Address review comments

  - clarify the intent of tests
  - Consolidate stripExistentialsAndTypeVars with similar logic in
    mergePrefixAndArgs
  - Refactor special cases in maybeRewrap

The name isn't great, but I'm struggling to come up with
a pithy way to describe the rogue band of types.

1 files changed, 7 insertions, 0 deletions

diff --git a/test/files/neg/lub-from-hell-2.scala b/test/files/neg/lub-from-hell-2.scala
index 96760c6edf..18c99dfada 100644
--- a/test/files/neg/lub-from-hell-2.scala
+++ b/test/files/neg/lub-from-hell-2.scala
@@ -4,3 +4,10 @@ class Test {
   def bar(a: Boolean, b: scala.collection.mutable.SetLike[Any,scala.collection.mutable.Set[Any]], c: scala.collection.mutable.Buffer[Any]) = if (a) b else c
   // bar produces an ill-bounded LUB in 2.11.8. After this commit, which fixes a bug in existential+refinement lubs, foo also fails.
 }
+// This test case minimizes a case that stated to fail compile after my fixes in SI-5294.
+// `foo` used to compile for the wrong reason, `mergePrefixAndArgs` failed to transpose a
+// ragged matrix and skipped to the next level of the base type sequences to find a common type symbol.
+//
+// My changes fixed the root cause of the ragged matrix, which uncovered the latent bug.
+// For comparison, `bar` failed to compile before _and_ after my changes for the same reason:
+// f-bounded types involved in LUBs can sometimes produce an ill-bounded LUB.