In the following code, it's not safe for the field `C._f` to undergo
type promotion, because a variable with static type `C` might have
type `D` at runtime, in which case `C._f` will get dispatched to
`noSuchMethod`, which is not guaranteed to return a stable result.
class C {
final int? _f;
}
class D implements C {
noSuchMethod(_) => ...;
}
foo(C c) {
if (c._f != null) {
print(c._f + 1); // UNSAFE!
}
}
Therefore, in order to determine which fields are promotable, the
implementations need to analyze enough of the class hierarchy to
figure out which field accesses might get dispatched to
`noSuchMethod`.
Currently, the CFE does this by following its usual algorithm for
generating `noSuchMethod` forwarders before trying to determine which
fields are promotable. The analyzer, on the other hand, doesn't have
an algorithm for generating `noSuchMethod` forwarders (since it
doesn't implement execution semantics); so instead it has its own
logic to figure out when a `noSuchMethod` forwarder is needed for a
field, and disable promotion for that field.
But there's a chicken-and-egg problem in the CFE: the CFE needs to
determine which fields are promotable before doing top-level inference
(since the initializers of top-level fields might make use of field
promotion, affecting their inferred types--see #50522). But it doesn't
decide where `noSuchMethod` forwarders are needed until after
top-level inference (because the same phase that generates
`noSuchMethod` forwarders also generates forwarders that do runtime
covariant type-checking, and so it has to run after all top level
types have been inferred).
To fix the chicken-and-egg problem, I plan to rework the CFE so that
it uses the same algorithm as the analyzer to determine which fields
are promotable. This CL makes a first step towards that goal, by
reworking the analyzer's field promotability algorithm into a form
where it can be shared with the CFE, and moving it to
`package:_fe_analyzer_shared`. Since this required a fairly
substantial rewrite, I went ahead and fixed#52938 in the process.
Fixes#52938.
Change-Id: I9e68f51b3ea9a967f55f15bdc445cc1c0efdabdd
Bug: https://github.com/dart-lang/sdk/issues/52938
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313293
Reviewed-by: Johnni Winther <johnniwinther@google.com>
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Sigmund Cherem <sigmund@google.com>
Commit-Queue: Paul Berry <paulberry@google.com>
I added the entries to dart.dev/resources/glossary in site-www/#5066.
I was only able to test whether the links worked by putting them in the same place in the diagnostics markdown on site-www, which worked. So, I think the way I did it in generate.dart should work but if anyone knows otherwise please let me know!
There were also a couple small changes from my last CL on the diagnostic messages that I didn't get in before merging, so those are in here too (https://dart-review.googlesource.com/c/sdk/+/309780).
Change-Id: Ie5561fa72c7f99f5c86d2112294edccee41d4544
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313960
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Marya Belanger <mbelanger@google.com>
Change-Id: Id4b9fb27990dd02f4c1d94faf11c6e25f056a5a7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313221
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Samuel Rawlins <srawlins@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Each phase now only has a single interface, instead of several of them. This
reduces the number of classes dramatically and also reduces the number of
objects actually sent over the wire.
It also means fewer things to name and a less polluted namespace.
Change-Id: Ib84b76ac4c0a04abfac5fd5650a228046b1bf1d4
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313721
Reviewed-by: Bob Nystrom <rnystrom@google.com>
Commit-Queue: Jake Macdonald <jakemac@google.com>
To be removed after Dart SDK 3.1 is published, and the analyzer
SDK constraints updated.
See
https://dart-review.googlesource.com/c/sdk/+/254903
Change-Id: I2c3321d991cd3e123f08c5a360487362be16f258
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313920
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Samuel Rawlins <srawlins@google.com>
Minor changes to last few 3.0 diagnostic message changes (late follow up). Mostly just a practice CL for me, after setting my sdk environment up for the first time.
Fixes https://github.com/dart-lang/site-www/issues/4740
Change-Id: I8f6871a270089627538928dd95bcbf38a29b74e7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/309780
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Marya Belanger <mbelanger@google.com>
This allows changing the in-process tests to use Windows-style paths on macOS for convenience.
Change-Id: I0e85a4f8e831471925b8308ad348f75b6867a53b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313385
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
This reverts commit fb9d0e6dc3.
Reason for revert: Performance regression https://github.com/dart-lang/sdk/issues/52922
Original change's description:
> Tweaks for mixin inference.
