There is a bug in the existing handling for cross-file references.
Suppose there are two files, module.ts and component.ts.
component.ts declares two components, one of which uses the other.
In the Ivy model, this means the component will get a directives:
reference to the other in its defineComponent call.
That reference is generated by looking at the declared components
of the module (in module.ts). However, the way ngtsc tracks this
reference, it ends up comparing the identifier of the component
in module.ts with the component.ts file, detecting they're not in
the same file, and generating a relative import.
This commit changes ngtsc to track all identifiers of a reference,
including the one by which it is declared. This allows toExpression()
to correctly decide that a local reference is okay in component.ts.
PR Close#25080
When ngtsc encounters a reference to a type (for example, a Component
type listed in an NgModule declarations array), it traces the import
of that type and attempts to determine the best way to refer to it.
In the event the type is defined in the same file where a reference
is being generated, the identifier of the type is used. If the type
was imported, ngtsc has a choice. It can use the identifier from the
original import, or it can write a new import to the module where the
type came from.
ngtsc has a bug currently when it elects to rely on the user's import.
When writing a .d.ts file, the user's import may have been elided as
the type was not referred to from the type side of the program. Thus,
in .d.ts files ngtsc must always assume the import may not exist, and
generate a new one.
In .js output the import is guaranteed to still exist, so it's
preferable for ngtsc to continue using the existing import if one is
available.
This commit changes how @angular/compiler writes type definitions, and
allows it to use a different expression to write a type definition than
is used to write the value. This allows ngtsc to specify that types in
type definitions should always be imported. A corresponding change to
the staticallyResolve() Reference system allows the choice of which
type of import to use when generating an Expression from a Reference.
PR Close#25080
@ContentChild[ren] and @ViewChild[ren] can contain a forwardRef() to a
type. This commit allows ngtsc to unwrap the forward reference and
deal with the node inside.
It includes two modes of support for forward reference resolution -
a foreign function resolver which understands deeply nested forward
references in expressions that are being statically evaluated, and
an unwrapForwardRef() function which deals only with top-level nodes.
Both will be useful in the future, but for now only unwrapForwardRef()
is used.
PR Close#25080
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
This commit moves the compiler compliance tests into compiler-cli,
and uses ngtsc to run them instead of the custom compilation
pipeline used before. Testing against ngtsc allows for validation
of the real compiler output.
This commit also fixes a few small issues that prevented the tests
from passing.
PR Close#24862
This change adds support for host bindings to ngtsc, and parses them
both from decorators and from the metadata in the top-level annotation.
PR Close#24862
@NgModule()s get compiled to two fields: ngModuleDef and ngInjectorDef.
Both fields contain imports, as both selector scopes and injectors have
the concept of composed units of configuration. Previously these fields
were generated by static resolution of imports and exports in metadata.
Support for ModuleWithProviders requires they be generated differently.
ngModuleDef's imports/exports are generated as resolved lists of types,
whereas ngInjectorDef's imports should reflect the raw expressions that
the developer wrote in the metadata.
This change modifies the NgModule handler and properly copies raw nodes
for the imports and exports into the ngInjectorDef.
PR Close#24862
Previously ngtsc had a few bugs handling special token types:
* Injector was not properly translated to INJECTOR
* ChangeDetectorRef was not injected via injectChangeDetectorRef()
This commit fixes these two bugs, and also adds a test to ensure
they continue to work correctly.
PR Close#24862
Within an @NgModule it's common to include in the imports a call to
a ModuleWithProviders function, for example RouterModule.forRoot().
The old ngc compiler was able to handle this pattern because it had
global knowledge of metadata of not only the input compilation unit
but also all dependencies.
The ngtsc compiler for Ivy doesn't have this knowledge, so the
pattern of ModuleWithProviders functions is more difficult. ngtsc
must be able to determine which module is imported via the function
in order to expand the selector scope and properly tree-shake
directives and pipes.
This commit implements a solution to this problem, by adding a type
parameter to ModuleWithProviders through which the actual module
type can be passed between compilation units.
The provider side isn't a problem because the imports are always
copied directly to the ngInjectorDef.
PR Close#24862
for non-inline templates
- Non-inline templates used to ouput the path to the component TS file
instead of the path to the original HTML file.
