feat(zink-codegen): introduce derive macro for events

[malik672]

Resolves #255

• generate ABIs for events and gathered events as enums
• introduce util function for converting type to string

0x30bE4D758d86cfb1Ae74Ae698f2CF4BA7dC8d693

[malik672]

@clearloop seems like log does not exist

[clearloop]

Sorry for the late! was busy these days, almost finished! your new tests are totally correct!

at the current stage, you now need to handle the log function in the visitor, see https://github.com/zink-lang/zink/blob/main/codegen/src/visitor/log.rs#L62

If you need more details or have specific questions, free feel to pin me in the telegram group! In case this issue is a bit harder than the previous, just updated the bounty to large level!

[malik672]

Sorry for the late! was busy these days, almost finished! your new tests are totally correct!

at the current stage, you now need to handle the log function in the visitor, see https://github.com/zink-lang/zink/blob/main/codegen/src/visitor/log.rs#L62

If you need more details or have specific questions, free feel to pin me in the telegram group! In case this issue is a bit harder than the previous, just updated the bounty to large level!

[malik672]

    pub fn log(&mut self, count: usize) -> Result<()> {
        let topics = Vec::<Vec<u8>>::default();
 
        let (offset, size) = self.data()?;
        let size = size as usize;
        let data = self.env.data.load(offset, size)?;

        tracing::debug!("log1 name: {:?}", data);

        for topic in topics {
            self.masm.push(&topic)?;
        }

    
        // 1. write data to memory
        let MemoryInfo { offset, size } = self.masm.memory_write_bytes(&data)?;

        // 3. prepare the offset and size of the data.
        self.masm.push(&size.to_ls_bytes())?;
        self.masm.push(&offset)?;

        // 4. run log for the data
        match count {
            0 => self.masm._log0(),
            1 => self.masm._log1(),
            2 => self.masm._log2(),
            3 => self.masm._log3(),
            4 => self.masm._log4(),
            _ => unreachable!("invalid topics"),
        }?;

        Ok(())
    }``` @clearloop 

[malik672]

    pub fn log(&mut self, count: usize) -> Result<()> {
        let topics = Vec::<Vec<u8>>::default();
 
        let (offset, size) = self.data()?;
        let size = size as usize;
        let data = self.env.data.load(offset, size)?;

        tracing::debug!("log1 name: {:?}", data);

        for topic in topics {
            self.masm.push(&topic)?;
        }

    
        // 1. write data to memory
        let MemoryInfo { offset, size } = self.masm.memory_write_bytes(&data)?;

        // 3. prepare the offset and size of the data.
        self.masm.push(&size.to_ls_bytes())?;
        self.masm.push(&offset)?;

        // 4. run log for the data
        match count {
            0 => self.masm._log0(),
            1 => self.masm._log1(),
            2 => self.masm._log2(),
            3 => self.masm._log3(),
            4 => self.masm._log4(),
            _ => unreachable!("invalid topics"),
        }?;

        Ok(())
    }``` @clearloop 

[clearloop]

    pub fn log(&mut self, count: usize) -> Result<()> {
        let data = self.env.data.load(offset, size)?;

        // 1. write data to memory
        let MemoryInfo { offset, size } = self.masm.memory_write_bytes(&data)?;

        // 3. prepare the offset and size of the data.
        self.masm.push(&size.to_ls_bytes())?;
        self.masm.push(&offset)?;

        // 4. run log for the data
        match count {
            0 => self.masm._log0(),
            1 => self.masm._log1(),
            2 => self.masm._log2(),
            3 => self.masm._log3(),
            4 => self.masm._log4(),
            _ => unreachable!("invalid topics"),
        }?;

        Ok(())

like this

[malik672]

let (offset, size) = self.data()?;

quite similar to what i did first, made the tests fail

[malik672]

    pub fn log(&mut self, count: usize) -> Result<()> {
        let data = self.env.data.load(offset, size)?;

        // 1. write data to memory
        let MemoryInfo { offset, size } = self.masm.memory_write_bytes(&data)?;

        // 3. prepare the offset and size of the data.
        self.masm.push(&size.to_ls_bytes())?;
        self.masm.push(&offset)?;

        // 4. run log for the data
        match count {
            0 => self.masm._log0(),
            1 => self.masm._log1(),
            2 => self.masm._log2(),
            3 => self.masm._log3(),
            4 => self.masm._log4(),
            _ => unreachable!("invalid topics"),
        }?;

        Ok(())

like this

it sorts normally:
--- STDERR: zink::log log1 ---
thread 'log1' panicked at tests/log.rs:25:5:
assertion left == right failed
left: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 112, 111, 110, 103]
right: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 105, 110, 103]

[clearloop]

assertion left == right failed left: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 112, 111, 110, 103] right: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 105, 110, 103]

looks closed! now you can decode this output to see what exactly the log is!

let left = String::from_utf8_lossy(&[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 112, 111, 110, 103]);
let right = String::from_utf8_lossy(&[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 80, 105, 110, 103])

[clearloop]

see my fix here: malik672/zink@sync...zink-lang:zink:cl/sync

[clearloop]

Everything works now!

Now we need to wrap these log functions into our proc-macro

for the event abi part, let's handle it in a separated PR (#299), since it modifies logic in compiler

[clearloop]

need to generate sugar functions in our proc-macro, the implementation till 583dbe2 makes sure our deep logic are working now, so the last thing is wrap them in proc-macro

[malik672]

@clearloop might need your help here

[clearloop]

Instead of quote!, construct the functions with the ImplItemType could make things more dynamic

[clearloop]

Thanks for contributing on this issue! however, I can not send you the full bounty because I solved the issue instead finally XD, it's unfair to other contributors who solve tricky issues independently, hope you can understand!

Apart from that, thank you again for spending time on digging the feature! see the transaction 🎁 : https://polygonscan.com/tx/0x1b588cab3bdb014fe247acf4c06f825515af6b8ee4c1df1c505de220f7af39b1

[CalabashSquash]

How do you specify indexed topics?

[g4titanx]

@CalabashSquash you can use the #[indexed] attribute on the fields you want to index within each variant. for example:

#[derive(Event)]
pub enum MyEvent {
    /// Event with one topic, indexed
    Topic1(#[indexed] U256),
    /// Event with two topics, first one indexed
    Topic2(#[indexed] U256, U256),
    /// Event with three topics, first two indexed
    Topic3(#[indexed] U256, #[indexed] U256, U256),
    /// Event with four topics, first three indexed
    Topic4(#[indexed] U256, #[indexed] U256, #[indexed] U256, U256),
}

the #[indexed] attribute marks a field as an indexed topic, which will be included in the event's topics for efficient filtering on the Ethereum blockchain. also, note that non-anonymous events can have up to 3 indexed fields, anonymous events can have up to 4 indexed fields and fields without #[indexed] are stored in the event's data payload, not as topics.
the logic lies here