| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
[recommit] Backend optimization: Inline exception handlers. Review by
dragos.
|
|
|
|
|
|
| |
Literals now take Any as argument (used to represent free values in
reified trees).
|
| |
|
|
|
|
|
| |
New utilitly method to create scope with some elements.
|
|
|
|
|
|
|
|
|
| |
Major rewrite of the testing infrastructure for the presentation
compiler. Added several new tests that will be part of the nightly
build. Once the move to SBT is completed I will look into how to extract
the test infrastructure (as it should really not be living in the
compiler codebase). Review by dragos
|
| |
|
| |
|
|
|
|
|
|
| |
Improved the error message given when a concrete method implementation
doesn't match the abstract one. No review.
|
|
|
|
|
|
|
| |
Removing the code which has been deprecated since 2.8.0. Contributed by
Simon Ochsenreither, although deleting code is such fun one hesitates to
call it a contribution. Still, we will. Closes SI-4860, no review.
|
|
|
|
|
|
|
| |
Fixed a method which got hosed in the reflection refactor because the
meaning of "this" changed. This is a perfect opportunity to yell "What
is the meaning of this!" No review.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A conceivably pretty bad performance bug in builders.
SI-4821 pointed out that ArrayBuffer's ++ checks for a cheap size method
by matching on IndexedSeq, but mutable.IndexedSeq, so all immutable
collections are thrown in the same group as linear seqs. I went looking
for other examples of this and found them, in key classes like Builder.
The "type shadowing trap" is a serious issue in the collections. Closes
SI-4821, no review.
|
|
|
|
|
|
| |
Reverted r25440 (update of forkjoin library) until all build problems
are resolved. No review.
|
|
|
|
|
|
| |
Split up adapt in some smaller methods; added code lifting for
non-function types to it.
|
|
|
|
|
| |
Added debug output to track down Fregor's "failing to lift" problem.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Pulled up the SuperMixin that calls closeOperations for frames into
window and let dialogs mix it in as well. Closes #3708.
|
|
|
|
|
|
| |
Added GraphicsConfiguration to window, frame, dialog constructors.
Closes #3853.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
| |
Turned up the defense against compiler exceptions in the repl. And gave
in and disabled an assertion I should have disabled two years ago.
Closes SI-4874, no review.
|
|
|
|
|
|
|
|
|
|
|
| |
Expanded the range of a warning, and made suppressed warnings visible.
Modified the positioning of "permanently hidden" errors so that when
there is more than one, the later ones are not ignored. Also changed the
error suppression code to emit the error anyway if -Ydebug was given (it
is prefixed with "[suppressed] ".) Since I can't be the only one who
wondered where his errors were vanishing to. No review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Reversed the values of "is" and "is not" in recent for comprehension
deprecation.
DO NOT BLOW HATCH
REPEAT
DO NOT BLOW HATCH
"Roger! Hatch blown."
Events reveal it was all baby, no bathwater. It turns out that the
specification is merely a document, not infallible holy writ as we
had all previously believed. So it is not the ABSENCE of val in a for
comprehension assignment which is deprecated, it is the PRESENCE of val.
Summarizing again, more accurately perhaps:
for (x <- 1 to 5 ; y = x) yield x+y // THAT's the one
for (val x <- 1 to 5 ; y = x) yield x+y // fail
for (val x <- 1 to 5 ; val y = x) yield x+y // fail
for (x <- 1 to 5 ; val y = x) yield x+y // deprecated
No review.
|
|
|
|
|
|
| |
Optimizations for Seq's implementations of sequence search algorithms.
Contributed by Rex Kerr. Closes SI-4828, no review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Moved the classes and objects which are defined in package objects out.
In principle this is something you should be able to do. In practice
right now it means bugs, to no advantage.
I also deprecated RangeUtils, an unused, undocumented trait in the
immutable package. It seems like there is a ton of stuff in the public
API which should not be in the public API. It's really tedious having to
go through a whole deprecation cycle to dispose of what could have been
an internal-only class (and was presumably intended as such given the
the absence of documentation.)
