It’s a shame that the authors of this standard didn’t wait a bit longer before flagging it as final as I’m noticing slight inconsistencies with the surrounding ecosystem.
E.g. EIP-721 which this contract aspires to be compatible with has a similar function to function userOf(uint256 tokenId) but in case the tokenId doesn’t exist, it throws. See OZ’s implementation. So it is inconsistent.
EIP-4907, however, made the odd choice of returning the zero address as a signal, which I judge as a bad decision given that I conceptually see function ownerOf to be very similar to function userOf.
And before someone misunderstands: 4907 mandates returning the zero address in the standard and 721 mandates throwing so these are NOT implementation details, they are interface inconsistencies!
eip-721.md
/// @notice Find the owner of an NFT
/// @dev NFTs assigned to zero address are considered invalid, and queries
/// about them do throw.
/// @param _tokenId The identifier for an NFT
/// @return The address of the owner of the NFT
function ownerOf(uint256 _tokenId) external view returns (address);
eip-4907.md
/// @notice Get the user address of an NFT
/// @dev The zero address indicates that there is no user or the user is expired
/// @param tokenId The NFT to get the user address for
/// @return The user address for this NFT
function userOf(uint256 tokenId) external view returns(address);