* Revert "Revert "Roll engine to fed2ea458ed49088d33eddabc546ba56d600c717 (includes dart roll) (#19044)" (#19276)"
This reverts commit cf932490b7 as it also
includes fix for type error that broke tests.
* Add type cast for dart2 type checks.
* Move up to latest goldens
* Make inDirectory() type-parameterized.
* Add typecasting to transitions_perf_test.dart and microbenchmarks.
* Add boolean flag initialization in save_catalog_screenshots.dart
* Add type conversion to gallery transition test
* Roll engine to rolled dart
Add connectionTimeout property to class that extends HttpOverrides.
Upgrade packages.
Add more type fixes.
Update goldens hash.
Add more type fixes.
Upgrade packages.
Fix tests.
Remove strong-mode from analysis options yaml file.
Increase dev/devicelab timeout from 2 to 10 seconds.
Added getVM to waitForViews. Fix type errors in compileExpression.
* Pick up engine revision with fix for microbenchmarks regression.
Increase default timeout from 20/30 to 60s.
Instead, if the surface is 0x0 semantics will just generate one root semantics node with size of 0x0. The node will have no children because all children are invisible and (except for the root node with this change) invisible children are dropped.
This should reduce the number of flakes without actually increasing
the timeout, so we'll still find out quickly if a test is hanging.
The numbers here might need tweaking. Maybe the default two seconds is
too short for CI bots.
This allows test environments other than `flutter test` to have a hook
into the test exception reporting. Some test environments, for example,
don't just dump error details to the console, but rather require them
to be reported to a separate server.
Our style guide says the k's are not necessary, and it seems like a good idea to make all the code be consistent on this.
Only naming changes to private vars: no logic changes.
Some APIs (such as the ones in package:test) assume you're running
in a child zone of the test zone by attempting to extract information
from the zone values. When we run runAsync() in the root zone, those
zone values are lost, and such API methods don't work.
The solution is to run in a child zone, but with a specification that
says to use the Root zone for task scheduling (both timers and
microtakss).
1. Make goldenFileComparator getter return `null` if it's set to the
uninitialized comparator, which matches the behavior of the setter
(it sets it to the uninitialized comparator if the caller specifies
`null`).
2. Make the uninitialized comparator return trivial success (and print
a message) when asked to compare as opposed to throwing. This ensures
that the comparator will play nicely with live widget bindings
3. Augment documentation
4. Add assert that test doesn't modify the value of `autoUpdateGoldenFiles`
* use HttpOverrides and dart:io HttpClient in flutter
* add missing package:http dependency
* update flutter packages and remove comment about createHttpClient from flutter_test
* move byte loading logic to common class, move string parsing logic to base class
* addAll doesn't work for a Uint8List
* use bytes.setRange
* undo addition to hello_world
* add newline to end of binding.dart
* and a newline for hello world
* refactor to function and add tests
* address comments on unknown length case
* alignment
* sort alaphabetically
* rename convertResponse to consolidateClientHttpClientResponseBytes. Add header
* fix alignment in test
Also:
* Remove find.byIcon since it's identical to find.icon. (I sent mail to flutter-dev about this.)
* Fix IconData's operator== and hashCode, which had not been updated when we added fields.
* Add the byTooltip finder to the list of suggested finders.
* Make the suggested Key finder prettier.
This allows the scheduler library to depend on the services library
and the painting library to depend on the scheduler library without
the services library having to depend on the scheduler library.
While I was at it I also cleaned up some of the binding logic: the
licenses logic can now be overridden (and the test library does so),
and the image cache can now be overridden as well.
I got tired of drive-by spelling fixes, so I figured I'd just take care of them all at once.
This only corrects errors in the dartdocs, not regular comments, and I skipped any sample code in the dartdocs. It doesn't touch any identifiers in the dartdocs either. No code changes, just comments.
* fix updrade script; upgrade to the latest package versions
* exclude special dependencies from transitive closure
* fix stack trace handling in flutter_test due to stack_trace change
* change type on _emptyStackTrace
Also:
* Provide a better message when you lerp from infinity to finity
constraints.
* Make the striped marker support RTL.
* By popular demand, dump a warning to the console the first time
a particular Flex overflows. (Resets on hot reload.)
Since the called function can't throw, there is no need for the "guarded".
Since the function returns something, running in guarded mode doesn't really work. The `uncaught` handler wouldn't know what to return (except for `null`).
* Introduce a Directionality inherited widget which sets the ambient LTR vs RTL mode (defaulting to null, which means you cannot use directionality-influenced values).
* Make it possible to configure Padding (including Container.padding and Container.margin) using a directionality-agnostic EdgeInsets variant.
* Provide textDirection and verticalDirection controls on Row and Column to make them RTL-aware.
* Introduce a variant of FractionalOffset based on the EdgeInsets variant. Not yet actually used.
* Fix all the tests that depended on Row defaulting to LTR.
It was 8.0. It's now arbitrarily 18.0.
Changing this required adjusting some tests. Adjusting the tests
required debugging the tests. Debugging the tests required some tools
to help debugging gesture recognizers and gesture arenas, so I added
some. It also required updating some toString() methods which resulted
in some changes to the tree diagnostics logic.
