Contract refactoring, API, README, etc

[volovyks]

• fixed contract tests
• renamed some errors and error messages (opinionated, open for a discussion about some of them)
• moved user-facing functions into one impl block
• added README with an API and a list of envs (mostly existing descriptions moved to a text file and formatted for better onboarding).

I've also tried to narrow down return error types in user-facing functions so they know what they can get exactly.

/// from
pub fn sign(&mut self, request: SignRequest) -> Result<near_sdk::Promise, MpcContractError>
/// to
pub fn sign(&mut self, request: SignRequest) -> Result<near_sdk::Promise, SignError>

But to do that we will need to get rid of the MpcContractError enum since it's implementation of

impl near_sdk::FunctionError for MpcContractError {
    fn panic(&self) -> ! {
        crate::env::panic_str(&self.to_string())
    }
}

conflicts with newly added impl for SignError, PublicKeyError, etc.
Let me know if you guys think this is an important change, or we can stick to MpcError everyvere.

@ppca based on your branch, since it has a lot of changes

[gagdiez]

@volovyks should this be marked as [#private]?

otherwise, people could just call it jumping over sign (except if there is something about using the DATA_ID_REGISTER that I am not aware of)

[volovyks]

Indeed. Changed. cc @ppca

[gagdiez]

@volovyks could we return the money if they exceed the payment? This would cost < 3Tgas, but help with the deposit estimate

if required > deposit {
  Promise::new(env::predecessor_account_id()).transfer(deposit - required_deposit);
}

[volovyks]

Good point, added to our roadmap: #708

[gagdiez]

@volovyks the current cost is 1 yoctoNEAR when there are less than 4 requests, and then it ramps to 50milliNEAR * (1...4), which ranges from 0.05N to 0.20N... this might be a bit expensive.

Did we intended to use yoctoNear instead of milliNEAR?

match pending_requests {
    0..=CHEAP_REQUESTS => 1,
    _ => {
        (pending_requests - CHEAP_REQUESTS) as u128
            * NearToken::from_millinear(50).as_yoctonear()
    }
}

[volovyks]

Audit is fixed in

@volovyks the current cost is 1 yoctoNEAR when there are less than 4 requests, and then it ramps to 50milliNEAR * (1...4), which ranges from 0.05N to 0.20N... this might be a bit expensive.

Did we intended to use yoctoNear instead of milliNEAR?

match pending_requests {
    0..=CHEAP_REQUESTS => 1,
    _ => {
        (pending_requests - CHEAP_REQUESTS) as u128
            * NearToken::from_millinear(50).as_yoctonear()
    }
}

We use yocto everywhere.
It may be expensive, but the intention is to prevent system overload. Attacks must be expensive in order to prevent them.

[github-actions]

Terraform Feature Environment Destroy (dev-706)

Terraform Initialization ⚙️success

Terraform Destroy success

Show Destroy Plan

No changes. No objects need to be destroyed.

Either you have not created any objects yet or the existing objects were
already deleted outside of Terraform.

Destroy complete! Resources: 0 destroyed.

Pusher: @ppca, Action: pull_request, Working Directory: ``, Workflow: Terraform Feature Env (Destroy)