Commit Graph

5785 Commits

Author SHA1 Message Date
JiaLiPassion 416c786774 fix(zone.js): should not try to patch fetch if it is not writable (#36311)
Close #36142

In Firefox extensions, the `window.fetch` is not configurable, that means

```
const desc = Object.getOwnPropertyDescriptor(window, 'fetch');
desc.writable === false;
```

So in this case, we should not try to patch `fetch`, otherwise, it will
throw error ('fetch is ReadOnly`)

PR Close #36311
2020-04-06 15:34:33 -07:00
Keen Yee Liau eb8c6c7eff test(language-service): Move pipe tests to TEST_TEMPLATE (#36407)
This commit simplifies the completion tests for pipes by moving them to TEST_TEMPLATE.

PR Close #36407
2020-04-06 15:32:33 -07:00
Paul Gschwendtner d43c30688a fix(core): avoid migration error when non-existent symbol is imported (#36367)
In rare cases a project with configured `rootDirs` that has imports to
non-existent identifiers could fail in the migration.

This happens because based on the application code, the migration could
end up trying to resolve the `ts.Symbol` of such non-existent
identifiers. This isn't a problem usually, but due to a upstream bug
in the TypeScript compiler, a runtime error is thrown.

This is because TypeScript is unable to compute a relative path from the
originating source file to the imported source file which _should_
provide the non-existent identifier. An issue for this has been reported
upstream: https://github.com/microsoft/TypeScript/issues/37731. The
issue only surfaces since our migrations don't provide an absolute base
path that is used for resolving the root directories.

To fix this, we ensure that we never use relative paths when parsing
tsconfig files. More details can be found in the TS issue.

Fixes #36346.

PR Close #36367
2020-04-06 13:21:54 -07:00
Pete Bacon Darwin ee70a18a75 fix(ngcc): don't crash on cyclic source-map references (#36452)
The source-map flattening was throwing an error when there
is a cyclic dependency between source files and source-maps.
The error was either a custom one describing the cycle, or a
"Maximum call stack size exceeded" one.

Now this is handled more leniently, resulting in a partially loaded
source file (or source-map) and a warning logged.

Fixes #35727
Fixes #35757
Fixes https://github.com/angular/angular-cli/issues/17106
Fixes https://github.com/angular/angular-cli/issues/17115

PR Close #36452
2020-04-06 13:19:53 -07:00
Alan Agius 76a8cd57ae fix(ngcc): add process title (#36448)
Add process.title, so it's clearly in the task manager when ngcc is running

See: https://github.com/angular/angular/issues/36414#issuecomment-609644282

PR Close #36448
2020-04-06 13:19:17 -07:00
Pete Bacon Darwin f9fb8338f5 fix(ngcc): support ignoring deep-imports via package config (#36423)
Recently we added support for ignoring specified deep-import
warnings by providing sets of regular expressions within the
`ngcc.config.js` file. But this was only working for the project
level configuration.

This commit fixes ngcc so that it will also read these regular
expressions from package level configuration too.

Fixes #35750

PR Close #36423
2020-04-06 11:32:09 -07:00
Pete Bacon Darwin 6b3aa60446 fix(ngcc): support simple `browser` property in entry-points (#36396)
The `browser` package.json property is now supported to the same
level as `main` - i.e. it is sniffed for UMD, ESM5 and CommonJS.

The `browser` property can also contain an object with file overrides
but this is not supported by ngcc.

Fixes #36062

PR Close #36396
2020-04-06 11:31:10 -07:00
Pete Bacon Darwin 2463548fa7 fix(ngcc): sniff `main` property for ESM5 format (#36396)
Previously, `main` was only checked for `umd` or `commonjs`
formats. Now if there are `import` or `export` statements in the
source file it will be deemed to be in `esm5` format.

Fixes #35788

PR Close #36396
2020-04-06 11:31:10 -07:00
Keen Yee Liau 1140bbc25c refactor(language-service): reformat using clang-format (#36426)
clang-format was recently updated and any PRs that touch files in the
language service will have to reformat all the files.

Instead of changing the formatting in those PRs, this PR formats all
files in language-service package once and for all.

PR Close #36426
2020-04-06 09:29:42 -07:00
Ayaz Hafiz e893c5a330 fix(compiler-cli): pass real source spans where they are empty (#31805)
Some consumers of functions that take `ParseSourceSpan`s currently pass
empty and incorrect source spans. This fixes those cases.

