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 I work around this now by introducing a bug in the subtype implementation at run-time, but it is not type correct behavior and will lead to subtle bugs elsewhere in the future.
After pattern matching a list out of something, like ... <{A ","}* l> ... the l will be bound to appl(regular(\iter-seps(sort("A"),[lit(",")])), ... instead of the correct appl(regular(\iter-seps(sort("A"),[layouts(...),lit(","),layouts(...)])), .... This is due to the auto-wrapping code which the compiler has to generate into the pattern matcher which should know the right regular prod to apply. It seems the layout separators are either not introduced yet there or they have been removed again (it depends on the inner workings of the type checker). Please help.
The text was updated successfully, but these errors were encountered:
I think I work around this now by introducing a bug in the subtype implementation at run-time, but it is not type correct behavior and will lead to subtle bugs elsewhere in the future.
After pattern matching a list out of something, like
... <{A ","}* l> ...
thel
will be bound toappl(regular(\iter-seps(sort("A"),[lit(",")])), ...
instead of the correctappl(regular(\iter-seps(sort("A"),[layouts(...),lit(","),layouts(...)])), ...
. This is due to the auto-wrapping code which the compiler has to generate into the pattern matcher which should know the rightregular
prod to apply. It seems the layout separators are either not introduced yet there or they have been removed again (it depends on the inner workings of the type checker). Please help.The text was updated successfully, but these errors were encountered: