Fix #1442: add new phase, SelectStatic #1445
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
GenBCode has an implicit assumption that I wasn't aware of: GetStatic should not be emitted against a valid selector. If it is, GenBCode messes up the stack by not pop-ing the selector. Surprisingly, this transformation is perfumed in nsc by flatten.
DarkDimius
changed the title
Blocks are not denoting trees(why aren't they?) For now, I'm fixing this using a quick fix. For future, it may make sense to discuss this on dotty meeting and make blocks be a Denoting tree and return denotation of expo. Another option is to move regularisation logic into tree transformers.
The new behaviour is more reasonable. Now the module if forced consistently in both examples. Note that this is deviation from behaviour of scalac.
| case Select(Block(stats, qual), nm) => | ||
| Block(stats, cpy.Select(t)(qual, nm)) | ||
| case Apply(Block(stats, qual), nm) => | ||
| Block(stats, Apply(qual, nm)) |
There was a problem hiding this comment.
Any particular reason not to copy Apply and TypeApplyas you do with Select?
There was a problem hiding this comment.
Copying does not copy the tree here, as it always has a different structure.
cpy.Select can preserve the symbol though, in case method is overloaded. This is not needed for Apply\TypeApply as by the time they are applied overloads are already resolved.
There was a problem hiding this comment.
Yes, I asked because I ran into the overloading issues some time ago. Great!
|
LGTM, thanks for the quick and elegant fix @DarkDimius! |
|
A general comment. Try to keep the rest loosely coupled with the backend so they can evolve independently and cages in one area does not necessitate changes elsewhere. |
sjrd
added a commit
to dotty-staging/dotty
that referenced
this pull request
Sep 14, 2020
… objects. This restores parity with the run-time semantics of Scala 2. When accessing an inner static class or object, we avoid loading the enclosing objects as values. The discrepancy in run-time semantics was introduced as a by-product of scala#1445 The change in `t4859.check` restores it to its previous state, before that PR.
sjrd
added a commit
to dotty-staging/dotty
that referenced
this pull request
Sep 14, 2020
… objects. This restores parity with the run-time semantics of Scala 2. When accessing an inner static class or object, we avoid loading the enclosing objects as values. The discrepancy in run-time semantics was introduced as a by-product of scala#1445 The change in `t4859.check` restores it to its previous state, before that PR.
sjrd
added a commit
to dotty-staging/dotty
that referenced
this pull request
Sep 15, 2020
… objects. This restores parity with the run-time semantics of Scala 2. When accessing an inner static class or object, we avoid loading the enclosing objects as values. The discrepancy in run-time semantics was introduced as a by-product of scala#1445 The change in `t4859.check` restores it to its previous state, before that PR.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
GenBCode has an implicit assumption that I wasn't aware of:
GetStatic should not be emitted against a non-trivial selector.
If it is, GenBCode messes up the stack by not pop-ing the selector.
Surprisingly, this transformation is perfumed in nsc by flatten.