chore: use near-abi 0.3.0 #954
Conversation
| &MethodType::Regular => { | ||
| quote! { near_sdk::__private::AbiFunctionKind::Call } |
There was a problem hiding this comment.
I think it's worth noting for clarity that until #890 is merged, the ABI will represent self-less functions as call functions. Meaning clients will think they have to send signed transactions and not simply making view calls. Then again, I don't have any stats on how often these are used or if at all, so it might be a trivial nit.
There was a problem hiding this comment.
Yeah, but that is also the current behavior, so doesn't have to do anything with this particular change.
| assert_eq!(args[1].name, "b".to_string()); | ||
| } | ||
| AbiParameters::Borsh { .. } => { | ||
| assert!(false); |
There was a problem hiding this comment.
Well, it is not actually unreachable if the thing we are trying to test happens to have borsh serialization though
There was a problem hiding this comment.
Hm, but isn't that the point of the test? Asserting that it is JSON serialized and not borsh? (at least in this case)
There was a problem hiding this comment.
Yeah, but unreacheable! to me is something you know (statically) will never be reached, but can't convince compiler to be the case. This test, on the other hand, depends on some external input (a wasm file) and tests its properties.
This is arguing very subtle semantics and honestly it feels like AbiParameters should just have json()? and borsh()? methods instead of this match/fail pattern.
There was a problem hiding this comment.
That's fair, wasn't a strong opinion. I'm not sure if we're at the point where we need those helper methods, so we can proceed as-is. 👍🏽
There was a problem hiding this comment.
In that case, still seems like panic!() might be more readable?
Related: near/cargo-near#83