PR Close #31805
2020-04-06 09:28:27 -07:00
Pete Bacon Darwin 8be8466a00 style(ngcc): reformat of ngcc after clang update (#36447)
PR Close #36447
2020-04-06 09:26:57 -07:00
Mikel Ward 568e9df1d6 fix(router): allow UrlMatcher to return null (#36402)
The matcher is allowed to return null per
https://angular.io/api/router/UrlMatcher#usage-notes

And run `yarn gulp format` to pick up recent clang format changes.

Closes #29824

BREAKING CHANGE: UrlMatcher's type now reflects that it could always return
    null.

    If you implemented your own Router or Recognizer class, please update it to
    handle matcher returning null.

PR Close #36402
2020-04-03 11:16:23 -07:00
Joey Perrott 6402a9ae2a build: rebuild yarn lock from scratch (#36377)
Rebuild the yarn lock file from scratch to collapse instances where
one package is able to satisfy multiple dependencies.  Currently we
have some situations where we have multiple versions when one would
work.

Example:
```
"@babel/code-frame@^7.0.0":
  version "7.0.0"
  resolved "https://registry.yarnpkg.com/@babel/cod
  integrity sha512-OfC2uemaknXr87bdLUkWog7nYuliM9Ij
  dependencies:
    "@babel/highlight" "^7.0.0"

"@babel/code-frame@^7.5.5":
  version "7.5.5"
  resolved "https://registry.yarnpkg.com/@babel/cod
  integrity sha512-27d4lZoomVyo51VegxI20xZPuSHusqbQ
  dependencies:
    "@babel/highlight" "^7.0.0"

"@babel/code-frame@^7.8.3":
  version "7.8.3"
  resolved "https://registry.yarnpkg.com/@babel/cod
  integrity sha512-a9gxpmdXtZEInkCSHUJDLHZVBgb1QS0j
  dependencies:
    "@babel/highlight" "^7.8.3"
```

becomes

```
"@babel/code-frame@^7.0.0", "@babel/code-frame@^7.5.5", "@babel/code-frame@^7.8.3":
  version "7.8.3"
  resolved "https://registry.yarnpkg.com/@babel/cod
  integrity sha512-a9gxpmdXtZEInkCSHUJDLHZVBgb1QS0j
  dependencies:
    "@babel/highlight" "^7.8.3"
```

PR Close #36377
2020-04-03 11:09:17 -07:00
George Kalpakas ca25c957bf fix(ngcc): correctly detect imported TypeScript helpers (#36284)
The `NgccReflectionHost`s have logic for detecting certain known
declarations (such as `Object.assign()` and TypeScript helpers), which
allows the `PartialEvaluator` to evaluate expressions it would not be
able to statically evaluate otherwise.

In #36089, `DelegatingReflectionHost` was introduced, which delegates to
a TypeScript `ReflectionHost` when reflecting on TypeScript files, which
for ngcc's case means `.d.ts` files of dependencies. As a result, ngcc
lost the ability to detect TypeScript helpers imported from `tslib`,
because `DelegatingReflectionHost` was not able to apply the known
declaration detection logic while reflecting on `tslib`'s `.d.ts` files.

This commit fixes this by ensuring `DelegatingReflectionHost` calls the
`NgccReflectionHost`'s `detectKnownDeclaration()` method as necessary,
even when using the TypeScript `ReflectionHost`.

NOTE:
The previous commit exposed a bug in ngcc that was hidden due to the
tests' being inconsistent with how the `ReflectionHost`s are used in the
actual program. The changes in this commit are verified by ensuring the
failing tests are now passing (hence no new tests are added).

PR Close #36284
2020-04-03 11:08:46 -07:00
George Kalpakas 93f07aee6c test(ngcc): use `DelegatingReflectionHost` for testing `NgccReflectionHost`s (#36284)
In #36089, `DelegatingReflectionHost` was introduced. Under the hood, it
delegates another `NgccReflectionHost` in order to reflect over the
program's source files, while using a different TypeScript
`ReflectionHost` to reflect over `.d.ts` files (which is how external
dependencies are represented in the program).

Previously, the `NgccReflectionHost`s were used directly in tests. This
does not exercise them in the way they are exercised in the actual
program, because (when used directly) they will also reflect on `.d.ts`
files too (instead of delegating to the TypeScript `ReflectionHost`).
This could hide bugs that would happen on the actual program.

This commit fixes this by using the `DelegatingReflectionHost` in the
various `NgccReflectionHost` tests.

NOTE:
This change will cause some of the existing tests to start failing.
These failures demonstrate pre-existing bugs in ngcc, that were hidden
due to the tests' being inconsistent with how the `ReflectionHost`s are
used in the actual program. They will be fixed in the next commit.

PR Close #36284
2020-04-03 11:08:46 -07:00
George Kalpakas 0af6e9fcbb refactor(ngcc): move logic for identifying known declarations to method (#36284)
The `NgccReflectionHost`s have logic for detecting certain known
declarations (such as `Object.assign()` and TypeScript helpers), which
allows the `PartialEvaluator` to evaluate expressions it would not be
able to statically evaluate otherwise.

This commit moves the logic for identifying these known declarations to
dedicated methods. This is in preparation of allowing ngcc's
`DelegatingReflectionHost` (introduced in #36089) to also apply the
known declaration detection logic when reflecting on TypeScript sources.

PR Close #36284
2020-04-03 11:08:46 -07:00
Wagner Maciel 008e12edda fix(benchpress): fix typings in lview_debug.ts (#36236)
* Installing & using benchpress was throwing typescript errors because typings were not defined for the getters in lview_debug.ts

PR Close #36236
2020-04-03 11:06:07 -07:00
Paul Gschwendtner c24ad560fa feat(core): undecorated-classes migration should handle derived abstract classes (#35339)
In version 10, undecorated base classes that use Angular features need
to be decorated explicitly with `@Directive()`. Additionally, derived
classes of abstract directives need to be decorated.

The migration already handles this for undecorated classes that are
not explicitly decorated, but since in V9, abstract directives can be
used, we also need to handle this for explicitly decorated abstract
directives. e.g.

```
@Directive()
export class Base {...}

// needs to be decorated by migration when updating from v9 to v10
export class Wrapped extends Base {}

@Component(...)
export class Cmp extends Wrapped {}
```

PR Close #35339
2020-04-02 10:51:49 -07:00
Paul Gschwendtner 3d2db5c5f0 test: add integration test for undecorated-classes-with-decorated-fields migration (#35339)
We don't have an integration test for the `undecorated-classes-with-decorated-fields
migration. For consistency and to cover for the latest changes, we add
it to the `ng update` integration test.

PR Close #35339
2020-04-02 10:51:48 -07:00
Paul Gschwendtner 32eafef6a7 fix(core): undecorated-classes-with-decorated-fields migration does not decorate derived classes (#35339)
The `undecorated-classes-with-decorated-fields` migration has been
introduced with 904a2018e0, but misses
logic for decorating derived classes of undecorated classes which use
Angular features. Example scenario:

```ts
export abstract class MyBaseClass {
  @Input() someInput = true;
}

export abstract class BaseClassTwo extends MyBaseClass {}

@Component(...)
export class MyButton extends BaseClassTwo {}
```

Both abstract classes would need to be migrated. Previously, the migration
only added `@Directive()` to `MyBaseClass`, but with this change, it
also decorates `BaseClassTwo`.

This is necessary because the Angular Compiler requires `BaseClassTwo` to
have a directive definition when it flattens the directive metadata for
`MyButton` in order to perform type checking. Technically, not decorating
`BaseClassTwo` does not break at runtime.

We basically want to enforce consistent use of `@Directive` to simplify the
mental model. [See the migration guide](https://angular.io/guide/migration-undecorated-classes#migrating-classes-that-use-field-decorators).

Fixes #34376.

PR Close #35339
2020-04-02 10:51:48 -07:00
Paul Gschwendtner 2366480250 refactor(core): move schematic import manager to shared utils (#35339)
The import manager has been created for both the `missing-injectable`
and `undecorated-classes-with-di` migration. Both initial PRs brought
in the manager class, so the manager is duplicated in the schematics.

In order to reduce this duplication, and to expose the manager to other
schematics/migrations, we move it into the shared schematic utils.

PR Close #35339
2020-04-02 10:51:48 -07:00
Paul Gschwendtner 8e55a11283 refactor(core): move schematic base classes logic into shared utils (#35339)
Moves the `findBaseClassDeclarations` method into the shared
schematic utilities. This method will be useful for future
migrations, and for planned changes to the
`undecorated-classes-with-decorated-fields` migration.

PR Close #35339
2020-04-02 10:51:48 -07:00
crisbeto e9de28111d build: enable service-worker tests on saucelabs (#36129)
Enables the `service-worker` tests on Saucelabs and fixes some issues that were preventing them from running on IE. The issues were:
1. We were serving es2017 code during tests. I've set it to es5.
2. The check which was verifying whether the environment is supported ended up hitting a `require` call in the browser which caused it to fail on browsers that don't support the `URL` API.

