Sync with the stable documentation branch #17614
Commits on Jan 23, 2023
Commits on Jan 27, 2023
Commits on Feb 15, 2023
-
Added jpath check to
ClassLikeSupportgetParentsAsTreeSymbolTuplesFixes #15927 Check for whether the non-scala3 parent exists before checking the start and end of the span to confirm whether the span exists in getParentsAsTreeSymbolTuples.
Commits on Feb 17, 2023
-
-
-
-
-
-
-
-
-
-
-
-
Add support for disabling redirected output in the REPL driver for us…
…age in worksheets in the Scala Plugin for IntelliJ IDEA - Calling `setOut/setErr` in a concurrent environment without any synchronization (such as the Scala compile server in the Scala Plugin for IntelliJ IDEA, which is used to execute Scala 3 worksheets) can lead to unpredictable outcomes where the out/err streams are not restored properly after changing. - This change adds a new default method `redirectOutput` which can be overriden by others to control the redirecting behavior of the REPL driver.
-
-
Avoid timeouts in community–build-C
Increased timeout due to timeouts when running on dotty community build
-
-
Commits on Feb 21, 2023
-
Fix HK quoted pattern type variables
The issue was in the encoding into `{ExprMatchModule,TypeMatchModule}.unapply`. Specifically with the `TypeBindings` argument. This arguments holds the list of type variable definitions (`tpd.Bind` trees). We used a `Tuple` to list all the types inside. The problem is that higher-kinded type variables do not conform with the upper bounds of the tuple elements. The solution is to use an HList with any-kinded elements. -
Fix HK quoted pattern type variables (#16980)
The issue was in the encoding into `{ExprMatchModule,TypeMatchModule}.unapply`. Specifically with the `TypeBindings` argument. This arguments holds the list of type variable definitions (`tpd.Bind` trees). We used a `Tuple` to list all the types inside. The problem is that higher-kinded type variables do not conform with the upper bounds of the tuple elements. The solution is to use an HList with any-kinded elements. Backport of #16907
Commits on Apr 13, 2023
-
Fix caching issue caused by incorrect isProvisional check
A static TypeRef can still be provisional if it's currently being completed (see the logic in `Namer#TypeDefCompleter#typeSig`). Fixes #16950.
-
-
-
Commits on Apr 14, 2023
-
- Ignore synthetic local private - Update test suit
-
Register usage of symbols in non-inferred type trees in CheckUnused
fixes lampepfl#16930
-
Traverse annotations instead of just registering
- Traverse the tree of annotations - Update test suits
-
- Do not report parameter of accessors - Update test suit
-
Improve override detection in CheckUnused
- CheckUnused detects override from base type in addition of `override` flag - Update test suit
-
-
-
-
-
Add failsafe for a case where prefixes in CheckUnused/prepareIndent f…
…ormed an infinite cycle
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Commits on Apr 17, 2023
Commits on Apr 18, 2023
-
Check the status of coursier download in CoursierScalaTests.scala
This should provide more insight on why #17119 happens.
-
-
Commits on Apr 26, 2023
-
-
-
-
Wunused: Check if symbol exists before isValidMemberDef check
closes lampepfl#17309
-
Wunused: Include import selector bounds in unused checks
closes lampepfl#17314
-
-
Commits on Apr 27, 2023
Commits on May 9, 2023
-
Patmat: Use less type variables in prefix inference
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.
Commits on May 12, 2023
-
-
Revert exact match in overlaps
As suggested by @som-snytt in #17366 (review)
-
Raise a warning instead of an error for a type ascription on a patter…
…n other than a variable or a number literal. This partially reverts the changes from #16150. This change is motivated by not breaking source compatibility for a number of projects in the Open Community Build.
