I still find this behavior to be a bit surprising:
#include "share/atspre_staload.hats"
fun immval(): void = let
val xx: int = 1
val _ = xx = 2
val yy: int = 1
val yy: int = 2
in (
println!("xx is ", xx);
println!("yy is ", yy)
) end
implement
main0 () = immval()
The result of running th
There are no assignments here.
The following line binds '_' to the value 'true':
val _ = xx = 2
xx = 2 is a boolean expression; it is not assignment.
Also, the following line means creating a name yy for the
value 1:
val yy: int = 1
Again, no assignment. For assignments, you need to change
'v
Ah, right, I'm still warming up to having "=" used for both assignment and
equality depending on placement, sorry for the mixup.
In the second case, then, I guess what I'm suggesting is that the same name
shouldn't be allowed to bind more than once in the same scope.
On Sunday, March 11, 2018 a
>>In the second case, then, I guess what I'm suggesting is that the same
name shouldn't be allowed to bind more than once in the same scope.
This really a taste, I would argue.
When coding in Java, I use a lot of 'final' variables. I find it annoying
that
two 'final' variables can not have the sa
Maybe now isn't the time, but perhaps (in ATS2 or ATS3) there could be
scope-based or project-based overrides for certain linting. For instance,
the following library is a build plugin and a normal
library: http://www.wartremover.org/doc/install-setup.html
In the first sense, as a build plugin,
