| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Changing some code to not use collect on top of a possibly changing
filesystem. Anonymous partial functions used with collect risk match
errors if guard results ever change, due to the guard being run twice
(isDefinedAt and then apply). No review.
|
|
|
|
|
|
| |
Primarily a cleanup of the organization of keywords for the scala and
java parsers, and a few other Name-related items. No review.
|
|
|
|
|
|
| |
Securing some of the blessings of whitespace for the sbt build. No
review.
|
|
|
|
|
| |
trying to fix CyclicReference errors for type completion
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Some overdue organization and cleanups in the repl. No review.
|
|
|
|
|
|
|
| |
Another naked assert is brought up on obscenity charges. That second
argument is not just a conversation piece, it can be used to communicate
data! No review.
|
|
|
|
|
|
| |
A few minor hashcode tidbits I had stored away from previous hashcoding,
and some fixups in treedsl. No review.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Since we generate no invalid generic signatures at the moment, I decided
to strike while the iron is hot and turned on warnings for when the
sig parser fails. It is likely we still generate bad signatures in
some cases, but they're no longer uncovered by compiling the compiler:
hopefully the community will help us out by reporting any found in the
wild. (The issued warning helpfully suggests exactly this.) No review.
|
|
|
|
|
|
|
|
|
| |
In a code review performance for the ages, I found an instance of the
bug being fixed in the commit under review still going strong elsewhere
in the file. Since the function with the bug wasn't actually called from
anywhere I fixed it by deleting it. No review could possibly improve
upon that.
|
| |
|
|
|
|
|
|
| |
Opening packages when browsing gets rid of spurious presentation co
mpiler errors.
|
|
|
|
|
|
| |
workaround for see #3936, allows building everything inside intellij. no
review
|
| |
|
|
|
|
|
|
| |
rename .classpath to classpath.SAMPLE and include .classpath in the
gitignore.SAMPLE. no review.
|
| |
|
|
|
|
|
|
| |
removed a activeLocks tracking in Symbols, since no longer needed for
presentation compiler.
|
|
|
|
|
|
| |
Imporved interrupt handling, which should avoid further "tree not found"
exceptions. It also seems to make hyperlinking much faster.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
One last attempt at getting the presentation compiler pass some tests.
no review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Another thing we are all a bit too used to is having lots of unreadable
HTML in our code. This is key, central documentation and we all hum to
ourselves and look past it rather than risk being assaulted by a swarm
of angle brackets. That's why the erasure docs can still talk about
BoxedAnyArray and x.asInstanceOf$erased. They could be talking about who
killed JFK, we wouldn't know the difference.
I also addressed some questions from decades past like "why is this
here?" and "!!! needed?" One thing I don't know is whether
{scala#Int)Array[T] is intentional notation or the aftermath of a
tragic punctuation accident.
How about I mass fix all the HTML in trunk? Wouldn't that be great? Just
say the word... no review.
|
|
|
|
|
|
| |
Conflicts:
src/library/scala/concurrent/SyncVar.scala
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
replaced the epic outer.outer.outer by something more robust that goes
to the context corresponding to the enclosing class
nextOuter determines which context is searched next for implicits (after
`this`, which contributes `newImplicits` below) in most cases, it is
simply the outer context if we're owned by a constructor, the actual
current context and the conceptual context are different when it comes
to scoping: the current conceptual scope is the context enclosing the
blocks that represent the constructor body (TODO: why is there more than
one such block in the outer chain?)
review by odersky, extempore
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
closes #2741, #4079: pickling now ensures that a local type param with
a non-local owner, which will thus get a localized owner, will only get
a class as its localized owner if its old owner was a class (otherwise,
NoSymbol) this ensures that asSeenFrom does not treat typerefs to this
symbol differently after pickling.
todo: should we pro-actively set the owner of these type params to
something else than the type alias that they originate from? see notes
in typeFunAnon
review by odersky
|
|
|
|
|
|
| |
review by extempore
(because of recent sin against this holy principle)
|
|
|
|
|
| |
bug. I think this will fix it. No review.
|
|
|
|
|
|
|
|
|
|
|
|
| |
partest --grep has never worked quite right (at least not in the
"sideported" version.) Now it looks harder for matching tests: the check
file or any java or scala source file can match. And directory based
tests will be properly included. Try it out:
test/partest --grep java
No review.
|
|
|
|
|
|
|
| |
SyncVar should not swallow interrupts and should be robust against
spurious wakeups. Closes #4094, review by prokopec. (What did you mean
in your comment that it is vulnerable against the take method?)
|
|
|
|
|
|
|
|
| |
is right for the situation, but positions are not my strongest thing. I
used atPos on the entire new block and marked it with the position of
the expression (+ makeTransparent.) If that sounds suboptimal to anyone
let me know. Closes #4182, no review.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A new murmur3 hashcode implementation contributed by rex kerr. I'm
really happy with it. There is benchmarking code included but you can
use the pudding for proof: the compiler compiling itself is clearly
faster. I deleted the existing murmur2 implementation which has never
left trunk in a release.
There remain some possible points of improvement with inlining and/or
synthetic method generation, but this is a major improvement over the
status quo. Closes #2537, review by prokopec.
|
|
|
|
|
|
|
|
| |
Skip getClass (signature changes between JDK 1.5 and 1.6) and resolve
encoding issue of unicode method name. Disabled part of a test that was
not handled correctly by the presentation compiler (yet). Added correct
encoding for the presentation compiler test. no review.
|
|
|
|
|
|
| |
Added methods askStructure and askLoadedTyped, which should replace any
calls to askType. Next: remove askType.
|
|
|
|
|
| |
Stripped one more embarrasing absolute path from test output. no review.
|
|
|
|
|
|
| |
Increased the timeout for presentation compiler tests and stripped
aboslute paths from the check file. no review.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
made DivergentImplicit a def so we see what goes wrong lately in Eclipse.
|
|
|
|
|
|
| |
Added log message to tell what version of the presentation compiler is
running.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The AnyVal types become source files instead of polite compiler
fictions.
!! You'll need a serious "ant all.clean" now. !!
As of this commit the system is fully bootstrapped and the
synthetic code eliminated: only the source files remain. The
sort-of-AnyVal-companions in scala.runtime.* have all been eliminated
because the actual companions can do everything; deprecated vals in
the scala.runtime package object point to the companions. This left
AnyValCompanion as the only AnyVal related thing in the runtime package:
that made little sense, so I deprecated and moved it as well.
Starr is based on r24066 plus this commit. Closes #4121. Review by rytz,
odersky.
|
|
|
|
|
|
|
| |
Tackled source generation from a new angle rather than letting genprod
get any more creaky. This code generates the AnyVal source files (but
those are not included here.) No review.
|
| |
|