| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|\
| |
| | |
SI-7599 Modify methods to be non-tail-callable ...
|
| |
| |
| |
| |
| | |
This commit modifies methods in the test to be non-tail-callable to
prevent Avian from eliding the stack frames we want to inspect.
|
|\ \
| | |
| | | |
SI-7584 Fix typer regression with by-name parameter types
|
| | |
| | |
| | |
| | |
| | |
| | | |
The enclosed test case exercises by-name closures, which were
the subject of the previous commit. In the process, a spurious
warning was eliminated.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
It regressed in fada1ef6b#L4L614. Partially reverting just
this change restores the correct behaviour:
```
- if (sym.isStable && pre.isStable && !isByNameParamType(tree.tpe) &&
+ if (treeInfo.admitsTypeSelection(tree) &&
```
This patch embeds the check for by-name parameter types into
`TreeInfo.isStableIdentifier`. That code already checks for
`Symbol#isStable`, which exludes direct references to by-name
parameters. But the additional check is required to deal with
by-name parameters in function types, e.g `(=> Int) => Any`.
Open question: should we go further and embed this check in `isStable`?
Currently:
final def isStable = isTerm && !isMutable && !(hasFlag(BYNAMEPARAM)) && (!isMethod || hasStableFlag)
Such function types are an underspecified corner of the language,
albeit one that is pretty useful writing, for example, in the
signature of a lazy foldRight that can operate over infinite
structures:
def foldRight[A, B](fa: F[A], z: => B)(f: (A, => B) => B): B
The next commit subjects them to a little testing.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
introduced in 5b54681: the end position of Postfix operators should
take the operator length into account.
review by @som-snytt
|
|\ \ \
| | | |
| | | | |
SI-7433 Fix spurious warning about catching control throwable
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
In the same vein as SI-6994, we have to be careful not to warn
about synthetic code. In that case, the spurious warnings came
because we warned in the typechecker, which was also called in
erasure. In this case, we are issuing the warning in Uncurry,
so we must be mindful of the pattern matchers translations of
non-trivial catch patterns, which look like:
case (ex8 @ _) => {
<synthetic> val x5: Throwable = ex8;
case11(){
if ({
case14(){
if (x5.$isInstanceOf[NullPointerException]())
matchEnd13(true)
else
case15()
};
case15(){
if (x5.$isInstanceOf[RuntimeException]())
matchEnd13(true)
else
case16()
};
case16(){
matchEnd13(false)
};
matchEnd13(x: Boolean){
x
}
})
This commit detects that `ex8` is synthetic and disables the warning.
|
|\ \ \
| |_|/
|/| | |
SI-7439 Avoid NPE in `isMonomorphicType` with stub symbols. …
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
`originalInfo` can return null for stub symbols; deal with that
as we used to before a regression in 016bc3db.
After this change, we can once again delete A_1.class and still compile
code instantiating B_1. (A_1 is only referred to in a method signature
of B_1 which is not called from our code.)
scala> new B_1
warning: Class A_1 not found - continuing with a stub.
res0: B_1 = B_1@5284b8f9
In practice, this situation arises when someone uses a third
party class that was compiled against other libraries not avaialable
on the current compilation classpath.
|
| | |
| | |
| | |
| | |
| | | |
Looks like partest's confusion about / vs. \ was fixed between
the original Jenkins run of the fix for SI-7292 and its merge.
|
|\ \ \
| | | |
| | | | |
SI-7292 Deprecate octal escape literals
|
| | | | |
|
|\ \ \ \
| | | | |
| | | | | |
SI-6899, prohibit dangerous, useless implicit conversions.
|
| | |/ /
| |/| |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Increase eligibility requirements for implicit conversions,
such that T => U is ineligible if
T <: Null <or> AnyRef <: U
This has the salutary effect of allowing us to ditch 16
ridiculous implicits from Predef, since they existed solely
to work around the absence of this restriction.
There was one tiny impact on actual source code (one line
in one file) shown here, necessitated because the literal null
is not eligible to be implicitly converted to A via <:<.
def f[A](implicit ev: Null <:< A): A = null // before
def f[A](implicit ev: Null <:< A): A = ev(null) // after
As impositions go it's on the tame side.
|
|\ \ \ \
| | | | |
| | | | | |
SI-7364 Allow raw types in parent position in Java sources
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
To make this work, this commit simply restricts parent
type argument inference to Scala source files.
