Previously, the flow control logic for patterns didn't use the
`FlowModel.split` or `FlowModel.unsplit` methods at all. This meant
that if a control flow join point occurred in pattern logic, flow
analysis would consider the split point to be whatever split point was
established by the enclosing expression or statement. In the case of
an if-case statement, it would consider the split point to be at the
beginning of the scrutinee expression.
Split points are used by flow analysis for the sole purpose of
ensuring that joins propagate type promotions the same way in dead
code as they do in live code (so that users introducing temporary
`throw` expressions or `return` statements into their code do not have
to deal with nuisance compile errors in the (now dead) code that
follows. The consequence of flow analysis considering the split point
to be at the beginning of the scrutinee expression is that if the
scrutinee expression is proven to always throw, then joins that arise
from the pattern or guard may not behave consistently with how they
would have behaved otherwise. For example:
int getInt(Object o) => ...;
void consumeInt(int i) { ... }
test(int? i) {
if (
// (1)
getInt('foo')
case
// (2)
int()
// (3)
when i == null) {
} else {
// (4)
consumeInt(i);
}
}
In the above code, there is a join point at (4), joining control flows
from (a) the situation where the pattern `int()` failed to match, and
(b) the situation where `i == null` evaluated to `false` (and hence
`i` is promoted to non-nullable `int`). Since the return type of
`getInt` is `int`, it's impossible for the pattern `int()` to fail, so
at the join point, control flow path (a) is considered
unreacable. Therefore the promotion from control flow path (b) is
kept, and so the call to `consumeInt` is valid.
In order to decide whether to preserve promotions from one of the
control flow paths leading up to a join, flow analysis only considers
reachability relative to the corresponding split point. Prior to this
change, the split point in question occurred at (1), so if the
expression `getInt('foo')` had been replaced with `getInt(throw
UnimplementedError())`, flow analysis would have considered both
control flow paths (a) and (b) to be unreachable relative to the split
point, so it would not have preserved the promotion from (b), and
there would have been a compile time error in the (now dead) call to
`consumeInt`.
This change moves the split point from (1) to (2), so that changing
`getInt('foo')` to `getInt(throw UnimplementedError())` no longer
causes any change in type promotion behavior.
The implementation of this change is to add calls to `FlowModel.split`
and `FlowModel.unsplit` around all top-level patterns. At first glance
this might appear to affect the behavior of all patterns, but actually
the only user-visible effect is on patterns in if-case statements,
because:
- In switch statements and switch expressions, there is already a
split point before each case.
- In irrefutable patterns, there is no user-visible effect, because
irrefutable patterns cannot fail to match, and therefore don't do
any control flow joins.
This change allows the split points for patterns to be determined by a
simple syntactic rule, which will facilitate some refactoring of split
points that I am currently working on.
Change-Id: I55573ba5c28b2f2e6bba8731f9e3b02613b6beb2
Bug: https://github.com/dart-lang/sdk/issues/53167
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/319381
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Commit-Queue: Paul Berry <paulberry@google.com>
Since user @staticInterop types are erased to JavaScriptObject, they
should only be able to subtype other types that are :> JavaScriptObject,
which are just JSObject and JSAny. Eventually all the other JS types
will move to extension types and we can remove this check.
Change-Id: If56b6770e141238b583937880ca87496780c8fac
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/316865
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Since external APIs can only use primitives and JS types in static interop,
we should require that all type parameters on static interop APIs extend
another static interop type. This is the minimum required to ensure all
type parameters can be erased to JSValue. This only affects dart:js_interop
users and replaces the previous type parameter static error check.
Change-Id: Ia546874da73c808aa25deb8d54d581db783987df
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/316140
Reviewed-by: Joshua Litt <joshualitt@google.com>
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
This is a reland of commit 16fcfe7eae
Original change's description:
> [dart:js_interop] Remove Object.toJS and JSNumber.toDart
>
> Modifies JSBoxedDartObject reified type on JS backends and also
> modifies JSBoxedDartObject.toDart now that a proper box is introduced.
> Also uses a JS symbol in JSBoxedDartObject.toDart for a property
> check so that different Dart apps don't accidentally share Dart
> objects. It's now an error to call this function on non-boxed objects.