Also I cleaned up some docs while I was at it.
Mainly, this adds documentation to members that were previously
lacking documentation.
It also adds a big block of documentation about improving performance
of widgets.
This also removes some references to package:collection and adds
global setEquals and listEquals methods in foundation that we can use.
(setEquals in particular should be much faster than the
package:collection equivalent, though both should be faster as they
avoid allocating new objects.) All remaining references now qualify
the import so we know what our remaining dependencies are.
Also lots of code reordering in Flutter driver to make the code
consistent and apply the style guide more thoroughly.
This splits the frame pipeline into two, beginFrame and drawFrame.
As part of making this change I added some debugging hooks that helped
debug the issues that came up:
* I added debugPrintScheduleFrameStacks which prints a stack whenever
a frame is actually scheduled, so you can see why frames are being
scheduled.
* I added some toString output to EditableText and RawKeyboardListener.
* I added a scheduler_tester.dart library for scheduler library tests.
* I changed the test framework to flush microtasks before pumping.
* Some asserts that had the old string literal form were replaced by
asserts with messages.
I also fixed a few subtle bugs that this uncovered:
* setState() now calls `ensureVisualUpdate`, rather than
`scheduleFrame`. This means that calling it from an
AnimationController callback does not actually schedule an extra
redundant frame as it used to.
* I corrected some documentation.
* Manually fix every use of Point.x and Point.y
Some of these were moved to dx/dy, but not all.
* Manually convert uses of the old gradient API
* Remove old reference to Point.
* Mechanical changes
I applied the following at the root of the Flutter repository:
git ls-files -z | xargs -0 sed -i 's/\bPoint[.]origin\b/Offset.zero/g'
git ls-files -z | xargs -0 sed -i 's/\bPoint[.]lerp\b/Offset.lerp/g'
git ls-files -z | xargs -0 sed -i 's/\bnew Point\b/new Offset/g'
git ls-files -z | xargs -0 sed -i 's/\bconst Point\b/const Offset/g'
git ls-files -z | xargs -0 sed -i 's/\bstatic Point /static Offset /g'
git ls-files -z | xargs -0 sed -i 's/\bfinal Point /final Offset /g'
git ls-files -z | xargs -0 sed -i 's/^\( *\)Point /\1Offset /g'
git ls-files -z | xargs -0 sed -i 's/ui[.]Point\b/ui.Offset/g'
git ls-files -z | xargs -0 sed -i 's/(Point\b/(Offset/g'
git ls-files -z | xargs -0 sed -i 's/\([[{,]\) Point\b/\1 Offset/g'
git ls-files -z | xargs -0 sed -i 's/@required Point\b/@required Offset/g'
git ls-files -z | xargs -0 sed -i 's/<Point>/<Offset>/g'
git ls-files -z | xargs -0 sed -i 's/[.]toOffset()//g'
git ls-files -z | xargs -0 sed -i 's/[.]toPoint()//g'
git ls-files -z | xargs -0 sed -i 's/\bshow Point, /show /g'
git ls-files -z | xargs -0 sed -i 's/\bshow Point;/show Offset;/g'
* Mechanical changes - dartdocs
I applied the following at the root of the Flutter repository:
git ls-files -z | xargs -0 sed -i 's/\ba \[Point\]/an [Offset]/g'
git ls-files -z | xargs -0 sed -i 's/\[Point\]/[Offset]/g'
* Further improvements and a test
* Fix minor errors from rebasing...
* Roll engine
* Fix tests to use Ahem, and helpful changes around that
- Fix fonts that had metric-specific behaviours.
- LiveTestWidgetsFlutterBinding.allowAllFrames has been renamed
to LiveTestWidgetsFlutterBinding.framePolicy.
- LiveTestWidgetsFlutterBinding now defaults to using a frame policy
that pumps slightly more frames, to animate the pointer crosshairs.
- Added "flutter run --use-test-fonts" to enable Ahem on devices.
- Changed how idle() works to be more effective in live mode.
- Display the test name in live mode (unless ahem fonts are enabled).
- Added a toString to TextSelectionPoint.
- Style nit fixes.
* Roll engine to get Ahem changes.
* Update tests for dartdoc changes.
* Fix flutter_tools tests
- output single-line JSON because iOS 9.x cannot output more than one line
- move the workaround for #7433 into LiveTestWidgetsFlutterBinding so that it works in all cases (e.g. animation_bench.dart on iOS)
```dart
expect(x, moreOrLessEquals(0.0));
```
...for those cases where `x` might be 1e-11 or whatever.
Also, be more resilient when dumping the tree from inside the test framework.
Also, add an assert that helped me debug something the other day.
This requires all AnimationController objects to be given a
TickerProvider, a class that can create the Ticker.
It also provides some nice mixins for people who want to have their
State provide a TickerProvider. And a schedulerTickerProvider for those
cases where you just want to see your battery burn.
Also, we now enforce destruction order for elements.