The surrounding code has also been refactored to avoid a var.
|
|\ \ \ \ \
| | | | | |
| | | | | | |
SI-7151 Emit final in bytecode for final inner classes.
|
| | |/ / /
| |/| | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
As we did before a regression in 18efdedfb / SI-5676.
This commit tightens up the condition in which the FINAL
modifier is omitted; it now *only* does this for the module
classes of nested objects.
|
|\ \ \ \ \
| | | | | |
| | | | | | |
SI-5022 Retain precise existentials through pattern matching
|
| | |/ / /
| |/| | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
From the dawn of scalac's existentials, the typer widens
existentials pt-s by substituting wildcard types in places
of existential quantifiers.
In this example:
class ForSomeVsUnapply {
def test {
def makeWrap: Wrap = ???
def useRep[e](rep: (e, X[e])) = ()
val rep = makeWrap match {
case Wrap(r) => r
};
useRep(rep) // error
}
}
the type of `r` is the result of typechecking:
Apply(
fun = TypeTree(
tpe = (rep#12037: (e#12038, X#7041[e#12038]) forSome { type e#12038 })
args = Bind(r @ _) :: Nil
}
This descends to type the `Bind` with:
pt = (e#12038, X#7041[e#12038]) forSome { type e#12038 }
`dropExistential` clobbers that type to `Tuple2#1540[?, X#7041[?]]`,
which doesn't express any relationship between the two instances
of the wildcard type. `typedIdent` sort of reverses this with a call
to `makeFullyDefined`, but only ends up with:
pt = (Any#3330, X#7041[_1#12227]) forSome { type _1#12227; type e#12038 }
I suspect that this existential dropping only makes sense outside of
typechecking patterns. In pattern mode, type information flows from the
expected type onwards to the body of the case; we must not lose precision
in the types.
For SIP-18 friendly existentials, one `dropExistential` is invertable with
`makeFullyDefined`, so this hasn't been such a big problem.
The error message improvement conferred by SI-4515 took a hit.
That might be a good example to consider when reviewing this change:
Does it tell us anything interesting about this `dropExistential`
business?
|
|\ \ \ \ \
| | | | | |
| | | | | | |
Merge JUnit support into master
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
The WeakHashSetTest was written as unit test but put into partest's
`run` category as we were missing direct unit testing support.
That got fixed so moving the test now.
|
| |\ \ \ \ \
| | |_|_|_|/
| |/| | | | |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Add `test.junit` ant target that compiles and runs JUnit tests
found in `test/junit` directory.
Add `scala.tools.nsc.SampleTest` that demonstrates working
testing infrastructure.
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Previously, the `split` method of the `ParTrieMap` iterator threw
an exception when splitting a splitter that iterated over nodes
whose hash codes collide.
This was due to reusing the iterator of the list of colliding keys
rather than creating a new splitter.
This commit changes the `subdivide` method to create a new
iterator using the factory method of the current trie map
iterator rather than returning a `LinearSeqLike` iterator.
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | | |
Add Duration.toCoarsest method
|
| | | | | | | |
|
|\ \ \ \ \ \ \
| |_|/ / / / /
|/| | | | | | |
SI-7479 Make test/files/run/tailcalls.scala pass on Avian
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
This commit modifies the test and check file to use the new diff filter.
|
| | |_|_|/ /
| |/| | | |
| | | | | |
| | | | | |
| | | | | | |
... so that the different results of the test on Avian
can be specified in the check file.
|
|\ \ \ \ \ \
| | |_|/ / /
| |/| | | |
| | | | | |
| | | | | | |
Conflicts:
src/partest/scala/tools/partest/DirectTest.scala
|
| |\ \ \ \ \
| | | | | | |
| | | | | | | |
SI-7505 Test case for pattern matcher + type alias bug
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Progressed along with SI-7214 in acd74cae09.
|
| |\ \ \ \ \ \
| | | | | | | |
| | | | | | | | |
Test for reading JDK 8 (classfile format 52) class files.
