[LSP2] Adding 2 bytes as length of the compact bytes array

[YamenMerhi]

What this issue is about ?

Problem

• Currently there is no way to know the number of elements of a compacted bytes array

Solution

• Adding 2 bytes as length of the compact bytes array at the start

[CJ42]

This is a good suggestion as so far, the only way to do this is by looping through the bytes, which can be expensive if there are a lot of elements.
This might be a good to have.

What do you think @frozeman @skimaharvey @b00ste ?

[skimaharvey]

Definitely could be useful to have an easy way to know the number of elements without having to loop over all the elements. The only thing I am not sure about is that it will kind of force us to verify the validity of the length value in comparison to each length element (length value = the total number of length elements) so we will need to loop over the entire compact byte array when storing.
In the LSP6 context, I am not sure we want to do that but at the same time, I don't think we want to store a length value that does not make sense.
I am just wondering what real value we will get from it since it is not trustable

[JeneaVranceanu]

I'm currently in the process of opening a lengthy issue on CBAs and decided to look if there's anything open related to CBAs. Found this issue and I'd like to join the "brainstorming" 🙂

---

The only thing I am not sure about is that it will kind of force us to verify the validity of the length value

@skimaharvey Why will it force you to validate the length? Are we forced to validate the length on Solidity's arrays?

I'll put the question differently: if we are calling setData with key whose valueType is some_type[] (the emphasis is on the [] part) we do not validate it on the smart contract side? I do not expect it to be validated on the smart contract side.

If that's true we can receive some ABI from getData for type some_type[] which will not be valid as well and the decoding attempt will fail.

It all results in the following question: if we do not validate ABI in setData then is anything from getData really trustable?

[CJ42]

Is this implemented already I think?

[CJ42]

Is this implemented already I think?

It's actually not. It's an extra 2 byte at the beginning to return the CompactBytesArray.length

[CJ42]

I'll put the question differently: if we are calling setData with key whose valueType is some_type[]

In the case of setData, the type for the data key will always be bytes32

[JeneaVranceanu]

I think you are talking about key and I was talking about value for setData(key, value).
To keep it short: we need to add array length support for CompactBytesArray.