You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think the reason for this issue is that we currently use two TailSwitch type logic application for computing the output of the ~ operator, whereby no distinction is made between the application for a contravariant type parameter (the first) and the application for a covariant type parameter (the second).
The text was updated successfully, but these errors were encountered:
There is a bug in (the application of) the
TailSwitch
logic, which causes certain parsers to compile when they shouldn't:The problem appears only when the 2nd rule consumes all of the 1st rule's output.
This parser for example correctly doesn't compile:
I think the reason for this issue is that we currently use two
TailSwitch
type logic application for computing the output of the~
operator, whereby no distinction is made between the application for a contravariant type parameter (the first) and the application for a covariant type parameter (the second).The text was updated successfully, but these errors were encountered: