Write a separate section on how return values are computed automatically

[kevaundray]

No description provided.

[Savio-Sou]

• Aim
  • Start a new section --> increase awareness
  • Also that they are syntactic sugar of constrained inputs for readability / Rust-likeness
• Reference:
  • Existing writeups:
    • https://noir-lang.github.io/book/getting_started/hello_world.html#prove-our-noir-program
    • https://noir-lang.github.io/book/getting_started/breakdown.html#anatomy-of-a-nargo-project
    • Old writeup:

There are two ways to "supply" public inputs.

fn main(x : Field, y : pub Field) {
  constrain x == y
}

In the above program, you are directly supplying the public input because it cannot be computed for some reason from the initial input.

fn main(x: Field, y : Field) -> pub Field {
 constrain x == y;
 return y
}

In the above program, you are return a value from the main function which we have defined to always be public, in this case, the output would get automatically computed for you from the initial inputs.
We floated the idea around of introducing the concept of a public input and a public output. This is a bit confusing depending on your background however because on the proving system layer, there are only public inputs

[critesjosh]

@Globallager what exactly needs to be documented for this to be considered complete?

[Savio-Sou]

@Globallager what exactly needs to be documented for this to be considered complete?

Mainly how public inputs & return values are automatically computed and stored in Verifier.toml when running nargo prove.

It was already documented here, but we still got a few queries from the community on it back in Jan. Hence we've created the issue for ways to further increase awareness (e.g. separating that into a new section).

If we weren't getting queries recently however we can close this without doc changes.

[signorecello]

to my knowledge this didn't prompt any more questions, so I'm closing for the time being