| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Add an example benchmark for OpenHashMap.
|
| |\ \ \ \ \ \
| | | | | | | |
| | | | | | | | |
SI-4625 Recognize App in script
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Fixed the warning when main module is accompanied by snippets.
Minor clean-up so even I can follow what is returned.
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
It's pretty confusing when your script object becomes a local
and then nothing happens. Such as when you're writing a test and
it takes forever to figure out what's going on.
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
In an unwrapped script, where a `main` entry point is discovered
in a top-level object, retain all top-level classes.
Everything winds up in the default package.
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
Cheap name test: if the script object extends "App",
take it for a main-bearing parent.
Note that if `-Xscript` is not `Main`, the default,
then the source is taken as a snippet and there is
no attempt to locate an existing `main` method.
|
| |\ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | | |
SI 9766 - allow ++ on empty ConcatIterator
|
| | |/ / / / / / |
|
|\| | | | | | | |
|
| | |_|/ / / /
| |/| | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
The kv field of scala.collection.immutable.HashMap.HashMap1 can be null. This
commit corrects the behavior of updated0 (which is on call path for merged) to
work in such cases, instead of throwing NPE.
Commit contains regression test.
|
|\| | | | | |
| |_|_|/ / /
|/| | | | | |
|
| |/ / / /
| | | | |
| | | | |
| | | | | |
Follow-up for https://github.com/scala/scala/pull/4117
|
| |_|_|/
|/| | |
| | | |
| | | |
| | | |
| | | |
| | | | |
We delegate `String`'s extension methods `toInt`, `toFloat`, etc to corresponding
methods in the Java standard library. These differ in the way they handle
whitespace in the original string.
This commit documents and tests the current behaviour.
|
|\ \ \ \
| | | | |
| | | | | |
Improvements to deprecations related to `since` parameter
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
|\ \ \ \ \
| | | | | |
| | | | | | |
SI-9382 Privatize enhanced x in Tuple2Zipped.Ops
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Consolated JUnit tests and heeded comment about private def and
code beauty.
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Probably there should be an Abide rule to avoid leaking
the "underlying value" of a value class. The spec or SIP defines
"underlying type" but doesn't mention the underlying value.
The argument for concealing the member is that it is redundant
and makes autocompletion results harder to read. Also, possibly
an additional implicit might want to add a member so-named.
|
|\ \ \ \ \ \
| |_|/ / / /
|/| | | | | |
SI-9794 Error advice uses decoded method name
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
So much work went into polishing this error message,
it's worth buffing the method name when it's an operator.
The message now says `+` instead of `$plus`.
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | | |
SI-2712 Add support for partial unification of type constructors
|
| | | | | | | |
|
| | | | | | | |
|
|\ \ \ \ \ \ \
| | | | | | | |
| | | | | | | | |
Debug flag to print a summary of the inliner's work
|
| |/ / / / / /
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Example output below. Note that inlining List.map fails because the
trait forwarder uses `INVOKESPECIAL` for now, will change with pr 5177.
$ cat Test.scala
class C {
def foo = Map(1 -> 'a', 2 -> 'b')
def bar(l: List[Int]) = l.map(_ + 1)
}
$ qsc -Yopt-log-inline _ -Yopt:l:classpath Test.scala
Inlining into C.foo (initially 36 instructions, ultimately 72):
- Inlined scala/Predef$ArrowAssoc$.$minus$greater$extension (8 instructions) 2 times: the callee is annotated `@inline`
Inlining into C.bar (initially 12 instructions, ultimately 12):
- Failed to inline scala/collection/immutable/List.map (the callee is a higher-order method, the argument for parameter (bf: Function1) is a function literal): The callee scala/collection/immutable/List::map(Lscala/Function1;Lscala/collection/generic/CanBuildFrom;)Ljava/lang/Object; contains the instruction INVOKESPECIAL scala/collection/TraversableLike.map (Lscala/Function1;Lscala/collection/generic/CanBuildFrom;)Ljava/lang/Object;
that would cause an IllegalAccessError when inlined into class C.
|
|\ \ \ \ \ \ \
| |_|/ / / / /
|/| | | | | | |
Fully qualify types in REPL generated code
|
| | | | | | | |
|
|/ / / / / /
| | | | | |
| | | | | |
| | | | | | |
Keep -Yopt-inline-heuristics and -Yopt-trace unchanged
|
|\ \ \ \ \ \
| |_|/ / / /
|/| | | | | |
SI-9522 release key reference when deleting from OpenHashMap
|
| |/ / / /
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This sets the key field in the hash table entry to its default value
when an entry is deleted, so as not to unexpectedly retain an object
reference, leading to a memory leak.
Also includes incidental changes to the slot location algorithm that
reduce the number of deleted entries.
|
|\ \ \ \ \
| | | | | |
| | | | | | |
SI-8044 Allow binding backquoted varid in patterns
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Allows arbitrary identifier in `X @ pat`, including
non-varids. This goes to regularity.
Users of this syntax are not likely to be confused
by the "backquoted var id is stable" rule.
Also for sequence pattern, `X @ _*`.
|
| | | | | | |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Previously, a varid could not be backquoted, so that it was not
possible to introduce variables with names such as `type` in a
match expression.
This commit allows backquoted varids in `case x @ _` and
`case x: Int`. In neither position is a stable id accepted,
that is, an id with leading uppercase.
Therefore, this commit merely relaxes the backquoted varid to
be taken as a normal varid in these contexts.
|
|\ \ \ \ \ \
| |_|/ / / /
|/| | | | | |
SI-9656 Distinguish Numeric with step type
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
For Range and NumericRange, toString will indicate the step
if it is not 1.
Additionally, indicate empty ranges and ranges which are not
"exact".
For a "mapped" range, used by `Range.Double`, toString
includes the underlying range and the simple type of the step
(to distinguish Double from BigDecimal).
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | | |
Check left side of an assignment expression for errors; fixes SI-9781
|
| | |/ / / /
| |/| | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
`convertToAssignment` is triggered on a type error but it doesn't seem
to really care what the error is as long as the expression can be
converted to an assignment expression.
This patch fixes that by checking whether the qualifier of the selection
contains any errors before deciding to continue with the conversion.
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | | |
SI-9760 Fix for higher-kinded GADT refinement
|
| | | | | | | |
|
| | | | | | | |
|
| | | | | | | |
|
|\ \ \ \ \ \ \
| | | | | | | |
| | | | | | | | |
SI-9776 Fix type of PriorityQueue.newBuilder and improve performance
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
- Fix the return type of mutable.PriorityQueue.newBuilder to be
Builder[A, PriorityQueue[A]] rather than PriorityQueue[A].
- Improve performance of bulk inserts from O(N log N) to O(N),
primarily in the builder, ++=, and reverse. These changes
indirectly benefit the many other methods that use the builder or ++=.
- Improve performance of clone.
- Fix SI-9757 space leak in dequeue.
|
|\ \ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | | |
Improve Future documentation (+ minor code cleanups)
|
| | | | | | | | | |
|
|\ \ \ \ \ \ \ \ \
| | | | | | | | | |
| | | | | | | | | | |
SI-9121 test case (fixed in new optimizer), SI-9179 test case
|
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
Also adds a mising phase travel in the backend. A comment already
points out why it's necessary, but it was actually forgotten.
|