Previously, if a StatefulWidget was marked dirty, then removed from the
build, then reinserted using the exact same widget under a widget under
a LayoutBuilder, it wouldn't rebuild.
This fixes that.
It also introduces an assert that's supposed to catch SizeObserver-like
behaviour. Rather than make this patch even bigger, I papered over two
pre-existing bugs which this assert uncovered (and fixed the other
problems it found):
https://github.com/flutter/flutter/issues/5751https://github.com/flutter/flutter/issues/5749
We should fix those before 1.0 though.
Don't suggest filing an issue when we can definitively say that the
assertion in question was not thrown from within the flutter package.
Fixes https://github.com/flutter/flutter/issues/3812.
Put the stack trace first after the message, with more details below the
stack trace, since the stack is often very useful and the stacks are no
longer stupidly long.
Better document the 'rescheduling' feature, and improve the error
handling asserts when it is misused.
Fixes https://github.com/flutter/flutter/issues/3888.
Improve the wording around the stack dump for exceptions during
callbacks.
Improve the color and font size of messages in live tests.
Fixes https://github.com/flutter/flutter/issues/4018.
* Add a "build" phase to EnginePhase for completeness.
* Ignore events from the device during test execution.
* More dartdocs
* Slightly more helpful messages about Timers in verifyInvariants.
* Add widgetList, elementList, stateList, renderObjectList.
* Send test events asynchronously for consistency with other APIs.
* Fix a test that was depending on test events being synchronous (or
rather, scheduled in a microtask that came before the microtask for
the completer of the future that the tap() function returned).
This makes it possible to substitute 'flutter run' for 'flutter test'
and actually watch a test run on a device.
For any test that depends on flutter_test:
1. Remove any import of 'package:test/test.dart'.
2. Replace `testWidgets('...', (WidgetTester tester) {`
with `testWidgets('...', (WidgetTester tester) async {`
3. Add an "await" in front of calls to any of the following:
* tap()
* tapAt()
* fling()
* flingFrom()
* scroll()
* scrollAt()
* pump()
* pumpWidget()
4. Replace any calls to `tester.flushMicrotasks()` with calls to
`await tester.idle()`.
There's a guarding API that you can use, if you have particularly
complicated tests, to get better error messages. Search for
TestAsyncUtils.
* Refactor widget test framework
Instead of:
```dart
test("Card Collection smoke test", () {
testWidgets((WidgetTester tester) {
```
...you now say:
```dart
testWidgets("Card Collection smoke test", (WidgetTester tester) {
```
Instead of:
```dart
expect(tester, hasWidget(find.text('hello')));
```
...you now say:
```dart
expect(find.text('hello'), findsOneWidget);
```
Instead of the previous API (exists, widgets, widget, stateOf,
elementOf, etc), you now have the following comprehensive API. All these
are functions that take a Finder, except the all* properties.
* `any()` - true if anything matches, c.f. `Iterable.any`
* `allWidgets` - all the widgets in the tree
* `widget()` - the one and only widget that matches the finder
* `firstWidget()` - the first widget that matches the finder
* `allElements` - all the elements in the tree
* `element()` - the one and only element that matches the finder
* `firstElement()` - the first element that matches the finder
* `allStates` - all the `State`s in the tree
* `state()` - the one and only state that matches the finder
* `firstState()` - the first state that matches the finder
* `allRenderObjects` - all the render objects in the tree
* `renderObject()` - the one and only render object that matches the finder
* `firstRenderObject()` - the first render object that matches the finder
There's also `layers' which returns the list of current layers.
`tap`, `fling`, getCenter, getSize, etc, take Finders, like the APIs
above, and expect there to only be one matching widget.
The finders are:
* `find.text(String text)`
* `find.widgetWithText(Type widgetType, String text)`
* `find.byKey(Key key)`
* `find.byType(Type type)`
* `find.byElementType(Type type)`
* `find.byConfig(Widget config)`
* `find.byWidgetPredicate(WidgetPredicate predicate)`
* `find.byElementPredicate(ElementPredicate predicate)`
The matchers (for `expect`) are:
* `findsNothing`
* `findsWidgets`
* `findsOneWidget`
* `findsNWidgets(n)`
* `isOnStage`
* `isOffStage`
* `isInCard`
* `isNotInCard`
Benchmarks now use benchmarkWidgets instead of testWidgets.
Also, for those of you using mockers, `serviceMocker` now automatically
handles the binding initialization.
This patch also:
* changes how tests are run so that we can more easily swap the logic
out for a "real" mode instead of FakeAsync.
* introduces CachingIterable.
* changes how flutter_driver interacts with the widget tree to use the
aforementioned new API rather than ElementTreeTester, which is gone.
* removes ElementTreeTester.
* changes the semantics of a test for scrollables because we couldn't
convince ourselves that the old semantics made sense; it only worked
before because flushing the microtasks after every event was broken.
* fixes the flushing of microtasks after every event.
* Reindent the tests
* Fix review comments
Previously the widgets layer only provided a concrete binding, which
makes it awkward to extend it compared to other bindings. This moves
widgets to the same style as the other layers.
In a subsequent patch I'll use this to make the tests layer saner.
