Summoning covariant class leads to Nothing
in type argument
#17708
soronpo
started this conversation in
Feature Requests
Replies: 3 comments
-
I think this should be a feature request then. |
Beta Was this translation helpful? Give feedback.
0 replies
-
OK, I wasn't sure this isn't a bug. Tried to transfer it to lampepfl/dotty-feature-requests but I have no privileges to so. |
Beta Was this translation helpful? Give feedback.
0 replies
-
OK, I transferred it now. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Removing covariance of
T
inFuzzy
results in successful compilation.Not sure that this is a bug, since this could be relevant to Scala's eager widening of non-invariant parameters, since Scala 2 has a similar issue.
Compiler version
v3.2.0-RC2
Minimized code
https://scastie.scala-lang.org/DB2gSGGzT5SLtkbSkskmNA
The following code fails similarly in Scala 2 (and succeeds when covariance is removed):
https://scastie.scala-lang.org/hRFQ9LLCQf2Jbxi9jRy76g
Output
Expectation
No error, if this is indeed a bug. If it's expected result from the widening mechanism, then we need a future language feature to specify the type argument must be precise/exact/narrow.
Beta Was this translation helpful? Give feedback.
All reactions