fix: specify encoding when calling bytes [APE-1186]

[z80dev]

What I did

Specify encoding as utf-8 when calling bytes

fixes: #1535

fixes: APE-1181

How I did it

in this case we should always be dealing with utf8 strings

How to verify it

I kept getting contract deployment failures like reported in the discord and in #1535
with this fix I was able to deploy contracts successfully

Checklist

• All changes are completed
• New test cases have been added
• Documentation has been updated

[antazoey]

test?

[antazoey]

test can be really simple:

def test_txn_str_when_data_is_bytes(ethereum):
    txn = ethereum.create_transaction(data=HexBytes("0x123"))
    actual = str(txn)
    assert isinstance(actual, str)

^ this fails on main

[victor-ego]

I recently encountered an analogous issue, and the resolution was accomplished via the installation of this specific branch.

This is to shed light on a plausible implication associated with the constant handling of UTF-8 formatted strings, particularly when they are passed as ape strings that represent parameters such as the Universal Resource Identifier (URI) for NFT images.

In the wake of recent updates to the Vyper, Python, and Ape environments, I have started experiencing difficulties with a previously operational contract. This has specifically impacted the ability to load NFT image assets on various marketplaces including, but not limited to, Rarible and OpenSea.

Token Minted Transaction: Goerli Etherscan
Base URI: ipfs://QmVTXG1tr3QpgwQ3n67qYmPqHgVbSMpDQnNAyqFtLD2HPH
Contract Source: GitHub - Ape Academy ERC721

Please note the associated links for further perusal and analysis.

[antazoey]

LGTM - I added the test for you.