Don Syme
Don Syme
The intention is reasonable - though this is pretty much in the "already decided in F# 1.x" category - it was something we considered. Also, if all types were sealed...
@charlesroddie I do agree with all of the above. Users of all levels below "expert" probably would make inaccurate assessments of whether a type is sealed. They would loosely learn...
Marked this approved-in-principle
@ovatsus Yes, we do call `ApplyStaticArguments` many times, basically each time the file is checked. The type provider should cache to give good performance in these cases However in the...
This is by design. Parentheses are required. There is a type declaration form `type R - struct ... end` and that takes precedence. I'll note this for clarification in the...
The suggestion here is misnamed, it's actually for typescript style literals-as-types
https://www.typescriptlang.org/docs/handbook/2/everyday-types.html#literal-types
There have been previous suggestions along these lines, usually too ungrounded in practice and I rejected them, but we should dig them out. The TS experience shows these are highly...
This is by design - a use of SRTP has caused an infinite regression in type inference (generally this is not possible in F#, but with non-recommended uses of SRTP...
There's a rough outline of where the code would have to go here: https://github.com/dotnet/fsharp/compare/master...dsyme:sd?expand=1 I started looking at the implementation but I notice that the implementation for set comparison looks...