PR Close #36129
2020-04-01 15:37:47 -07:00
JoostK 75afd80ae8 refactor(compiler): add `@nocollapse` annotation using a synthetic comment (#35932)
In Ivy, Angular decorators are compiled into static fields that are
inserted into a class declaration in a TypeScript transform. When
targeting Closure compiler such fields need to be annotated with
`@nocollapse` to prevent them from being lifted from a static field into
a variable, as that would prevent the Ivy runtime from being able to
find the compiled definitions.

Previously, there was a bug in TypeScript where synthetic comments added
in a transform would not be emitted at all, so as a workaround a global
regex-replace was done in the emit's `writeFile` callback that would add
the `@nocollapse` annotation to all static Ivy definition fields. This
approach is no longer possible when ngtsc is running as TypeScript
plugin, as a plugin cannot control emit behavior.

The workaround is no longer necessary, as synthetic comments are now
properly emitted, likely as of
https://github.com/microsoft/TypeScript/pull/22141 which has been
released with TypeScript 2.8.

This change is required for running ngtsc as TypeScript plugin in
Bazel's `ts_library` rule, to move away from the custom `ngc_wrapped`
approach.

Resolves FW-1952

PR Close #35932
2020-04-01 15:37:06 -07:00
George Kalpakas 326240eb91 fix(ngcc): allow ngcc configuration to match pre-release versions of packages (#36370)
Ngcc supports providing a project-level configuration to affect how
certain dependencies are processed and also has a built-in fallback
configuration for some unmaintained packages. Each entry in these
configurations could be scoped to specific versions of a package by
providing a version range. If no version range is provided for a
package, it defaults to `*` (with the intention of matching any
version).

Previously, the installed version of a package was tested against the
version range using the [semver][1] package's `satisfies()` function
with the default options. By default, `satisfies()` does not match
pre-releases (see [here][2] for more details on reasoning). While this
makes sense when determining what version of a dependency to install
(trying to avoid unexpected breaking changes), it is not desired in the
case of ngcc.

This commit fixes it by explicitly specifying that pre-release versions
should be matched normally.

[1]: https://www.npmjs.com/package/semver
[2]: https://github.com/npm/node-semver#prerelease-tags

PR Close #36370
2020-04-01 13:32:32 -07:00
Alan Agius b59bc0e38c fix(platform-server): update `xhr2` dependency (#36366)
Previous versions of xhr2 used the depreciated "new Buffer()".

Closes #36358

PR Close #36366
2020-04-01 13:31:38 -07:00
Keen Yee Liau 5516802142 fix(compiler): avoid undefined expressions in holey array (#36343)
From G3 bug ID: 116443331

The View Engine compiler crashes when it tries to compile a test in JIT mode
that includes the d3-scale-chromatic library [1]. The d3 package initializes
some arrays using the following pattern:

```js
export var scheme = new Array(3).concat(
  "d8b365f5f5f55ab4ac",
  // ... more entries
).map(colors);
```

The stack trace from the crash is as follows:

```
TypeError: Cannot read property 'visitExpression' of undefined
    at ../../../third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts:505:39
    at JitEmitterVisitor.AbstractEmitterVisitor.visitAllObjects third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=526
    at JitEmitterVisitor.AbstractEmitterVisitor.visitAllExpressions third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=505
    at JitEmitterVisitor.AbstractEmitterVisitor.visitLiteralArrayExpr third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=484
    at LiteralArrayExpr.visitExpression third_party/javascript/angular2/rc/packages/compiler/src/output/output_ast.ts?l=791
    at ../../../third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts:492:19
    at JitEmitterVisitor.AbstractEmitterVisitor.visitAllObjects third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=526
    at JitEmitterVisitor.AbstractEmitterVisitor.visitLiteralMapExpr third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=490
    at LiteralMapExpr.visitExpression third_party/javascript/angular2/rc/packages/compiler/src/output/output_ast.ts?l=819
    at ../../../third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts:505:39
    at JitEmitterVisitor.AbstractEmitterVisitor.visitAllObjects third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=526
    at JitEmitterVisitor.AbstractEmitterVisitor.visitAllExpressions third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=505
    at JitEmitterVisitor.AbstractEmitterVisitor.visitInvokeFunctionExpr third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_emitter.ts?l=318
    at JitEmitterVisitor.AbstractJsEmitterVisitor.visitInvokeFunctionExpr third_party/javascript/angular2/rc/packages/compiler/src/output/abstract_js_emitter.ts?l=112
    at InvokeFunctionExpr.visitExpression third_party/javascript/angular2/rc/packages/compiler/src/output/output_ast.ts?l=440
```

This is because the corresponding output AST for the array is of the form

```ts
[
  undefined,
  undefined,
  undefined,
  o.LiteralExpr,
  // ...
]
```

The output AST is clearly malformed and breaks the type representation of
`LiteralArrayExpr` in which every entry is expected to be of type `Expression`.

