SI-9946 don't null field in lazy accessors that turn out to be live

If a non-transient lazy val is the only user of a private field in
a class, the field is nulled out at the end of the lazy initializer.
This is tested in the existing test `run/lazy-leaks.scala`.

The analysis of which fields could be nulled out was recently moved
from `mixin` to the new `fields` phase. This introduced a regression
as a reference from an inner- or companion-classes had not yet
been processed by `explicitouter` to publicise private fields.

This commit delays the analysis to mixin (after explicit outer has
done its work.)

Navigating from `foo$lzycompute()` to `foo()` to `foo` is a little
dirty now. I'm not sure whether there is a more robust way to
structure things.

1 files changed, 1 insertions, 1 deletions

diff --git a/src/reflect/scala/reflect/internal/TypeDebugging.scala b/src/reflect/scala/reflect/internal/TypeDebugging.scala
index 63f897cd32..e9050b4e33 100644
--- a/src/reflect/scala/reflect/internal/TypeDebugging.scala
+++ b/src/reflect/scala/reflect/internal/TypeDebugging.scala
@@ -110,7 +110,7 @@ trait TypeDebugging {
         val hi_s = if (noPrint(hi)) "" else " <: " + ptTree(hi)
         lo_s + hi_s
       case _ if (t.symbol eq null) || (t.symbol eq NoSymbol) => to_s(t)
-      case _                                                 => "" + t.symbol.tpe
+      case _                                                 => "" + t.symbol.rawInfo.safeToString
     }
     def ptTypeParam(td: TypeDef): String = {
       val TypeDef(_, name, tparams, rhs) = td