Silq REPL

[ode]

I think it'd be great to have a repl for the language, for toying with code before writing it down. Shouldn't be too hard to implement (not too sure). Thoughts?

[tgehr]

Reading is easy, but it's not entirely clear how it would evaluate and print for quantum state. E.g.

> x:=H(0:𝔹);
???
> y:=dup(x);
???
> x
???
> dump()
/────────
QUANTUM STATE
(1+0i)·|1⟩₀⊗|1⟩₁

VARIABLES
x ↦ ref(0)
y ↦ ref(1)
────────/

How should this work? Otherwise, implementation should be pretty straightforward.

[tgehr]

Also, measurements can lead to destructive state updates, maybe there should be a way to undo them.

[ode]

well, clearly measurements shouldn't be done unless explicitly asked. Printing out the ket notation is a good alternative imo

[tgehr]

In my example, x and y are entangled, so I am not sure how to show the partial state. What would you show instead of ??? in the REPL output?

[ode]

we could either disregard entanglement and only show the probabilities or do something like

> x = H(0:𝔹);
(1/√2) (|0⟩ + |1⟩)

> y = dup(x);
|xy⟩ = (1/√2) (|00⟩ + |11⟩)

in other words, detecting variables that are entangled and showing all of them when one of them is asked. Note that x evaluates to the same thing after typing in the second statement

[j6k4m8]

As a user I like the probability approach @LordPos mentions — and I think for beginners it'll feel similar to what people have seen in tools like Quantum Experience etc!