Wallet: Support transactions to multiple recipients

[JoshOrndorff]

Currently the wallet allows sending inputs from multiple owners in a single transaction, but only allows a single recipient to all specified outputs.

To support specifying a recipient on a per-output basis we will need to expand the cli somehow.

[NadigerAmit]

@JoshOrndorff

I am planning to add the multiple recipient support as below, Please let me know if it is ok not ?

Approach1:

/// Default recipient is  SHAWN_KEY and output amount is 0
pub const DEFAULT_RECIPIENT: &str = "d2bf4b844dfefd6772a8843e669f943408966a977e3ae2af1dd78e0f55f4df67 0";

#[derive(Debug, Args)]
pub struct SpendArgs {
    /// An input to be consumed by this transaction. This argument may be specified multiple times.
    /// They must all be coins.
    #[arg(long, short, verbatim_doc_comment, value_parser = output_ref_from_string)]
    pub input: Vec<OutputRef>,

    /// Variable number of recipients and their associated coins.
    /// For example, "--recipients 0x1234 1 2 --recipients 0x5678 3 4 6"
   #[arg(long, short, verbatim_doc_comment, value_parser = parse_recipient_coins, action = Append, 
    default_value = DEFAULT_RECIPIENT)]
   pub recipients: Vec<(H256, Vec<u128>)>,
}

// I need to refine this API , to return the SHAWN_PUB_KEY
fn parse_recipient_coins(s: &str) ->  Result<(H256, Vec<u128>), &'static str> {
    let parts: Vec<&str> = s.split_whitespace().collect();
    if parts.len() >= 2 {
        let recipient = h256_from_string(parts[0]);
        let coins = parts[1..].iter().filter_map(|&c| c.parse().ok()).collect();
        match recipient {
            Ok(r) => {return Ok((r, coins));},
            _ => {},
        };
    }
    Err("Invalid input format")
}

And I need to send the command line args as below
./target/release/tuxedo-template-wallet spend-coins-multi-recipient -i a95362504966abd5ee55223d09e860bb1dd60eba1425b80fb05e1cc3ad66bf7100000000 -r "0xfa96831be67e2abe0378467e1156b401d2ab044b0a66fc7703b8d90e84c45c54 1 2" -r "0xc2f52af267b5a3f46510e39e6dcdbd5904b722396727866f22f272ba687a9953 3 4 6"

Result: I could receive the command line args successfully .

But I think ideal way below:

Approach2:

#[derive(Debug, Args)]
pub struct SpendArgs1 {
    /// An input to be consumed by this transaction. This argument may be specified multiple times.
    /// They must all be coins.
    #[arg(long, short, verbatim_doc_comment, value_parser = output_ref_from_string)]
    pub input: Vec<OutputRef>,

    #[arg(long, short, verbatim_doc_comment, value_parser = h256_from_string, default_value = SHAWN_PUB_KEY)]
    pub recipients:Vec<Recipients>
}

#[derive(Debug, Args,Clone)]
pub struct Recipients {
    #[arg(long, short, verbatim_doc_comment, value_parser = h256_from_string, default_value = SHAWN_PUB_KEY)]
    pub recipient: H256,
    
    #[arg(long, short, verbatim_doc_comment, action = Append)]
    pub output_amount: Vec<u128>,
}

For this approach I am not finding a correct way to send the cmd line arguments.
If possible, Please share your opinion.

[NadigerAmit]

the transfer to multiple recipient is working good :

[JoshOrndorff]

Nice job getting something working. Please open a PR with your work-in-progress already so I can play around with it as well. I'm excited about this.

The way you have parsing working is reasonable and good enough. Another alternative would be that for each -r you provide a single address and single value. So if you want multiple coins to go to the same address you provide two different -r with that address. Not sure if it is better than yours, just different.

Your idea for pub struct Recipient is nice. But that type shouldn't be #[derive(Args)] or any other clap stuff. Then your parse function returns that type instead of the (H256, Vec<u128>).

Anyway, you're doing well, please open a PR.

[NadigerAmit]

@JoshOrndorff,
Thank you so much.
I will create a PR soon. In the meantime, if you can access the below please check the 'spend_coins' function:
https://github.com/mlabs-haskell/Tuxedo/blob/main/wallet/src/money.rs

Note: The above code is still experimental and may not meet the standard quality. I appreciate your understanding in this regard.

I am successful with Approach 1. In the future will do some more experiments with Approach 2.

[JoshOrndorff]

@NadigerAmit I hope you will be able to open a PR for this feature. LMK if I can help.

[NadigerAmit]

@JoshOrndorff, Thanks. I will open a PR this week after one more round of test and the relevant modification in the README file.