- Inline templates keep the same behavior.
Fixes#24884
PR Close#24885
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
The current module resolution simply attaches .ts to the import/export path, which does
not work if the path is using Node / CommonJS behavior to resolve to an index.ts file.
This patch uses typescript's module resolution logic, and will attempt to load the original
typescript file if this resolution returns a .js or .d.ts file
PR Close#22856
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 commit adds support for templateUrl in component templates within
ngtsc. The compilation pipeline is split into sync and async versions,
where asynchronous compilation invokes a special preanalyze() phase of
analysis. The preanalyze() phase can optionally return a Promise which
will delay compilation until it resolves.
A ResourceLoader interface is used to resolve templateUrls to template
strings and can return results either synchronously or asynchronously.
During sync compilation it is an error if the ResourceLoader returns a
Promise.
Two ResourceLoader implementations are provided. One uses 'fs' to read
resources directly from disk and is chosen if the CompilerHost doesn't
provide a readResource method. The other wraps the readResource method
from CompilerHost if it's provided.
PR Close#24704
This change generates ngInjectorDef as well as ngModuleDef for @NgModule
annotated types, reflecting the dual nature of @NgModules as both compilation
scopes and as DI configuration containers.
This required implementing ngInjectorDef compilation in @angular/compiler as
well as allowing for multiple generated definitions for a single decorator in
the core of ngtsc.
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
Previously, the compileComponent() and compileDirective() APIs still required
the output of global analysis, even though they only read local information
from that output.
With this refactor, compileComponent() and compileDirective() now define
their inputs explicitly, with the new interfaces R3ComponentMetadata and
R3DirectiveMetadata. compileComponentGlobal() and compileDirectiveGlobal()
are introduced and convert from global analysis output into the new metadata
format.
This refactor also splits out the view compiler into separate files as
r3_view_compiler_local.ts was getting unwieldy.
Finally, this refactor also splits out generation of DI factory functions
into a separate r3_factory utility as the logic is utilized between different
compilers.
PR Close#23545
This commit adds a new compiler pipeline that isn't dependent on global
analysis, referred to as 'ngtsc'. This new compiler is accessed by
running ngc with "enableIvy" set to "ngtsc". It reuses the same initialization
logic but creates a new implementation of Program which does not perform the
global-level analysis that AngularCompilerProgram does. It will be the
foundation for the production Ivy compiler.
PR Close#23455
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
rxjs 6.0.0 breaks strictMetadataEmit as they now publish a .d.ts file with a
structure like:
declare export class Subscription {
static EMPTY: Subscription;
}
This generates metadata which contains an error, and fails the strictMetadataEmit
validation. There is nothing a library author can do in this situation except to
set strictMetadataEmit to false.
The spirit of strictMetadataEmit is to validate that the author's library doesn't
do anything that will break downstream users. This failure is a corner case which
causes more harm than good, so this commit disables validation for metadata
collected from .d.ts files.
Fixes#22210
PR Close#23275
Lowering expressions in flat module metadata is desirable, but it won't
work without some rearchitecting. Currently the flat module index source
is added to the Program and therefore must be determined before the rest
of the transforms run. Since the lowering transform changes the set of
exports needed in the index, this creates a catch-22 in the index
generation.
This commit causes the flat module index metadata to be generated using
only those transforms which are "safe" (don't modify the index).
PR Close#23226
Currently, the flat module index metadata is produced directly from
the source metadata. The compiler, however, applies transformations
on the Typescript sources during transpilation, and also equivalent
transformations on the metadata itself. This transformed metadata
doesn't end up in the flat module index.
This changes the compiler to generate the flat module index metadata
from its transformed version instead of directly from source.
PR Close#23129
Computing the value of loadChildren does not work externally, as the CLI
needs to be able to detect the paths referenced to properly set up
codesplitting. However, internally, different approaches to codesplitting
require hashed module IDs, and the computation of those hashes involves
something like:
{path: '...', loadChildren: hashFn('module')}
ngc should lower loadChildren into an exported constant in that case.
This will never break externally, because loadChildren is always a
string externally, and a string won't get lowered.
PR Close#23088