xermicus
xermicus
Ah yes this is a good idea
Actually, I think we can keep the note but it should point to the `import "substrate";` statement?
@LucasSte found another broken example ```solidity contract Test { struct S { int foo; S[] s; } function test(int foo) public returns (S) { S ss = S(foo, new S[](2));...
Hi @vinodsharmak The documentation you linked literally answers your question: ``` The way to implement this on Solana is to have an authority account for the contract that must be...
Fine by me
> Not a bug but a feature XD As discussed elsewhere. I disagree. Because it is hard to impossible to know inside our _macros_ whether any state mutating API is...
> As discussed before, I think we should restrict the direct usage of ink::env crate, This doesn't solve the issue fully, but would be helpful in a lint. Bottom line...
Alternatively: There isn't even a need to change the metadata at all to make this work. If we just specify that front-ends and calling contracts always must decode the `lang_error`...
I see why contracts want to return errors. But I'm still not convinced that a contract output can be the same in the revert case _and_ in the non-revert case....
> but it increases the size of the contracts and creates new weird types per message. Right now, `Result`, a generic type, is a common appearance in our metadata, and...