| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
Fix for interpreter issue with missing newlines wreaking havoc. No
review.
|
|
|
|
|
|
| |
Some minor sensibility improvements to previous patch, and tearing out
some obsolete pieces of scalap. No review.
|
|
|
|
|
|
| |
Making the scaladoc command line help look as pretty as all the others
ones do now. No review.
|
|
|
|
|
|
| |
Incidental work on scaladoc while trying to figure out what is up with
usecases. Did a little rewrite of recent patch. No review.
|
|
|
|
|
|
|
|
|
|
| |
Modified some typer logic to allow annotation arguments for constants
which don't have the form Literal(_). The current logic seems to be
avoided most of the time, but scaladoc breaks when it runs into it. This
closes #4301. I can't figure out from the ticket what the deal is with
#2764 and what is presently happening, but it seems like this patch
could only improve the situation. Review by rytz.
|
|
|
|
|
|
| |
[scaladoc] Scaladoc should expand macros recursively. Closes #3484.
Review by dubochet.
|
|
|
|
|
|
|
| |
Warn about unqualified names in scaladoc links because they don't work.
I think it would be better to just make them work, but the way things
are structured it looks involved. Closes #3696, no review.
|
|
|
|
|
|
| |
Addressing most of the warnings revealed by the patch to warn about
unknown scaladoc variables. Updated and reran genprod. No review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Issue warning when doccomments have $variables which go unfulfilled.
Started with patch by dmharrah. Noticed expandVariables never
incremented its recursion guard and ended up rewriting it. To avoid
spurious warnings you can escape $'s, as in this comment:
/** The decoded name of the symbol, e.g. `==` instead of `\$eq\$eq`.
*/
The above will be ignored during expansion and translated to $eq$eq for
output. Closes #4412, no review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Wanting to deprecate -make, first I had to write a way to deprecate
-make. So there's that, now you can do
val s = SomeSetting(...) withDeprecationMessage "don't use this"
And it will do the usual deprecation things. And, deprecated -make. And
couldn't resist fixing a bug in -make, it would crash if you gave it its
own default option (i.e. -make:all.) Let's deprecate more!
I also did further cleaning up of our help outputs. Do I smell a
pulitzer in the making? No review.
|
|
|
|
|
| |
Excluding presentation test from sbaz-packaged test suite.
|
|
|
|
|
|
| |
include files/specialized/*.txt in the set of files packaged in the test
sbp.
|
|
|
|
|
|
| |
Transcript pastes now work with leading whitespace (such as my commit
message in r24624), no review.
|
|
|
|
|
|
| |
Fixing the scalap test rig to work like other test types so it doesn't
break the distribution build. No review.
|
|
|
|
|
|
|
|
| |
the wrong places, tags saying the wrong thing. I sorted types and values
so deprecated ones are at the end. I think they should be hidden by
default, but this is a big improvement. Leaving #3914 open so they can
be made invisible. No review.
|
|
|
|
|
|
| |
[scaladoc] Don't create broken link to private companion object. Closes
#4374. Review by pedrofurla.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Polishing the programmatic interface to the repl and other bits
of machinery which we'll have to live with for a while. The repl
classloader now works more like you'd expect a classloader to, as seen
here:
% scala -Dscala.repl.power
scala> class Bippus extends Traversable[Int] { def foreach[U](f: Int => U) = () }
defined class Bippus
scala> intp.classLoader.getResourceAsStream("Bippus").bytes()
res0: Array[Byte] = Array(-54, -2, -70, ...
scala> res0.size
res1: Int = 23954
scala> case class Bippy(x: Int)
defined class Bippy
// classBytes is shorter way to say the same thing scala>
intp.classLoader.classBytes("Bippy").size
res2: Int = 2356
scala> intp.classLoader.classBytes("Bippy$").size
res3: Int = 1741
Closes #4399, no review.
|
|
|
|
|
|
|
| |
Made ScalaRunTime.stringOf more general by separating out the part which
formats it for a repl result, and cleaned up some string functions
around the repl. No review.
|
|
|
|
|
|
|
|
| |
Eliminated option -Xwarninit, which is warning about a change which took
place 2.5 years ago and is only likely to add to option confusion at
this point. No review. (But let me know if we still need this for some
reason: many more of our options should be culled.)
|
|
|
|
|
|
|
| |
Removed some bitrotted ant directives which were keeping us from getting
Manifest docs. Thanks to dmharrah. Also removed RichUnit, fare thee
well. Closes #4404, no review.
|
|
|
|
|
|
|
| |
Added some implicitNotFound annotations to commonly used classes, and
some documentation to Manifest. (Said documentation is invisible for the
moment due to #4404.) No review.
|
|
|
|
|
|
| |
[scaladoc] Remove whitespace around colon and parentheses. Closes #4372.
