| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
{{{
scala> def g { case class C(); object C; }
<console>:5: error: C is already defined as (compiler-generated) case class companion object C
def g { case class C(); object C; }
^
}}}
review by odersky
|
| |
|
|
|
|
|
|
| |
documentation for scala.collection package, contribution by Dave
Copeland. close #3793, no review.
|
|
|
|
|
|
| |
documentation for scala.xml.pull, contribution by Dave Copeland. close
#3786, no review.
|
| |
|
| |
|
|
|
|
|
|
| |
Removed the presumably unintentional empty "1" file which showed up on
monday and which ocd-me cannot handle. No review.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
(a) The bytecode that Scala.NET emitted had a tough time in
passing peverify due to valuetypes (aka structs) and their related
managed-pointer types. With these changes (details in [1] and [2])
external APIs exposing valuetypes can be used, yet the extra step of
supporting defining valuetypes in Scala programs has been left for
later. Supporting the unsigned integral valuetypes (used, among others,
by IKVM) is also pending.
(b) A very first step towards generics can be found in
TypeParser.parseClass, for the time being commented out (search for
the label "TODO CLR generics"). It's commented out because without
CLRManifests codegen won't work as expected. Details in [3].
review by rytz
Refs:
[1]
http://lamp.epfl.ch/~magarcia/ScalaCompilerCornerReloaded/2010Q3/Bootstr
apping3.pdf
[2]
http://lamp.epfl.ch/~magarcia/ScalaCompilerCornerReloaded/2010Q3/Bootstr
apping4.pdf
[3]
http://lamp.epfl.ch/~magarcia/ScalaCompilerCornerReloaded/2010Q2/SigToTy
pe.pdf
|
|
|
|
|
|
| |
Fix and test case for broken linked lists, contributed by Lucien
Pereira. Closes #3361, no review.
|
| |
|
| |
|
|
|
|
|
|
| |
in refchecks, visit the qualifier of irrefutable filters. close #3773.
review by moors.
|
| |
|
| |
|
|
|
|
|
|
| |
More robust icode loading and fixed an issue when dead-code eliminating
closures that have live static fields.
|
| |
|
|
|
|
|
| |
Renaming and small tweak in testing.Benchmark for the output format.
|
|
|
|
|
|
| |
Various fixes to optimizations, and mainly much better heuristics for
inlining. Now even the compiler is some 3% faster! Yeah!
|
|
|
|
|
| |
Fixed various issues with Icode, mainly fixing empty/open blocks.
|
| |
|
|
|
|
|
|
| |
Finally look at the right place for inherited methods that are
effectively final.
|
| |
|
| |
|
|
|
|
|
| |
Fix for not looking into Java types when specializing.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Don't box class parameters when passed to the super constructor, from
specialized subclasses.
|
| |
|
|
|
|
|
|
| |
Added an overridable setUp method that is called prior to each
benchmarking (but whose execution time is not measured).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
@implicitNotFound(msg="Custom error message that may refer to type parameters ${T} and ${U}")
trait Constraint[T, U]
whenever an implicit argument of type Constraint[A, B] cannot be found,
the custom error message will be used, where the type arguments are
interpolated in the obvious way
note: if the msg in the annotation references non-existing type params,
a warning is emitted
the patch also cleans up annotation argument retrieval (moved it to
AnnotationInfo from Symbol)
review by odersky
|
|
|
|
|
| |
no review
|
|
|
|
|
|
|
|
|
|
| |
cloneSymbol now preserves privateWithin -- need to reset it explicitly now when before it was assumed to be not to be carried over
rewrote accessibility in overriding checks so they're more readable, but
hopefully with same semantics
review by odersky
|
|
|
|
|
|
|
| |
Discovered ListMap.++ returns a Map instead of a ListMap. Does
preserving binary compatibility mean we can't fix this sort of thing?
Fixing for now, inquiring via: review by odersky.
|
|
|
|
|
|
|
|
| |
- building continuation plugin
- building and replacing fjgb, msil and forkjoin
- building and replacing starr
- layer cleaning tasks
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
[scaladoc] Merges Petr's change to Scaladoc from the Colladoc branch:
bugfixes, improvements and refactorings that make it possible to extends
Scaladoc into Colladoc.
- Layout of index uses jQuery UI panel instead of frameset.
- Search boxes have correct width on Firefox.
- Scaladoc models the original, untransformed source.
- Various internal bugfixes & refactorings.
Review by dubochet.
|
|
|
|
|
|
|
| |
the added flexibility'll cost ya one ClassManifest
review by phaller
|
|
|
|
|
| |
no review
|
| |
|
| |
|
| |
|