-
* Preserve the more restrictive syntax for typed patterns in the lang…
…uage specification * Make the parser's warning a migration warning
-
-
Commits on May 29, 2023
Commits on May 30, 2023
-
Update docs/_docs/reference/language-versions/source-compatibility.md
Co-authored-by: Jamie Thompson <[email protected]>
-
-
Update source-compatibility.md
reorder the source versions
Commits on Jun 1, 2023
Commits on Jun 21, 2023
-
sort language versions to match natural ordering
also fixes the bullet-point under 3.3 to actually be a bullet point
-
sort language versions to match natural ordering (#18030)
also fixes the bullet-point under 3.3 to actually be a bullet point. (the ordering now matches the enum in https://github.com/lampepfl/dotty/blob/main/compiler/src/dotty/tools/dotc/config/SourceVersion.scala)
Commits on Jun 26, 2023
Commits on Jun 27, 2023
-
Fix example that did not compile and following paragraph (#18065)
Compiling the code on latest main gives: ```scala { val x = f(x: Int, y => // Recursive value x needs type x * ( y + 1 ) + (x + x) ) } ``` Which makes sense as we basically write `val x = f(x)`
Commits on Jun 28, 2023
-
Fix accessibleType for package object prefixes
Making a package object explicit re-computes the denotations of an overloaded method. So it should not be done after we have pruned down those denotations by an accessibility test. We now do it before checking accessibility. Fixes #15821
-
-
-
Commits on Jul 5, 2023
-
Add clause for protected visibility from package objects
We usually have an access rule that the access to a protected member `foo` in class `C` must be from somewhere nested in a subclass of `C`. But that fails if the member is accessed from a package object `p.package`. In that case, the access does not need to be in the same object, it just has to be in package `p`. This clause was previously missing and is now added. Why was this only recently discovered? #18057 fixed an issue where toplevel protected members were always accessible because explicit package object prefixes were added after the accessibility check was done, and would re-establish the previous members without doing an accessibility check. The fix was done by adding package objects first, then doing he rest of the checks. But that also means that protected toplevel objects now get checked as members of their synthetic package object instead of as members of their package. The change here also makes specs2 compile again.
-
-
Commits on Jul 11, 2023
Commits on Jul 14, 2023
-
Update link to point to correct section (#18208)
In the reference, in Erased Definitions, link pointed to the Inline page, even though the content is in Compile Time Operations
Commits on Aug 2, 2023
-
Fix regression with Overloaded methods returning Functions
Before the regression, FunctionOf unapply would not try dealiasing, meaning that an aliased function type would be handled by a general case. To fix that, instead of handling Function types separately when filtering overloaded methods in `resolveOverloaded1`, we allow to fallback to the general case if the previous one returns nothing. Along with fixing the regression, this also improves other cases, one of which was added to the test. Readd a separate FunctionOf case, but with a fallback
Commits on Aug 9, 2023
-
A slightly more conservative version of #14128
Two changes - Fix `hasUpperBound` to work correctly for higher-kinded types - A more conservative fix in `IsFullyDefinedAccumulator`. We now maintain the symmetry that - if variance < 0, we maximize - if variance > 0 (and Nothing is admissible) we minimize - only if variance = 0, we use the upper bound as a tie breaker Previously, we maximized even if variance > 0 if there was an upper but no lower bound. But that was asymmetric since there is no corresponding case where we minimize at variance < 0 if there is a lower but no upper bound. -
-
-
-
Space: Revert how invariant targs are erased to fix regression
The motivating case (i16451) is complicated, because it involves unchecked type arguments. To fix the regression, I'm reverting the fix.
Commits on Aug 10, 2023
Commits on Aug 16, 2023
-
-
Make constructors effectively final
This is mainly a cleanup.
-
Do not compute
protoFormalifparam.tptis emptyThis was accidentally moved before of the `if (!param.tpt.isEmpty)` guard in 0f7c3ab#diff-8c9ece1772bd78160fc1c31e988664586c9df566a1d22ff99ef99dd6d5627a90R1534 Fixes #18276
Commits on Aug 22, 2023
-
Revert "Normalize match type usage during implicit lookup"
This reverts commit 5bafff7.
-
Revert "Normalize match type usage during implicit lookup" on 3.3.1 (#…
…18440) Reverts #17457. This fix caused significant challenges in maintaining the source compatibility. We decided not to include it in 3.3.1. It will still be a part of 3.4 and future Scala Next versions. It is possible that in the future, an improved version of this fix will make it to one of 3.3.x releases.
-
-
Commits on Aug 29, 2023
-
Tweak selection from self types
Previously, we rejected the case where a symbol of a self type selection was private if was not of the enclosing class. But that symbol could shadow a non-private symbol in a base class, so have to treat that case as well. Fixes #18631
-
Revert "Add reflect
defn.FunctionClassoverloads"This reverts commit 9571b42.
-
-
-
Commits on Sep 7, 2023
Commits on Oct 1, 2023
Commits on Oct 2, 2023
-
`-source future` is still required as of 3.3 LTS.