>
> Fixes a type issue where creating a new object literal with the JS
> foreign function was resulting in dart2js thinking toJSBox would
> always throw. Changing the typeDescription to PlainJavaScriptObject
> instead of =Object fixes that issue.
>
> CoreLibraryReviewExempt: Backend-specific library.
> Change-Id: I5cfb1f32ff4328fafdf9831b0d8da806c39391d9
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/309082
> Reviewed-by: Joshua Litt <joshualitt@google.com>
> Commit-Queue: Srujan Gaddam <srujzs@google.com>
CoreLibraryReviewExempt: Reland.
Change-Id: If6b190f12bdf840b0259c5739f50d9bdcd27fd47
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313600
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
This reverts commit 16fcfe7eae.
Reason for revert: Flutter changes haven't landed to google3 yet.
Original change's description:
> [dart:js_interop] Remove Object.toJS and JSNumber.toDart
>
> Modifies JSBoxedDartObject reified type on JS backends and also
> modifies JSBoxedDartObject.toDart now that a proper box is introduced.
> Also uses a JS symbol in JSBoxedDartObject.toDart for a property
> check so that different Dart apps don't accidentally share Dart
> objects. It's now an error to call this function on non-boxed objects.
>
> Fixes a type issue where creating a new object literal with the JS
> foreign function was resulting in dart2js thinking toJSBox would
> always throw. Changing the typeDescription to PlainJavaScriptObject
> instead of =Object fixes that issue.
>
> CoreLibraryReviewExempt: Backend-specific library.
> Change-Id: I5cfb1f32ff4328fafdf9831b0d8da806c39391d9
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/309082
> Reviewed-by: Joshua Litt <joshualitt@google.com>
> Commit-Queue: Srujan Gaddam <srujzs@google.com>
Change-Id: I469ad04db7b49ffef47a46ccac97e909e4865719
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313580
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
Modifies JSBoxedDartObject reified type on JS backends and also
modifies JSBoxedDartObject.toDart now that a proper box is introduced.
Also uses a JS symbol in JSBoxedDartObject.toDart for a property
check so that different Dart apps don't accidentally share Dart
objects. It's now an error to call this function on non-boxed objects.
Fixes a type issue where creating a new object literal with the JS
foreign function was resulting in dart2js thinking toJSBox would
always throw. Changing the typeDescription to PlainJavaScriptObject
instead of =Object fixes that issue.
CoreLibraryReviewExempt: Backend-specific library.
Change-Id: I5cfb1f32ff4328fafdf9831b0d8da806c39391d9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/309082
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
This reverts commit a356f71b71.
Reason for revert: This should be handled by throwing an exception when
the methods of the returned window are called, not when it is opened.
This would be a noisy breaking change that we don't want for 3.1. For
now, revert until the change that affects the individual methods is
landed.
Original change's description:
> [dart:html] Throw exception if Window.open opens null window
>
> Window.open silently allows a null window to be opened, and
> issues arise later when users try to use the non-null wrapper.
> This CL changes that to throw an exception if the window is null.
> This exception can be caught and recovered from. This avoids the
> larger breaking change of making this API nullable.
>
> CoreLibraryReviewExempt: Backend-specific library.
> Change-Id: I9a53a477cb370c3bc6bc26b2162ce66c5af166aa
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/306910
> Reviewed-by: Sigmund Cherem <sigmund@google.com>
> Commit-Queue: Srujan Gaddam <srujzs@google.com>
CoreLibraryReviewExempt: Revert in backend-specific library.
Change-Id: I5007b7d7aa608bfc8e5827b5f967af5573d0b758
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/309000
Commit-Queue: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
This reverts commit 14a3051552.
Reason for revert: b/289195983
Original change's description:
> [dart:html] Move NullWindowException to implementation
>
> Window.open may open a null window in more cases than expected.
> Users may not care that the window they get back is invalid if
> they never use it. Therefore, this CL moves the exception to
> the implementation of the returned window in order to reduce
> noise, but still give a way for users to recover if they wish
> to do so.
>
> CoreLibraryReviewExempt: Backend-specific library.