This commit fixes the bug by using a plain `for` loop to iterate over the
entire length of the holey array and convert undefined elements to
`LiteralExpr`.

[1]: https://github.com/d3/d3-scale-chromatic/blob/master/src/diverging/BrBG.js

PR Close #36343
2020-04-01 13:31:14 -07:00
Pete Bacon Darwin cc4b813e75 fix(ngcc): handle bad path mappings when finding entry-points (#36331)
Previously, a bad baseUrl or path mapping passed to an `EntryPointFinder`
could cause the original `sourceDirectory` to be superceded by a higher
directory. This could result in none of the sourceDirectory entry-points being
processed.

Now missing basePaths computed from path-mappings are discarded with
a warning. Further, if the `baseUrl` is the root directory then a warning is
given as this is most likely an error in the tsconfig.json.

Resolves #36313
Resolves #36283

PR Close #36331
2020-04-01 13:30:46 -07:00
Ayaz Hafiz fe2b6923ba fix(language-service): infer type of elements of array-like objects (#36312)
Currently the language service only provides support for determining the
type of array-like members when the array-like object is an `Array`.
However, there are other kinds of array-like objects, including
`ReadonlyArray`s and `readonly`-property arrays. This commit adds
support for retrieving the element type of arbitrary array-like objects.

Closes #36191

PR Close #36312
2020-04-01 13:24:53 -07:00
Pete Bacon Darwin 38ad1d97ab fix(ngcc): handle entry-points within container folders (#36305)
The previous optimizations in #35756 to the
`DirectoryWalkerEntryPointFinder` were over zealous
with regard to packages that have entry-points stored
in "container" directories in the package, where the
container directory was not an entry-point itself.

Now we will also walk such "container" folders as long
as they do not contain `.js` files, which we regard as an
indicator that the directory will not contain entry-points.

Fixes #36216

PR Close #36305
2020-04-01 13:20:52 -07:00
Pete Bacon Darwin 372b9101e2 refactor(ngcc): simplify `DirectoryWalkerEntryPointFinder` (#36305)
This commit simplifies the `DirectoryWalkerEntryPointFinder` inter-method
calling to make it easier to follow, and also to support controlling
walking of a directory based on its children.

PR Close #36305
2020-04-01 13:20:52 -07:00
Pete Bacon Darwin 7e62aa0c6e refactor(ngcc): rename INVALID_ENTRY_POINT to INCOMPATIBLE_ENTRY_POINT (#36305)
This name better reflects its meaning.

PR Close #36305
2020-04-01 13:20:52 -07:00
JiaLiPassion 36e927a8c6 fix(zone.js): UNPATCHED_EVENTS and PASSIVE_EVENTS should be string[] not boolean (#36258)
__zone_symbol__UNPATCHED_EVENTS and __zone_symbol__PASSIVE_EVENTS should be string[] type not boolean.
For example:
```
const config = window as ZoneGlobalConfigurations;
config.__zone_symbol__UNPATCHED_EVENTS = ['scroll'];
config.__zone_symbol__PASSIVE_EVENTS = ['scroll'];
```

PR Close #36258
2020-04-01 13:20:00 -07:00
Wagner Maciel 9d8bb634f9 fix(benchpress): update dependencies (#36205)
* updated, added, and removed dependencies in package.json
* added dependencies to BUILD.bazel

PR Close #36205
2020-04-01 13:19:33 -07:00
JiaLiPassion b44f7b5e16 fix(zone.js): fix 2 bluebird test cases for each/mapSeries (#36295)
`Bluebird.each` and `Bluebird.mapSeries` will accept a callback with `value` parameter,
the `value` should be the item in the array, not array itself.

For example:
```
const arr = [1, 2];
Bluebird.each(arr, function(value, idx) {
  console.log(`value: ${value}, idx: ${idx}`);
})
```

the output will be
```
value: 1, idx: 0
value: 2, idx: 1
```

This PR fix the test cases for `each` and `mapSeries` APIs.

