`flush()` can now be used from within fakeAsync tests to simulate moving
time forward until all macrotask events have been cleared from the
event queue.
Previous to 2.x there were some source maps distrubted, but they didn't go
all the way back to the TypeScript sources and they weren't available for
all JavaScript distrubted to NPM.
With this change source maps will be available for FESM distributions as
well as UMD and will go all the way back to TypeScript sources.
Fixes#15184
TypeScript compiler will now build to ES2015 code and modules. Babili is used to minify ES2015
code, providing an initial optimization that we couldn't previously get just from Uglify. Uses
Babel to convert ES2015 to UMD/ES5 code, and Uglify to minimize the output.
This change installs HttpModule with ServerModule, and overrides bindings to
service Http requests made from the server with the 'xhr2' NPM package.
Outgoing requests are wrapped in a Zone macro-task, so they will be tracked
within the Angular zone and cause the isStable API to show 'false' until they
return. This is essential for Universal support of server-side HTTP.
All the docs related files (docs-app, doc-gen, content, etc)
are now to be found inside the `/aio` folder.
The related gulp tasks have been moved from the top level
gulp file to a new one inside the `/aio` folder.
The structure of the `/aio` folder now looks like:
```
/aio/
build/ # gulp tasks
content/ #MARKDOWN FILES for devguides, cheatsheet, etc
devguides/
cheatsheets/
transforms/ #dgeni packages, templates, etc
src/
app/
assets/
content/ #HTML + JSON build artifacts produced by dgeni from /aio/content.
#This dir is .gitignored-ed
e2e/ #protractor tests for the doc viewer app
node_modules/ #dependencies for both the doc viewer builds and the dgeni stuff
#This dir is .gitignored-ed
gulpfile.js #Tasks for generating docs and building & deploying the doc viewer
```
Closes#14361
Tsickle transforms typescript code, which can change the location of code.
This can cause issues such as runtime stack traces reporting that errors
were raised on the incorrect line in the orginal source. This change replaces
the DecoratorDownlevelCompilerHost and the TsickleCompilerHost with tsickle's
TsickleCompilerHost, which automatically composes tsickle's source maps with
typescript's source maps, so that line numbers are correctly reported at
runtime.
PR Close#14150
Angular 1.x -> AngularJS
Angular 1 -> AngularJS
Angular1 -> AngularJS
Angular 2+ -> Angular
Angular 2.0 -> Angular
Angular2 -> Angular
I have deliberately not touched any of the symbol names as that would cause big merge collisions with Tobias's work.
All the renames are in .md, .json, and inline comments and jsdocs.
PR Close#14132
* feat: add an env for testing closure builds
* build(npm): add dev dependency on yarn (and remove dev props for readability)
* build: refactor integration test runner
Now that rxjs is stable and the rxjs team follows semver, we can update and unpin the dependency safely.
From now on the Angular application/library developers are in charge of controlling the rxjs version as long as it's newer than 5.0.1.
closes#13561closes#13478closes#13572
* chore: update protractor and selenium-webdriver packages
As `karma-jasmine` has a peer dependency on `jasmine-core@2.3`, but `jasmine` and `protractor` are using `jasmine-core@2.4` we need to add `jasmine-core@2.3` explicitly. Previously, the peer dependency was
satisfied by accident because npm deduped the dependency
for `jasmine-core@2.3` as top level dependency.
Note that the shrink-wrap files changes quite a bit because
of the deduping mechanism of npm.
* fix(benchpress): make it work with latest protractor and seleniuv-webdriver
* fix(e2e_tests): make them work with latest protractor
Closes#10503
It is possible for code in `beforeEach` to capture and fork a zone
(for example creating `NgZone` in `beforeEach`). Subsequently the code
in `it` may chose to do `fakeAsync`. The issue is that because the
code in `it` can use `NgZone` from the `beforeEach`. it effectively can
escape the `fakeAsync` zone. A solution is to run all of the test in
`ProxyZone` which allows a test to dynamically replace the rules at any
time. This allows the `beforeEach` to fork a zone, and then `it` to
retroactively became `fakeAsync` zone.
- many entry points were previously missing (e.g. all testing entry points, http, etc)
- upgrade ts-api-guardian to 0.0.3 that adds support for more api surface
- add all info to the spec that was surfaced by ts-api-guardian@0.0.3
Added and used the cors middleware:
- add the module as a dev depedency in the package.json file
- require the module in the jsserve.js file
- add the module in the middleware list
Closes#7273Closes#7274
This tool lets us re-write TypeScript sources before entering the emit pipeline.
For example, we lower Decorators to the tree-shakable Annotation form.
Instead of using injectAsync and returning a promise, use the `async` function
to wrap tests. This will run the test inside a zone which does not complete
the test until all asynchronous tasks have been completed.
`async` may be used with the `inject` function, or separately.
BREAKING CHANGE:
`injectAsync` is now deprecated. Instead, use the `async` function
to wrap any asynchronous tests.
Before:
```
it('should wait for returned promises', injectAsync([FancyService], (service) => {
return service.getAsyncValue().then((value) => { expect(value).toEqual('async value'); });
}));
it('should wait for returned promises', injectAsync([], () => {
return somePromise.then(() => { expect(true).toEqual(true); });
}));
```
After:
```
it('should wait for returned promises', async(inject([FancyService], (service) => {
service.getAsyncValue().then((value) => { expect(value).toEqual('async value'); });
})));
// Note that if there is no injection, we no longer need `inject` OR `injectAsync`.
it('should wait for returned promises', async(() => {
somePromise.then() => { expect(true).toEqual(true); });
}));
```
Closes#7735
To workaround https://github.com/Microsoft/TypeScript/issues/7573
we must remove the readonly keyword from generated .d.ts files.
This solution will not scale, but will probably buy enough time to require our users move to a 2.0 beta.
Closes#8003