.
PR Close#24208
This PR tackles a simple case where ViewRef definition point (<ng-template>) is the
same as the insertion point (ViewContainerRef requested on the said <ng-template>).
For this particular case we can assume that we know a container into which a given
view will be inserted when a view is created. This is not true fall all the possible
cases so follow-up PR will be needed to extend this basic implementation.
PR Close#24179
This commit builds out enough of the JIT compiler to render
//packages/core/test/bundling/todo, and allows the tests to run in
JIT mode.
To play with the app, run:
bazel run --define=compile=jit //packages/core/test/bundling/todo:prodserver
PR Close#24138
In ngIvy directives matching (determining which directives are active based
on a CSS seletor) happens at runtime. This means that runtime needs to have
enough context to match directives. This PR takes care of cases where a directive's
selector should match bindings (ex. [foo]="exp") and event handlers (ex. (out)="do()").
In the mentioned cases we need to have binding / output "attributes" for directive's
CSS selector matching purposes. At the same time those are not regular attributes and
as such should not be reflected in the DOM.
Closes#23706
PR Close#23991
Bazel has a restriction that a single output (eg. a compiled version of
//packages/common) can only be produced by a single rule. This precludes
the Angular repo from having multiple rules that build the same code. And
the complexity of having a single rule produce multiple outputs (eg. an
ngc-compiled version of //packages/common and an Ivy-enabled version) is
too high.
Additionally, the Angular repo has lots of existing tests which could be
executed as-is under Ivy. Such testing is very valuable, and it would be
nice to share not only the code, but the dependency graph / build config
as well.
Thus, this change introduces a --define flag 'compile' with three potential
values. When --define=compile=X is set, the entire build system runs in a
particular mode - the behavior of all existing targets is controlled by
the flag. This allows us to reuse our entire build structure for testing
in a variety of different manners. The flag has three possible settings:
* legacy (the default): the traditional View Engine (ngc) build
* local: runs the prototype ngtsc compiler, which does not rely on global
analysis
* jit: runs ngtsc in a mode which executes tsickle, but excludes the
Angular related transforms, which approximates the behavior of plain
tsc. This allows the main packages such as common to be tested with
the JIT compiler.
Additionally, the ivy_ng_module() rule still exists and runs ngc in a mode
where Ivy-compiled output is produced from global analysis information, as
a stopgap while ngtsc is being developed.
PR Close#24056
Short-circuitable expressions (using ternary & binary operators) could not use
the regular binding mechanism as it relies on the bindings being checked every
single time - the index is incremented as part of checking the bindings.
Then for pure function kind of bindings we use a different mechanism with a
fixed index. As such short circuiting a binding check does not mess with the
expected binding index.
Note that all pure function bindings are handled the same wether or not they
actually are short-circuitable. This allows to keep the compiler and compiled
code simple - and there is no runtime perf cost anyway.
PR Close#24039
This commit adds a mechanism by which the @angular/core annotations
for @Component, @Injectable, and @NgModule become decorators which,
when executed at runtime, trigger just-in-time compilation of their
associated types. The activation of these decorators is configured
by the ivy_switch mechanism, ensuring that the Ivy JIT engine does
not get included in Angular bundles unless specifically requested.
PR Close#23833
Previously, ngOnDestroy was only called on services which were statically
determined to have ngOnDestroy methods. In some cases, such as with services
instantiated via factory functions, it's not statically known that the service
has an ngOnDestroy method.
This commit changes the runtime to look for ngOnDestroy when instantiating
all DI tokens, and to call the method if it's present.
Fixes#22466Fixes#22240Fixes#14818
PR Close#23755
Prior to this patch, if an element is queried and animated for 0 seconds
(just a style() call and nothing else) then the styles applied would not
be properly cleaned up due to their camelCased nature.
PR Close#23633
Fix a corner case where eager providers were getting constructed twice if the provider was requested before the initialization of the NgModule is complete.
PR Close#23559
The bug fixed here steams from the fact that we are traversing too far up
in the views tree hierarchy in the destroyViewTree function.
The logic in destroyViewTree is off if we start removal at an embedded view
without any child views. For such a case we should just clean up (cleanUpView)
this one view without paying attention to next / parent views.
PR Close#23482
Ivy definition looks something like this:
```
class MyService {
static ngInjectableDef = defineInjectable({
…
});
}
```
Here the argument to `defineInjectable` is well known public contract which needs
to be honored in backward compatible way between versions. The type of the
return value of `defineInjectable` on the other hand is private and can change
shape drastically between versions without effecting backwards compatibility of
libraries publish to NPM. To our users it is effectively an opaque token.
For this reson why declare the return value of `defineInjectable` as `never`.
PR Close#23383
Ivy definition looks something like this:
```
class MyService {
static ngInjectableDef = defineInjectable({
…
});
}
```
Here the argument to `defineInjectable` is well known public contract which needs
to be honored in backward compatible way between versions. The type of the
return value of `defineInjectable` on the other hand is private and can change
shape drastically between versions without effecting backwards compatibility of
libraries publish to NPM. To our users it is effectively an `OpaqueToken`.
By prefixing the type with `ɵ` we are communicating the the outside world that
the value is not public API and is subject to change without backward compatibility.
PR Close#23371
- Remove default injection value from `inject` / `directiveInject` since
it is not possible to set using annotations.
- Module `Injector` is stored on `LView` instead of `LInjector` data
structure because it can change only at `LView` level. (More efficient)
- Add `ngInjectableDef` to `IterableDiffers` so that existing tests can
pass as well as enable `IterableDiffers` to be injectable without
`Injector`
PR Close#23345
This change changes:
- compiler uses `directiveInject` instead of `inject` for `Directive`s
- unifies the flags in `di` as well as `render3`
- changes the signature of `directiveInject` to match `inject` In prep for #23330
- compiler now generates flags for injection.
Compiler portion of #23342
Prep for #23330
PR Close#23345
When compiling templates the compiler would often bind to
closest context rather than the component context.
The only time one should be binding to the cont component is
in explicit cases where the inner template declares local variable.
PR Close#23168
As we no longer create native (RNode) comment nodes for containers,
we need to execute logic for finding a next sibiling node with RNode
when inserting a view.
The mentioned logic need to be updated for the case of dynamically
created containers (LContainerNode). Indeed, we need to be able to
descend into dynamically inserted views while looking for a RNode.
To achieve this we need to have a pointer from a host LNode to a
dynamically created LContainerNode).
PR Close#23193
Pawel Kozlowski
Kara Erickson
Andrew Kushnir
Matias Niemelä
Igor Minar
Shino Kurian
Marc Laval
Alex Rickabaugh
Greg Magolan
Miško Hevery
Ben Lesh
Adam Plumer
George Kalpakas
Alex Eagle
Victor Berchet
Michael Giambalvo
Aldo Roman Nureña
Aldo Roman
Carlos Ortiz Garcia
Olivier Combe
Vikram Subramanian
Kevin Villela
Rado Kirov
Georgi Parlakov
Jason Aden
Alan Agius
JoostK
swseverance
Rafael
Joey Perrott