PR Close #36295
2020-03-31 10:59:56 -07:00
JiaLiPassion c5df9ce474 build(zone.js): update zone.js version to 0.10.3 (#36214)
PR Close #36214
2020-03-31 10:59:17 -07:00
Andrew Kushnir 80c68583d1 refactor(core): use more narrow `QueryList` import to avoid circular deps issue (#36286)
Prior to this commit, the `packages/core/src/render3/interfaces/query.ts` file used to import `QueryList` using `../../linker`, which contains a lot of re-exports and as a result, this one import caused a lot of circular deps cycles reported by the tool that checks such deps. In other places in the code the `QueryList` is imported using more narrow import (`linker/query_list`), so this commit uses the same pattern. This change allowed to reduce the number of known cycles from 343 to 207, the golden file was updated accordingly.

PR Close #36286
2020-03-30 15:31:39 -07:00
Greg Magolan 5b6ced5599 build: update to rules_nodejs 1.5.0 (#36307)
### New stuff

* The `ts_project` rule is a simple wrapper around the TypeScript compiler, `tsc`. This is an alternative to `ts_library` but not a replacement. Read more about the trade-offs at https://bazelbuild.github.io/rules_nodejs/TypeScript#alternatives or read the [API docs](https://bazelbuild.github.io/rules_nodejs/TypeScript#ts_project)
* `pkg_npm` can now be used as a dependency within your repo as well as for publishing to npm. It provides a `LinkablePackageInfo` which is our internal API to pass package name/path to downstream compilations, essentially providing the "Lerna" feature.
* There is experimental support for Bazel's "worker mode" in `rollup_bundle`, which essentially puts Rollup in watch mode. Add the `supports_workers = True` attribute to opt-in.
* Better support for [pre-defined label variables](https://docs.bazel.build/versions/master/be/make-variables.html#predefined_label_variables) like `$(rootpath)` and `$(execpath)` - we no longer recommend using `$(location)` at all.

See release notes https://github.com/bazelbuild/rules_nodejs/releases/tag/1.5.0 for more info.

PR Close #36307
2020-03-30 11:25:16 -07:00
Pete Bacon Darwin c6dd900f60 fix(ngcc): do not write entry-point manifest outside node_modules (#36299)
Fixes #36296

PR Close #36299
2020-03-30 11:03:26 -07:00
Pete Bacon Darwin 5ac308060d refactor(ngcc): rename `workerCount` to `maxWorkerCount` (#36298)
Now that we spawn workers lazily as needed, this private property is
really the upper limit on how many workers we might spawn.

PR Close #36298
2020-03-30 11:02:52 -07:00
George Kalpakas 5cee709266 fix(ngcc): do not spawn more processes than intended in parallel mode (#36280)
When running in parallel mode, ngcc spawns multiple worker processed to
process the various entry-points. The number of max allowed processes is
determined by the number of CPU cores available to the OS. There is also
currently an [upper limit of 8][1]. The number of active workers is in
turn inferred by the number of [task assignments][2].

In the past, counting the entries of `ClusterMaster#taskAssignments` was
enough, because worker processes were spawned eagerly at the beginning
and corresponding entries were created in `taskAssignments`.
Since #35719 however, worker processes are spawned lazily on an as
needed basis. Because there is some delay between
[spawning a process][3] and [inserting it][4] into `taskAssignments`,
there is a short period of time when `taskAssignment.size` does not
actually represent the number of spawned processes. This can result in
spawning more than `ClusterMaster#workerCount` processes.

An example of this can be seen in #36278, where the debug logs indicate
9 worker processes had been spawned (`All 9 workers are currently busy`)
despite the hard limit of 8.

This commit fixes this by using `cluster.workers` to compute the number
of spawned worker processes. `cluster.workers` is updated synchronously
with `cluster.fork()` and thus reflects the number of spawned workers
accurately at all times.

[1]: https://github.com/angular/angular/blob/b8e9a30d3b6/packages/compiler-cli/ngcc/src/main.ts#L429
[2]: https://github.com/angular/angular/blob/b8e9a30d3b6/packages/compiler-cli/ngcc/src/execution/cluster/master.ts#L108
[3]: https://github.com/angular/angular/blob/b8e9a30d3b6/packages/compiler-cli/ngcc/src/execution/cluster/master.ts#L110
[4]: https://github.com/angular/angular/blob/b8e9a30d3b6/packages/compiler-cli/ngcc/src/execution/cluster/master.ts#L199

PR Close #36280
2020-03-27 14:12:28 -07:00
Pete Bacon Darwin 995cd15a69 fix(ngcc): correctly identify the package path of secondary entry-points (#36249)
Previously we only searched for package paths below the set of `basePaths`
that were computed from the `basePath` provided to ngcc and the set of
`pathMappings`.

