xermicus

However, seems like for now I don't even need such breaking changes to the metadata. We have the language in the metadata for a reason. For my use case it'd...

Ah that is a good find. Well I think in #1207 and following discussions and PRs were solving a (very slightly) different problem; namely allowing to include something like language...

@SkymanOne to me the question as to why the contract execution output is wrapped inside a `Result`, while this seems unnecessary (because the revert flag already should tell you if...

> it makes sense to have Ok(Err) Sorry, I can't hold this off On a more serious note. I'm not questioning the concept of a `lang_error` per se. I'm just...

From another discussion; it could also be useful to provide some kind of "reference" data structure (akin to `Box`), that just holds a reference (the storage key) to a `Vec`...

You are right, it's very similar. I think what we had in mind was to give the contract authors some way of _trying_ to encode (or decode) the type, so...

Unfortunately `cargo publish` will fail, as we depend on `forge-fmt` which depends on solang-parser `=0.3.2`. While pinned dependencies in libraries is definitely annoying; I think depending on our own crate...

Right, given the licenses, both Apache 2.0, are compatible, I think this should work. The other solution coming to my mind would be to factor the language server out into...

> That doesn't solve the problem, as the language server would have the same issue. I see, so the solution would rather be factoring `solang-parser` out of the workspace I...

thank you so much for the fun contest and the super awesome prices!