> Change-Id: I005cf80630cfb4db2f5ec2012cfcd0161ad10ff1
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/311460
> Commit-Queue: Srujan Gaddam <srujzs@google.com>
> Reviewed-by: Sigmund Cherem <sigmund@google.com>
Change-Id: Id7c514a80fdcaa18a7eb0acdcb7f36477a04d966
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/311843
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Auto-Submit: Ivan Inozemtsev <iinozemtsev@google.com>
Window.open may open a null window in more cases than expected.
Users may not care that the window they get back is invalid if
they never use it. Therefore, this CL moves the exception to
the implementation of the returned window in order to reduce
noise, but still give a way for users to recover if they wish
to do so.
CoreLibraryReviewExempt: Backend-specific library.
Change-Id: I005cf80630cfb4db2f5ec2012cfcd0161ad10ff1
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/311460
Commit-Queue: Srujan Gaddam <srujzs@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Window.open silently allows a null window to be opened, and
issues arise later when users try to use the non-null wrapper.
This CL changes that to throw an exception if the window is null.
This exception can be caught and recovered from. This avoids the
larger breaking change of making this API nullable.
CoreLibraryReviewExempt: Backend-specific library.
Change-Id: I9a53a477cb370c3bc6bc26b2162ce66c5af166aa
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/306910
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
This annotation is unneeded as named arg external constructors
is enough to determine if an object literal is created. The one
exception is an empty object literal, but our guidance is to use
utility functions to create one instead. This is a better fit for
that purpose too as an interface for an empty object literal is
more uncommon.
CoreLibraryReviewExempt: Backend-specific library.
Change-Id: I10cf891601b28ff7e56129842d099ea28863626d
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/307506
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
This includes a single commit which fixes:
https://github.com/dart-lang/dart_style/issues/1224
This doesn't need to be coordinated with a pre-built SDK update.
Change-Id: Ifaabfde1d0d3dc180cf65a1052ba8dcff3324013
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/306122
Reviewed-by: Alexander Thomas <athom@google.com>
Commit-Queue: Bob Nystrom <rnystrom@google.com>
Auto-Submit: Bob Nystrom <rnystrom@google.com>
Breaking change request: https://dartbug.com/52334
Four abstract pure-interface classes in `stream.dart` have
been missed, and not made into `interface` declarations
like other similar abstract interface-only classes.
They should be marked `interface`, for consistency,
and to express intent.
The classes are:
* `StreamConsumer`
* `StreamIterator`
* `StreamTransformer`
* `MultiStreamController`
All are abstract classes with only abstract instance members.
Only `StreamConsumer` has uses where it is extended.
Most of those are inside the SDK, and are fixed here.
Two more are in packages (`http_server` and `streams`).
The former is discontinued and archived, and will never have a 3.0
release. The latter is internally developed and easy to fix.
Hope to cherry pick this for 3.0.3
Tested: Modified affected tests. Only adding restrictions.
CoreLibraryReviewExempt: Adding modifiers only.
Change-Id: I41aa47e48eaf769b7bd9d3206c1079a16ef3d476
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/302204
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
This has a few changes. One of them affects code even before Dart 3,
though it only touches parameters with metadata on them, which tend to
be rare (outside of uses of the freezed package). It may affect code in
the SDK repo.
The other two are specific records and patterns so won't affect much
existing code. But the formatted behavior before these bug fixes looks
pretty bad, so if it's not too much risky, I think it's worth cherry
picking into the next stable point release.
Change-Id: I1e60d251b474b099c5059ffa63ddb260f0ed1318
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/302500
Reviewed-by: Alexander Thomas <athom@google.com>
Reviewed-by: Michael Thomsen <mit@google.com>
Commit-Queue: Bob Nystrom <rnystrom@google.com>
We've changed the semantics for these members using dart:js_interop
so that not passing in optionals on the Dart side is equivalent to
not passing in optionals on the JS side. This CL makes that consistent
with package:js as well.
Modifies CHANGELOG to announce breaking change.
Change-Id: Ic5c33c9c797983a72edec9bc59f60fc1f29240b4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/300400
Reviewed-by: Joshua Litt <joshualitt@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>
Disallows tear-offs for all interop members when using
dart:js_interop, and disallow tear-offs of @staticInterop and
extension external members. Creates tear-off indexes in
InlineExtensionIndex to keep track of tear-off methods that
the CFE generates.