No review.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Adds documentation to the scala.math package object. Contributed by
Christian Krause. No review.
|
|
|
|
|
|
| |
Fix java signature generation for traits: no classes as parents. Closes
SI-4891, review by grek.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Don't discard deprecation/unchecked warnings regardless of settings.
Changed warnings code to accumulate them rather than thoughtlessly
discarding them and issuing its well-known taunt. In the repl you can
take advantage of this with the :warnings command, which will show the
suppressed warnings from the last line which had any. Be advised that at
the moment it has some issues: unchecked warnings aren't making it out,
and near repl startup neither are deprecation warnings, so don't open a
bunch of tickets please.
References SI-4594, no review.
|
|
|
|
|
|
| |
Big cleanups in the tracing utility and its usage for signature tracing.
To assist in my current java signature thrill-o-rama. No review.
|
|
|
|
|
|
|
| |
Removed unused import in collection.immutable.WrappedString.
No review.
|
|
|
|
|
|
|
|
|
| |
Looks like there was a little baby in the parser bathwater, or at least
some bathwater still in the general shape of a baby. Apparently both
old forms weren't deprecated. The spec does not suggest that the 'val'
in for comprehension assignment is optional, but clearly it can't go
anywhere for a while yet regardless. Will seek clarification. No review.
|
|
|
|
|
|
|
|
|
|
|
| |
Made the thread behavior of the repl a little bit configurable. Every
line will run in the same thread:
scala -Dscala.repl.no-threads
This mechanism is likely to change. Repl is obscenely overdue for a
config file. No review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When I tried to remove the old for comprehension syntax in r24958, I
managed to instead only remove the deprecation warnings. No wonder it
went so smoothly. Sorry to get your hopes up only to execute you, old
syntax. Now the for comprehensions have to get it right:
for (x <- 1 to 5 ; y = x) yield x+y // nope
for (val x <- 1 to 5 ; y = x) yield x+y // nope
for (val x <- 1 to 5 ; val y = x) yield x+y // nope
for (x <- 1 to 5 ; val y = x) yield x+y // that's the one
No review.
|
|
|
|
|
|
|
| |
Fixing all the tests and source which still use the old for
comprehension syntax with vals where there are no vals and no vals where
there are vals. No review.
|
|
|
|
|
|
|
|
|
|
| |
Fix for a bug in CharArrayReader which made triple quoted strings
fail to parse sometimes. Note: when the temptation strikes to adjust
for special cases by letting the regular case happen and subsequently
attempting to fix the ball of mutation by selectively applying what
seems like the inverse operation, please consider the possibility that
this is not the optimal approach. Closes SI-4785, no review.
|
|
|
|
|
|
| |
Better error message for case class/object match confusion. Closes
SI-4879, no review.
|
|
|
|
|
|
| |
Avoid some spurious errors after a cyclical reference error. Closes
SI-2388, no review.
|
|
|
|
|
|
| |
Improved structural type error messages, and other error message related
boosts. Closes SI-4877, review by odersky.
|
|
|
|
|
|
|
|
| |
Fixed bug in the disambiguation of f(foo='bar') style method calls in
the presence of overloading, parameterization, and by-name arguments.
Took the opportunity to clean things up a little bit. Closes SI-4592,
review by rytz.
|
|
|
|
|
|
|
| |
Disallowed implicit modifier on auxiliary constructors, since it either
silently accomplishes nothing or crashes the compiler. If it should do
something useful let me know. Closes SI-4882, review by odersky.
|
|
|
|
|
|
|
|
| |
Don't want to chase NPEs around for the rest of my life. Created
"NoCompilationUnit" and "NoSourceFile" objects to represent not-present
versions of these items. Seems a lot better than null. References
SI-4859, got past NPE only to uncover the actual problem. No review.
|