In some scenarios, such as hoisted packages, the entry-point is not within
any of the `basePaths` identified above. For example:

```
project
  packages
    app
      node_modules
        app-lib (depends on lib1)
  node_modules
    lib1 (depends on lib2)
      node_modules
        lib2 (depends on lib3/entry-point)
    lib3
      entry-point
```

When CLI is compiling `app-lib` ngcc will be given
`project/packages/app/node_modules` as the `basePath.

If ngcc is asked to target `lib2`, the `targetPath` will be
`project/node_modules/lib1/node_modules/lib2`.

Since `lib2` depends upon `lib3/entry-point`, ngcc will need to compute
the package path for `project/node_modules/lib3/entry-point`.

Since `project/node_modules/lib3/entry-point` is not contained in the `basePath`
`project/packages/app/node_modules`, ngcc failed to compute the `packagePath`
correctly, instead assuming that it was the same as the entry-point path.

Now we also consider the nearest `node_modules` folder to the entry-point
path as an additional `basePath`. If one is found then we use the first
directory directly below that `node_modules` directory as the package path.

In the case of our example this extra `basePath` would be `project/node_modules`
which allows us to compute the `packagePath` of `project/node_modules/lib3`.

Fixes #35747

PR Close #36249
2020-03-27 11:17:45 -07:00
Andrew Kushnir 16497438d6 fix(core): run `APP_INITIALIZER`s before accessing `LOCALE_ID` token in Ivy TestBed (#36237)
Prior to this commit, Ivy TestBed was accessing locale ID before `APP_INITIALIZER` functions were called. This execution order is not consistent with the app bootstrap logic in `application_ref.ts`. This commit updates Ivy TestBed execution order to call initializers first (since they might affect `LOCALE_ID` token value) and accessing and setting locale ID after that.

Fixes #36230.

PR Close #36237
2020-03-27 11:16:57 -07:00
Paul Gschwendtner 44acf6734b build: allow custom module resolution for ts-circular-deps tests (#36226)
Currently the `ts-circular-deps` tool uses a hard-coded module resolver
that only works in the `angular/angular` repository.

If the tool is consumed in other repositories through the shared
dev-infra package, the module resolution won't work, and a few
resolvable imports (usually cross-entry-points) are accidentally
skipped. For each test, the resolution might differ, so tests can
now configure their module resolution in a configuration file.

Note that we intentionally don't rely on tsconfig's for module
resolution as parsing their mappings rather complicates the
circular dependency tool. Additionally, not every test has a
corresponding tsconfig file.

Also, hard-coding mappings to `@angular/*` while accepting a
path to the packages folder would work, but it would mean
that the circular deps tool is no longer self-contained. Rather,
and also for better flexibility, a custom resolver should be
specified.

PR Close #36226
2020-03-27 11:14:49 -07:00
Sonu Kapoor 2d7c95fb70 fix(service-worker): prevent SW registration strategies from affecting app stabilization (#35870)
Previously, some of the built-in ServiceWorker registration strategies,
namely `registerWithDelay:<timeout>` and `registerWhenStable:<timeout>`,
would register potentially long-running timeout, thus preventing the app
from stabilizing before the timeouts expired. This was especially
problematic for the `registerWhenStable:<timeout>` strategy, which waits
for the app to stabilize, because the strategy itself would prevent the
app from stabilizing and thus the ServiceWorker would always be
registered after the timeout.

This commit fixes this by subscribing to the registration strategy
observable outside the Angular zone, thus not affecting the app's
stabilization.

PR Close #35870
2020-03-27 10:47:44 -07:00
Sonu Kapoor 29e8a64cf0 fix(service-worker): by default register the SW after 30s even the app never stabilizes (#35870)
Previously, when using the default ServiceWorker registration strategy
Angular would wait indefinitely for the [app to stabilize][1], before
registering the ServiceWorker script. This could lead to a situation
where the ServiceWorker would never be registered when there was a
long-running task (such as an interval or recurring timeout).

Such tasks can often be started by a 3rd-party dependency (beyond the
developer's control or even without them realizing). In addition, this
situation is particularly hard to detect, because the ServiceWorker is
typically not used during development and on production builds a
previous ServiceWorker instance might be already active.

