Not implemented chain_id function.

[smiasojed]

The ERC20 workspace in Remix is failing to compile.

How to reproduce:

1. Got to https://smiasojed.github.io/remix-project
2. Select ERC20 workspace
3. Compile

results:

Contract `contracts/MyToken.sol:MyToken` compiling error: thread 'main' panicked at /home/runner/work/revive/revive/crates/llvm-context/src/polkavm/evm/context.rs:46:5:
not yet implemented
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

eip: https://eips.ethereum.org/EIPS/eip-1344

[xermicus]

Can you rebase against latest main where it will always return a chain id of 0 instead of panicking? I snuck it in recently. We still have to agree on those.

[smiasojed]

With the new version I got another panic:
Contract contracts/MyToken.sol:MyToken compiling error: thread 'main' panicked at /Users/smiasojed/Development/revive/crates/llvm-context/src/polkavm/context/pointer.rs:65:9:\nassertion left != right failed: Stack pointers cannot be addressed\n left: Stack\n right: Stack\n

[xermicus]

Can you provide me the contents of MyToken.sol? I don't see the ERC20 workspace you mentioned originally

[xermicus]

Also any libraries, assuming it uses OZ

[xermicus]

Nevermind. It is caused by the immutables, which are still a WIP.

[smiasojed]

Can you provide me the contents of MyToken.sol? I don't see the ERC20 workspace you mentioned originally

// SPDX-License-Identifier: MIT
pragma solidity ^0.8.20;

import "@openzeppelin/contracts/token/ERC20/ERC20.sol";
import "@openzeppelin/contracts/token/ERC20/extensions/ERC20Permit.sol";

contract MyToken is ERC20, ERC20Permit {
    constructor() ERC20("MyToken", "MTK") ERC20Permit("MyToken") {}
}

on the remix main page in the section: Explore. Prototype. Create. Click ERC20

[athei]

I guess chain_id needs pallet support to implement properly. We should check what other L2, Moonbeam etc. return there.

[xermicus]

I planned to add it to the pallet as configuration. Eventually we'll have to settle on the ChainID for every chain we deploy, including testnets. So having it in the configuration seems reasonable to me.

[athei]

One obvious idea would be to use the parachain id. But it is not unique across relay chains. So maybe something like keccak("revive_chain" ++ relay_chain_id ++ parachain_id). But no matter what we pick: Exposing it as Config` item is the way to go as it has to be configured by the runtime.

But I assume it doesn't really matter as long it is unique and people can lookup what is what.

[xermicus]

Yeah, we might want to add them here too.

[athei]

Yes I was wondering how if numbers are coordinated. So basically just pick a unique one and get a PR merged. No hashing required :)