Fix check bounds #1012
Fix check bounds #1012
Conversation
This is important for IDEs who want to see the full tree. The tree now gets replaced by a TypeTree in PostTyper.
Previously, bounds of a TypeDef tree were not checked. We now make sure bounds are checked everywhere in PostTyper. The previous partial check in Applications gets removed (it was not complete even for TypeApplications because sometimes bounds were not yet known when the test was performed.)
Checking bounds everywhere revealed a problem in compileStdLib, which this commit fixes.
New test that exhibited the problem is ski.scala. Previously this did not fail with a bounds violation.
|
I haven't looked into it much yet, but I think the deep subtype error is worth investigating, it happens just by running: ./bin/dotc -Yno-deep-subtypes ./scala-scala/src/library/scala/collection/generic/ParMapFactory.scalahowever, if I enable the ./bin/dotc -Ylog:all -Yno-deep-subtypes ./scala-scala/src/library/scala/collection/generic/ParMapFactory.scalaThen the error disappears! So clearly it's possible to typecheck this without deep subtyping |
|
An explanation could be that LazyRefs forcing creates additional subtypes. On Wed, Dec 30, 2015 at 5:51 PM, Guillaume Martres <notifications@github.com
Martin Odersky |
|
I think I got it, see #1014. |
|
Superseded by #1014 |
I noticed that we do not check all bounds of type parameters. Fixing this revealed another
problem when compiling stdlib which this commit also fixes. However, bounds checking
in PostTyper now causes deep subtypes, so stdlib has to be compiled without -Yno-dee-subtypes.
Review by @smarter.