This commit fixes this by changing the default registration strategy
from `registerWhenStable` to `registerWhenStable:30000`, which will
ensure that the ServiceWorker will be registered after 30s at the
latest, even if the app has not stabilized by then.

Fixes #34464

PR Close #35870
2020-03-27 10:47:44 -07:00
Sonu Kapoor 00efacf561 feat(service-worker): support timeout in `registerWhenStable` SW registration strategy (#35870)
Previously, when using the `registerWhenStable` ServiceWorker
registration strategy (which is also the default) Angular would wait
indefinitely for the [app to stabilize][1], before registering the
ServiceWorker script. This could lead to a situation where the
ServiceWorker would never be registered when there was a long-running
task (such as an interval or recurring timeout).

Such tasks can often be started by a 3rd-party dependency (beyond the
developer's control or even without them realizing). In addition, this
situation is particularly hard to detect, because the ServiceWorker is
typically not used during development and on production builds a
previous ServiceWorker instance might be already active.

This commit enhances the `registerWhenStable` registration strategy by
adding support for an optional `<timeout>` argument, which guarantees
that the ServiceWorker will be registered when the timeout expires, even
if the app has not stabilized yet.

For example, with `registerWhenStable:5000` the ServiceWorker will be
registered as soon as the app stabilizes or after 5 seconds if the app
has not stabilized by then.

Related to #34464.

[1]: https://angular.io/api/core/ApplicationRef#is-stable-examples

PR Close #35870
2020-03-27 10:47:44 -07:00
JoostK d783519835 fix(common): let `KeyValuePipe` accept type unions with `null` (#36093)
`KeyValuePipe` currently accepts `null` values as well as `Map`s and a
few others. However, due to the way in which TS overloads work, a type
of `T|null` will not be accepted by `KeyValuePipe`'s signatures, even
though both `T` and `null` individually would be.

To make this work, each signature that accepts some type `T` has been
duplicated with a second one below it that accepts a `T|null` and
includes `null` in its return type.

Fixes #35743

PR Close #36093
2020-03-24 14:41:41 -07:00
Pete Bacon Darwin b8e9a30d3b fix(ngcc): use preserve whitespaces from tsconfig if provided (#36189)
Previously ngcc never preserved whitespaces but this is at odds
with how the ViewEngine compiler works. In ViewEngine, library
templates are recompiled with the current application's tsconfig
settings, which meant that whitespace preservation could be set
in the application tsconfig file.

This commit allows ngcc to use the `preserveWhitespaces` setting
from tsconfig when compiling library templates. One should be aware
that this disallows different projects with different tsconfig settings
to share the same node_modules folder, with regard to whitespace
preservation. But this is already the case in the current ngcc since
this configuration is hard coded right now.

Fixes #35871

PR Close #36189
2020-03-24 14:25:06 -07:00
JoostK 32ce8b1326 feat(compiler): add dependency info and ng-content selectors to metadata (#35695)
This commit augments the `FactoryDef` declaration of Angular decorated
classes to contain information about the parameter decorators used in
the constructor. If no constructor is present, or none of the parameters
have any Angular decorators, then this will be represented using the
`null` type. Otherwise, a tuple type is used where the entry at index `i`
corresponds with parameter `i`. Each tuple entry can be one of two types:

1. If the associated parameter does not have any Angular decorators,
   the tuple entry will be the `null` type.
2. Otherwise, a type literal is used that may declare at least one of
   the following properties:
   - "attribute": if `@Attribute` is present. The injected attribute's
   name is used as string literal type, or the `unknown` type if the
   attribute name is not a string literal.
   - "self": if `@Self` is present, always of type `true`.
   - "skipSelf": if `@SkipSelf` is present, always of type `true`.
   - "host": if `@Host` is present, always of type `true`.
   - "optional": if `@Optional` is present, always of type `true`.

   A property is only present if the corresponding decorator is used.

   Note that the `@Inject` decorator is currently not included, as it's
   non-trivial to properly convert the token's value expression to a
   type that is valid in a declaration file.

Additionally, the `ComponentDefWithMeta` declaration that is created for
Angular components has been extended to include all selectors on
`ng-content` elements within the component's template.

This additional metadata is useful for tooling such as the Angular
Language Service, as it provides the ability to offer suggestions for
directives/components defined in libraries. At the moment, such
tooling extracts the necessary information from the _metadata.json_
manifest file as generated by ngc, however this metadata representation
is being replaced by the information emitted into the declaration files.

Resolves FW-1870

PR Close #35695
2020-03-24 14:21:42 -07:00