>
> I started initially with the idea to move 'mixinInferenceCallback'
> to `ClassElementImpl`, but then realized that we do it for enums too.
> So, it should stay in `InterfaceElementImpl`.
>
> I left with code that I think slightly modernized, so decided to
> send it for review.
>
> Change-Id: Ib990392dc4985a71ffba1f4080237872d9a65ad2
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312521
> Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
> Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Change-Id: I33098d8509319e9d6e62d7f5174cae2c05977436
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313440
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Samuel Rawlins <srawlins@google.com>
Reviewed-by: Jaime Wren <jwren@google.com>
Commit-Queue: Samuel Rawlins <srawlins@google.com>
Added helpers `_field` and `_localVar` to allow tests to test their final fields and local const variables in the const tests.
The rest of the CL moves away from `_evaluateConstant` helpers (to avoid recomputing constants), to `_topLevelVar` or equivalent helpers which grabs the existing evaluation result.
This CL adds to the goal making all constant tests consistent and to avoid unnecessary const computations in the tests.
Change-Id: I508483714a51e5d060286256657ae460b65787c7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312889
Reviewed-by: Nate Bosch <nbosch@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
These become a single property of ExecutableElement, MethodElement.
No new visitors, no new nodes and elements.
Change-Id: Ic523e9d51356e5a8d2a6cd62cd508344f561ffb0
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312960
Reviewed-by: Samuel Rawlins <srawlins@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Right now `utf8.encode()` has a static return type of `List<int>`
due to extending `Encoding` (which extends `Codec<String, List<int>>`).
We cannot easily change `Encoding` to extend `Codec<String, Uint8List>`
because that would also change `utf8.decode()` to require `Uint8List`
which would be a breaking change.
So instead we override `utf8.encode()` to have more precise return type.
Some parts of our SDK are run using the checked-in SDK, so it cannot
rely on the changed return type yet (until checked-in SDK is rolled).
So we use `const Utf8Encoder().convert()` as a temporary change, as
that already has `Uint8List` return type.
Issue https://github.com/dart-lang/sdk/issues/52801
TEST=ci
CoreLibraryReviewExempt: More precise return type for existing API
Change-Id: I2861d1f0eb3d292d8e3ec8437c0d441a2d2bd193
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/254903
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Lasse Nielsen <lrn@google.com>
So, that we don't have to copy all class members for 99% cases when
there are no augmentaitons.
Change-Id: I5c225905a0460ce5162543553e207a2d4412ecb9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/313082
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Samuel Rawlins <srawlins@google.com>
Print type in `dart_object_printer.dart`.
Use `assertDartObjectText` instead of whatever was being used before.
Change-Id: Ifdfb6343013bb5fe654bcff870a9ff83b155402f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312885
Commit-Queue: Kallen Tu <kallentu@google.com>
Reviewed-by: Bob Nystrom <rnystrom@google.com>
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
We walk augmentations in the right order, so can update update
necessary data structures as we go.
Change-Id: Ibf3409658009b82a960caaacd538fcd6c779b007
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312883
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
So, we can read it more naturally, first the declaration, then its
augmentations.
Change-Id: I21805969bff6874660500d4477b39ce5aba7484a
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312880
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Made more of the tests consistent in `evaluation_test` (slowly, but surely).
Change-Id: I515c4d6d81cedcc1d8d20080e231ac1e0afb5ec7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312200
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Reviewed-by: Konstantin Shcheglov <scheglov@google.com>
Commit-Queue: Kallen Tu <kallentu@google.com>
This code was older, and was a prototype of NodeTextExpectationsCollector,
but we have it now, it is better, and we can de-duplicate.
Change-Id: I428abb04687f2e3ff5142907a71fcf80e460b830
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312700
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
We can use ElementPrinter now for printing elements as references.
Change-Id: Ie75dc96940cf3a1e60f87e34801494f55145901f
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312350
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Add ClassAugmentationElementImpl, no any tests yet because it
required parsing, AST, etc. And it is also not ready yet.
Change-Id: I601141545e9bf0638771f28bb9498be9de664659
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312345
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
I started initially with the idea to move 'mixinInferenceCallback'
to `ClassElementImpl`, but then realized that we do it for enums too.
So, it should stay in `InterfaceElementImpl`.
I left with code that I think slightly modernized, so decided to
send it for review.
Change-Id: Ib990392dc4985a71ffba1f4080237872d9a65ad2
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/312521
Commit-Queue: Konstantin Shcheglov <scheglov@google.com>
Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