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
We had several tests designed to only run if the JDK version was at
least some specified version. This commit refactors that common logic
into DirectTest.
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
This commit includes a test for reading JDK 8 (classfile format 52)
class files, in particular default (aka defender) methods. It uses ASM
to generate an interface with default methods then exercises that
interface from Scala.
Surprisingly no changes are necessary to the Scala code base to support
reading format 52 class files.
Because the test can only run under JDK 8, the JDK version is checked
and the expected output is synthesized for previous versions.
|
| |\ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | | |
SI-6841 Fix bug at the intersection of DelayedInit and named args
|
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
The DelayedInit transformation analyses the constructor to partition
regular initialization from calls to super constructors / trait
initializers.
It failed to find such super calls if they were nested in a Block,
which can happens when using named or default arguments.
This commit makes that code peer into Blocks to correctly partition
the constructor statements.
This change doesn't affect the result of run/t4680.scala, which was
mentioned in nearby comments and which chronicles bugs with DelayedInit
when used in inheritance hierarchies.
|
| |\ \ \ \ \ \ \ \
| | |_|_|/ / / / /
| |/| | | | | | | |
SI-7558 Fix capture of free local vars in toolbox compiler
|
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
It was creating an `ObjectRef[<notype>]` because of a small
bug in `capturedVariableType`.
|
| |\ \ \ \ \ \ \ \
| | |/ / / / / / /
| |/| | | | | | | |
SI-7556 Fix runtime reflection involving ScalaLongSignature
|
| | |/ / / / / /
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Scala type information is stored in classfiles in encoded in a String
in the ScalaSignature annotation. When it is too big for a single
String, it is split into an array of Strings in a different annotation,
ScalaLongSignature.
The enclosed test, with a class containing 3000 methods, uses the latter.
It exposes a bug in the way runtime reflection decodes that data.
It must concatentate and *then* decode, rather that the other way around.
|
|\ \ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | | |
SI-7264 Initialize owner when searching for companion.
|
| | |_|_|_|_|/ /
| |/| | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
From ClassSymbol:
protected final def companionModule0: Symbol =
flatOwnerInfo.decl(name.toTermName).suchThat(sym => sym.isModuleNotMethod && (sym isCoDefinedWith this))
protected final def flatOwnerInfo: Type = {
if (needsFlatClasses)
info
owner.rawInfo
}
Note the call to `rawInfo`; in the enclosed test case, that gives
us back an uninitialized type for the module class of `Foo`, and
consequently we don't find the companion for `Foo.Values`.
This commit forces the initialization of the owning symbol if it
was compiled in a prior run.
In addition, it adds a special case to `Run#compiles` for early
initialized symbols, which start out in life with the wrong owner.
As best as I can see, that complexity stems from allowing early
initialized members *without* return types to be used as value arguments
to the super call, which in turn is needed to infer parent type arguments.
The situation is described a little further in existing comments of
`typedPrimaryConstrBody`.
This bug is essentially another case of SI-6976. See the comments in pull
request of that patch (https://github.com/scala/scala/pull/1910) for
commit archaeology that shows why we're reluctant to force the owner
info more broadly than is done in this commit.
|
|\ \ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | | |
SI-7498 ParTrieMap.foreach no longer crashes
|
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
Previously, the `split` method of the `ParTrieMap` iterator threw
an exception when splitting a splitter that iterated over nodes
whose hash codes collide.
This was due to reusing the iterator of the list of colliding keys
rather than creating a new splitter.
This commit changes the `subdivide` method to create a new
iterator using the factory method of the current trie map
iterator rather than returning a `LinearSeqLike` iterator.
|
|\ \ \ \ \ \ \ \ \
| | | | | | | | | |
| | | | | | | | | | |
SI-7519 Less brutal attribute resetting in adapt fallback
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Prefers `resetLocalAttrs` over `resetAllAttrs`. The latter loses
track of which enclosing class of the given name is referenced by
a `This` node which prefixes the an applied implicit view.
The code that `resetAllAttrs` originally landed in: https://github.com/scala/scala/commit/d4c63b#L6R804
|