Merge 3.3.0 reference #17587
Merged
Merge 3.3.0 reference #17587
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Reduces number of type parameters in boundary from 2 to 1
Note that this does not rewrite to labeled returns yet because of some inlining limitations.
Co-authored-by: Guillaume Martres <[email protected]>
The test was minimized too far and did not actually have a stack size issue. Also added a test when in a local `def` and made the comment for the `lazy val` case more accurate.
… owner (also fix false positive for enum) (scala#16661) This solves the issue scala#16650. The issue is that `member(sym.name)` failed to resolve when a type contains multiple identical name. Here the `dotty.tools.dotc.util` package contains multiple "<init>" (constructor), so `member` fail to resolve the name The solution is to register the owner of the contructor instead of the constructor itself. :warning: I couldn't create a test where this appends, I tested it by `publishLocal` and testing the code in a different project with this code from scala#16650 (also didn't check on the metals build): ```scala //> using lib "org.scala-lang::scala3-compiler:3.3.0-RC1-bin-20230109-f56089b-NIGHTLY" //> using scala "3.3.0-RC1-bin-20230109-f56089b-NIGHTLY" //> using option "-Wunused:all" import dotty.tools.dotc.util.LinearSet @main def run = val a = 123 println("Hello!") ``` ```bash sbt> run [info] compiling 1 Scala source to .../target/scala-3.3.0-RC1-bin-SNAPSHOT/classes ... [warn] -- Warning: ..../src/main/scala/Main.scala:5:29 [warn] 5 |import dotty.tools.dotc.util.LinearSet [warn] | ^^^^^^^^^ [warn] | unused import [warn] -- Warning:.../src/main/scala/Main.scala:10:6 [warn] 10 | val a = 123 [warn] | ^ [warn] | unused local definition [warn] two warnings found [info] running run Hello! ``` ### EDIT Also add a related fix for import generated by enum (false positive)
* Simplify the flow for most common kinds of release/commit validation - compiling and running sources * Allow specifying a range of releases to bisect with `--releases` flag (needed for finding regressions in features introduced after 3.0.0) * Automatically assert correctness of the validation script - it should succeed for the first validated release and fail for the last one * Add dry run mode (`--dry-run` flag) to check the validation script without running the entire bisection * Use `--server=false` in scala-cli by default * this avoids false negatives of validation failing for some compiler versions for which it should succeed but fails because of some zinc/scala-cli related errors * this allows using a non-bootstrapped compiler for testing, which significantly speeds up `publishLocal`; non-boostrapped is now the default, bootstrapped can be turned on with `--bootstrapped` flag
Only check trees if there is an enclosing boundary
We track in a new variable `stackHeight` the number of stack slots that are "on hold" while loading some other tree. When we immediately consume stack elements without using `genLoad` on top, we do not need to update `stackHeight`. We store the target `stackHeight` of labels in their `LoadDestination.Jump`. This way, when we generate a jump, we can compute the difference and drop the excess stack.
The back-end can now handle labeled-returns across stack differences, so we can emit optimized blocks even in those situations.
scala#702 introduced a requirement that self types are closed. This means > If trait X has self type S and C is a class symbol of S, then S also conforms to the self type of C. An example that violates this requirement is ```scala trait X { self: Y => } // error: missing requirement: self type Y & X of trait X does not conform to self type Z of required trait Y trait Y { self: Z => } trait Z ``` But it's no longer clear what the requirement should achieve. If we let the example above typecheck and try to implement X with something like ```scala class C extends X, Y ``` we would at that point get an error saying that `C` does not conform to the self type Z of Y. So it would have to be ```scala class C extends X, Y, Z ``` and this one looks fine. The original change in scala#702 was made to avoid a crash in pending/run/t7933.scala. Unfortunately, we cannot reproduce this anymore since it depends on nsc.interpreter, which is no longer part of Scala 2.13. Since we are no longer sure what the restriction should achieve I think it's better to drop it for now. If people discover problems with code that uses "open" self types, we can try to fix those problems, and if that does not work, would fallback re-instituting the restriction. It's not ideal, but I don't see another way. Fixes scala#16407
Enable the addition of classes from a `MacroAnnotation`: * Can add new `class`/`object` definitions next to the annotated definition Special cases: * An annotated top-level `def`, `val`, `var`, `lazy val` can return a `class`/`object` definition that is owned by the package or package object. Related PRs: * Follows scala#16454
…dled-by-backend Add errorhandling stack handled by backend
This reverts commit 04e7bc1.
In code like:
class Outer:
sealed trait Foo
case class Bar() extends Foo
def mat(foo: Foo) = foo match
case Bar() =>
When in the course of decomposing the scrutinee's type, which is
`Outer.this.Foo`, we're trying to instantiate subclass `Outer.this.Bar`,
the `Outer.this` is fixed - it needn't be inferred, via type variables
and type bounds. Cutting down on type variables, particularly when GADT
symbols are also present, can really speed up the operation, including
making code that used to hang forever compile speedily.
As suggested by @som-snytt in scala#17366 (review)
…n other than a variable or a number literal. This partially reverts the changes from scala#16150. This change is motivated by not breaking source compatibility for a number of projects in the Open Community Build.
…uage specification * Make the parser's warning a migration warning
57 tasks
julienrf
reviewed
May 25, 2023
There was a problem hiding this comment.
I am not sure about the changes in this file. @bishabosha could you please double check?
There was a problem hiding this comment.
this change is ok, (all I see is h3 titles changing to h2)
prolativ
approved these changes
May 26, 2023
bishabosha
requested changes
May 26, 2023
There was a problem hiding this comment.
we really should update https://github.com/lampepfl/dotty/blob/release-3.3.0/docs/_docs/reference/language-versions/source-compatibility.md to list -source 3.3 and -source 3.3-migration modes
I know that we can always post-hoc merge to language-reference-stable, but does the 3.3.0 reference in the version picker only get published once? or is it always published until the next stable release?
|
I don't think we have a version picker for the language reference. It exists only for API docs. |
bishabosha
reviewed
May 29, 2023
Co-authored-by: Jamie Thompson <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.