diff options
author | Paul Phillips <paulp@improving.org> | 2011-06-24 05:42:08 +0000 |
---|---|---|
committer | Paul Phillips <paulp@improving.org> | 2011-06-24 05:42:08 +0000 |
commit | 034489b50158756b57091a04830d94160975bddb (patch) | |
tree | 4b695b2c00f7d464b893842ad8e729404b62c69b /src/compiler/scala/reflect/internal/Types.scala | |
parent | 5dc127e69cae1d88aa6910ea6378ad5dc1aaeaab (diff) | |
download | scala-034489b50158756b57091a04830d94160975bddb.tar.gz scala-034489b50158756b57091a04830d94160975bddb.tar.bz2 scala-034489b50158756b57091a04830d94160975bddb.zip |
Added sanity check to lub calculation to preven...
Added sanity check to lub calculation to prevent invalid lubs from
emerging. The underlying cause of said lubs is that higher-order
type parameters are not handled correctly: this is why the issue
is seen so frequently in the collections. See pending test
pending/pos/those-kinds-are-high.scala for a demonstration. Until that's
fixed, we can at least raise the bar a bit.
Closes #2094, #2322, #4501. Also, some test cases in neg have been
promoted into working programs: #2179, #3774. (They're not in neg for
the "shouldn't work" reason, but out of despair.)
In some cases, such as the original reported ticket in #3528, this
only pushes the problem downfield: it still fails due to inferred type
parameters not conforming to bounds. I believe a similar issue with
higher-order type parameters underlies that.
Look at how far this takes us though. All kinds of stuff which did not
work, now works. None of these even compiled until now:
scala> :type List(mutable.Map(1 -> 1), immutable.Map(1 -> 1))
List[scala.collection.Map[Int,Int]]
scala> :type Set(List(1), mutable.Map(1 -> 1))
scala.collection.Set[Iterable[Any] with PartialFunction[Int,Int]]
scala> :type Stream(List(1), Set(1), 1 to 5)
Stream[Iterable[Int] with Int => AnyVal{def getClass(): Class[_ >: Int with Boolean <: AnyVal]}]
scala> :type Map(1 -> (1 to 10), 2 -> (1 to 10).toList)
scala.collection.immutable.Map[Int,scala.collection.immutable.Seq[Int]
]
PERFORMANCE: compiling quick.lib and quick.comp, this patch results in
an extra 27 subtype tests. Total. Time difference too small to measure.
However to be on the safe side I made it really easy to disable.
private final val verifyLubs = true // set to false
Review by moors, odersky.
Diffstat (limited to 'src/compiler/scala/reflect/internal/Types.scala')
-rw-r--r-- | src/compiler/scala/reflect/internal/Types.scala | 22 |
1 files changed, 20 insertions, 2 deletions
diff --git a/src/compiler/scala/reflect/internal/Types.scala b/src/compiler/scala/reflect/internal/Types.scala index e8b9dc0865..b5777c099f 100644 --- a/src/compiler/scala/reflect/internal/Types.scala +++ b/src/compiler/scala/reflect/internal/Types.scala @@ -90,6 +90,8 @@ trait Types /*extends reflect.generic.Types*/ { self: SymbolTable => private final def decr(depth: Int) = if (depth == AnyDepth) AnyDepth else depth - 1 private final val printLubs = false + /** In case anyone wants to turn off lub verification without reverting anything. */ + private final val verifyLubs = true /** The current skolemization level, needed for the algorithms * in isSameType, isSubType that do constraint solving under a prefix. @@ -5310,9 +5312,25 @@ A type's typeSymbol should never be inspected directly. } } if (lubRefined.decls.isEmpty) lubBase + else if (!verifyLubs) lubRefined else { -// println("refined lub of "+ts+"/"+narrowts+" is "+lubRefined+", baseclasses = "+(ts map (_.baseTypeSeq) map (_.toList))) - lubRefined + // Verify that every given type conforms to the calculated lub. + // In theory this should not be necessary, but higher-order type + // parameters are not handled correctly. + val ok = ts forall { t => + (t <:< lubRefined) || { + if (settings.debug.value) { + Console.println( + "Malformed lub: " + lubRefined + "\n" + + "Argument " + t + " does not conform. Falling back to " + lubBase + ) + } + false + } + } + // If not, fall back on the more conservative calculation. + if (ok) lubRefined + else lubBase } } existentialAbstraction(tparams, lubType) |