Conversation
It seems not related to my fix. |
to confirm this, does it still fail if you revert your change? - there can always be knock-on effects |
I see, I tried to run locally, but I couldn't because of
|
|
Me too, but I'll take a look. It's also confusing when tests emit junk text. But this looks like an actual fail. Also I can't stand the progress bars at this point, but is there any point in complaining. |
|
Somehow, test ran locally, |
|
#24063 |
|
The summary is correct; the "progress" noise is just noise. But it has an option to suppress it, which I will propose. I can't read the linked PR output with the progress bar "supernoise", so I can't comment. |
fixes: scala#22513 Currently, if the upper bounds of opaque types are sealed, they pass exhaustivityCheckable, but exhaustivity checks are not handled correctly because the compiler preventing from decomposing them. The guard was introduced in [scala#19368](scala#19368), but the tests still pass even without this guard. [Cherry-picked f4e6239]
fixes: scala#22513 Currently, if the upper bounds of opaque types are sealed, they pass exhaustivityCheckable, but exhaustivity checks are not handled correctly because the compiler preventing from decomposing them. The guard was introduced in [scala#19368](scala#19368), but the tests still pass even without this guard.

fixes: #22513
Currently, if the upper bounds of opaque types are sealed, they pass exhaustivityCheckable, but exhaustivity checks are not handled correctly because the compiler preventing from decomposing them.
The guard was introduced in #19368, but the tests still pass even without this guard.