Review by pedrofurla.
|
| |
|
|
|
|
|
|
| |
(1) The msil library can now parse modreq and modopt, (2) bug fix in
GenMSIL. review by rytz. Updated msil.jar coming soon.
|
|
|
|
|
|
|
|
| |
Learned an interesting lesson about having an implicit object with a
"def apply(x: Any): List[String]" method imported into power mode. Let's
just say this is not advised. This and other rough edges eliminated from
power mode, no review.
|
|
|
|
|
|
| |
A paltry 570 cases failing in -Xcheckinit due to some uninitialized val
or other. It is but a trifle! No review.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Trying to get fsc doing the right thing with respect to absolute and
relative paths. My knowledge of the problem had heretofore been second
hand, and my understanding of it incomplete. The real problem I have
determined is that there are a bunch of different things which go wrong
if relative paths start being resolved from a different base, each of
which needs custom handling.
classpath-style options, e.g. fsc -cp ../foo.jar
path-style options, e.g. fsc -d ../mydir
file arguments, e.g. fsc ../foo.scala
So it was more work than I had realized, or I probably wouldn't have
even touched it. But now it seems to be working as one would want. I
also poured some readability onto the fsc help output.
Closes #4395, no review, but community input would be great.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix for linked lists closes #4080 and proves my desire not to ship
obviously broken code is even greater than my will to hold out for any
help. I threw in a free fix for this which I noticed while in there.
scala> scala.collection.mutable.LinkedList[Int]().head
res0: Int = 0
Also was reminded how useless tests can be:
val ten = DoubleLinkedList(1 to 10: _*)
ten.insert(DoubleLinkedList(11))
// Post-insert position test
require(ten.last == 11)
Fortunately a test confirming buggy behavior still serves a purpose by
breaking when you fix the bug which allowed it to pass, thus letting you
fix the broken test too. Life's (very) little compensations. Linked list
code should still be presumed broken. No review.
|
| |
|
|
|
|
|
|
| |
needed when compiling the (jdk2ivkm'd) Scala library with Scala.NET, and
doesn't hurt in forJVM mode.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Revert "All compiler control methods now do something sensible when
called from presentation compiler thread itself."
This reverts commit bb290ba26e1d4677a36b7278ab370e841cca2605.
It leads to consistent lockups on the IDE, so we revert this
temporarily. We'll re-enable it after we can make sure it works
correctly.
|
|
|
|
|
| |
Also removed some obsolete targets related to the old 1.4 version.
|
|
|
|
|
|
|
| |
Suppressed unnecessary checkcasts on Throwables which were letting
AnyRefs reach the backend. Looks like this was my doing when trying to
wake back up the icode checker. Closes #4393, no review.
|
| |
|
|
|
|
|
| |
Some accumulated work and polish on classpath and jars. No review.
|
|
|
|
|
|
| |
Added some explanatory output when jline reader fails to create. No
review.
|
|
|
|
|
|
| |
Fix for cygwin: jline.UnixTerminal is now
scala.tools.jline.UnixTerminal. No review.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Fixed the "poor resilience to error conditions" crasher mentioned in
#3625, no review.
|
|
|
|
|
|
| |
Figured out why -Ydebug has been such a deathtrap for a while. I'm the
culprit. Making -Ydebug less likely to kill a compile, no review.
|
|
|
|
|
|
| |
Set the original tree for TypeTree in self-types, and making sure there
are no cycles in TypeTree.original. review by odersky.
|
|
|
|
|
| |
forJVM behavior remains unchanged.
|
| |
|
|
|
|
|
|
| |
All compiler control methods now do something sensible when called from
presentation compiler thread itself.
|
|
|
|
|
|
| |
[scaladoc] Changes to template.css based on suggestions by Bill Venners.
No review.
|
|
|
|
|
|
| |
Added synchronous version of askStructure, needed by the IDE to avoid
deadlocks. review by odersky.
|
| |
|
|
|
|
|
|
|
| |
Three things to make Eclipse more robust against deadlocks: (1) catch
stale responses in presentation compile thread. (2) Avoid stale
responses by two try-finallys in getEnteredParsed, askLoadedTyped.
|