diff options
author | Jason Zaugg <jzaugg@gmail.com> | 2013-09-04 15:02:00 +0200 |
---|---|---|
committer | Jason Zaugg <jzaugg@gmail.com> | 2013-09-04 15:59:36 +0200 |
commit | fdd860df184f4c8bc6997ce2c1045b23c9fb61aa (patch) | |
tree | ab85d470988990e0e0da76e6e21d611e5dbece3b /src | |
parent | a8c05274f738943ae58ecefda4b012b9daf5d8dc (diff) | |
download | scala-fdd860df184f4c8bc6997ce2c1045b23c9fb61aa.tar.gz scala-fdd860df184f4c8bc6997ce2c1045b23c9fb61aa.tar.bz2 scala-fdd860df184f4c8bc6997ce2c1045b23c9fb61aa.zip |
SI-7801 Fix a nightmarish bug in Symbols#adaptInfos
The compiler-in-residence has always been a sketchy affair;
FSC and REPL offers a bounty of bugs that exploit the menagerie
of time-travel mechanisms in play for symbols' metadata (type, flags,
name and owner.) but are often cleverly masked by optimizations in
the compiler based on reference equality.
The latest: an innocuous change in Erasure:
https://github.com/scala/scala/commit/d8b96bb8#commitcomment-3995163
means that some `ErasureMap`-s over `MethodType`-s are now true
identities (as `UnitTpe` is always the same object, whereas
`erasedTypeRef(UnitClass)` returns an different `TypeRef` each
time.)
This, in turn, enables `TypeMap#mapOver` to reuse
the existing enclosing type, and so on. On such subtleties hinge
further optimizations, such as whether or not a given phase's
`InfoTransformer` needs to add an entry in a symbols type history.
When the REPL (or FSC / Presentation Compiler) creates a new
`Run`, `Symbol#rawInfo` tries to adapt the entries in the type
history for the new run. For packages, this was taken to be a
no-op; each entry is marked as being valid in the new run and
no further action is taken. This logic lurks in `adaptInfos`.
But, when the namer enters a new symbol in a package, it
*mutates* the Scope of that package classes info `enteringTyper`.
So the later entries in the type history *must* be invalidated
and recomputed.
We have two choices for a fix:
1) modify `Namers#enterInScope` to blow away the subsequent
type history for the owning symbol after inserting the
new member. Something like `owner.setInfo(owner.info)` would
have the desired effect.
2) Change `adaptInfos` to be more conservative when it comes
to package classes, and retain only the oldest entry in the
type history.
This commit goes for option 2.
Diffstat (limited to 'src')
-rw-r--r-- | src/reflect/scala/reflect/internal/Symbols.scala | 14 |
1 files changed, 10 insertions, 4 deletions
diff --git a/src/reflect/scala/reflect/internal/Symbols.scala b/src/reflect/scala/reflect/internal/Symbols.scala index d3a0ffb744..d58ea09386 100644 --- a/src/reflect/scala/reflect/internal/Symbols.scala +++ b/src/reflect/scala/reflect/internal/Symbols.scala @@ -1435,6 +1435,13 @@ trait Symbols extends api.Symbols { self: SymbolTable => assert(isCompilerUniverse) if (infos == null || runId(infos.validFrom) == currentRunId) { infos + } else if (isPackageClass) { + // SI-7801 early phase package scopes are mutated in new runs (Namers#enterPackage), so we have to + // discard transformed infos, rather than just marking them as from this run. + val oldest = infos.oldest + oldest.validFrom = validTo + this.infos = oldest + oldest } else { val prev1 = adaptInfos(infos.prev) if (prev1 ne infos.prev) prev1 @@ -1444,10 +1451,7 @@ trait Symbols extends api.Symbols { self: SymbolTable => _validTo = period(currentRunId, pid) phase = phaseWithId(pid) - val info1 = ( - if (isPackageClass) infos.info - else adaptToNewRunMap(infos.info) - ) + val info1 = adaptToNewRunMap(infos.info) if (info1 eq infos.info) { infos.validFrom = validTo infos @@ -3473,6 +3477,8 @@ trait Symbols extends api.Symbols { self: SymbolTable => "TypeHistory(" + phaseOf(validFrom)+":"+runId(validFrom) + "," + info + "," + prev + ")" def toList: List[TypeHistory] = this :: ( if (prev eq null) Nil else prev.toList ) + + def oldest: TypeHistory = if (prev == null) this else prev.oldest } // ----- Hoisted closures and convenience methods, for compile time reductions ------- |