Modifies the CHANGELOG to communicate the breaking change.
Change-Id: I900fdfd6ee6b198f2f34e9d9fd5f3d9c964680e3
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/299800
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Joshua Litt <joshualitt@google.com>
Also adds `firstOrNull`, `lastOrNull`, `singleOrNull`, and `elementAtOrNull`.
I chose `nonNulls` instead of `whereNotNull()`.
The latter should have been a getter, but isn't.
Making it a getter in the platform libraries would be confusing,
and make migrating from `package:collection`'s function harder,
so instead I gave it a completely new name.
The alternative would be to retain the non-idiomatic `whereNotNull()`
for familiarity. But then it would be tool late to fix.
Fixes#49928
BUG= https://github.com/dart-lang/sdk/issues/49928
CoreLibraryReviewExempt: Everybody's on vacation, everybody everywhere.
Change-Id: If464e3bd6bc97cbeefc3e5084b4cbaadac3f1e95
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/290760
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Makes `Iterable` a mixin class with a `const` constructor,
and `IterableBase` and `IterableMixin` (to-be-deprecated) aliases
for it.
Combining everything in one place avoids (some) code
duplication.
Also gave the methods a quick overhaul for formatting,
removing uses of `late` and unnecessary element accesses,
and some general tweaks.
CoreLibraryReviewExempt: Refactoring of redundant types.
Tested: Refactoring. If existing tests work, it works.
Change-Id: Ie49a88f713d386d2d118c53606a71bdd50e1eb11
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/287600
Reviewed-by: Brian Quinlan <bquinlan@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
It's always a RegExp, so it's a safe change.
It's also a very, very rarely used property,
since nobody ever noticed that it wasn't typed optimally.
Tested: Type change of getter to actual type of value. No new test.
CoreLibraryReviewExempt: Aske is out.
Change-Id: Ifb560a80c3cdb05be0164c593383539dc10ec0dd
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/267961
Reviewed-by: Stephen Adams <sra@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Change-Id: Ia6eda18bb4ad6ae9f2705846e262f793f73d1e4f
Tested: class type changes
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/291343
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Brian Quinlan <bquinlan@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
With the addition of records, it now makes sense to create an API
for parallelizing of futures of different types, like `Future.wait`
does for collections of same-typed futures.
The `wait` getters here apply to tuples of 2-9 futures of distinct types, which should be enough for most reasonable uses.
Planned addition for Dart 3.0.
CoreLibraryReviewExempt: Everybody's on vacation, everybody everywhere.
Change-Id: Iaa814e02e2274082bb8a29b9a18b4930bcc953bb
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/288903
Reviewed-by: Nate Bosch <nbosch@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Pure interfaces made `interface`s
implementation classes which cannot/should not be extended made `final`.
A class like `HasHMap` which provides the `Map` interface,
and no implementation except factory constructors
for internal implementations, is made `final`.
Unified {List,Set,Map}{Base,Mixin} into their `Base` class.
Deprecations are retained in comments for now, to be landed
separately. Search for '// TODO: @Deprecated'.
Tested: No new test, only adding restrictions on use.
CoreLibraryReviewExempt: Everybody's on vacation, everybody everywhere.
Change-Id: Ia83b8a3bb20b5b214546b328d4492de6658253db
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/288240
Reviewed-by: Aske Simon Christensen <askesc@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Lasse Nielsen <lrn@google.com>
Reviewed-by: Nate Bosch <nbosch@google.com>
Allows the following semantics:
- If the disable flag is passed, never check
- If we're in non-sound mode, never check to avoid a breaking
change (this was the behavior before)
- If the enable flag was passed and we're in sound mode,
always check
- If we're in sound mode and with >= -03, don't check
Discussion in https://github.com/dart-lang/sdk/issues/50710
Change-Id: If83aaebb2745e4c8bcaa3f5a38ff41d79a869b8b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/289451
Reviewed-by: Nicholas Shahan <nshahan@google.com>
Commit-Queue: Srujan Gaddam <srujzs@google.com>