Open alexon1234 opened 3 years ago
I like this idea, it would increase the coverage for tokens that return byte32
instead of string
for symbol and name.
I listed some other cases when token detection https://github.com/blockchain-etl/ethereum-etl/blob/develop/ethereumetl/jobs/export_contracts_job.py#L82 doesn't work https://github.com/blockchain-etl/ethereum-etl/issues/195
Hi,
I would like to propose to remove or allow to override the current filter is placed in the
ExtractTokensJob
class that filters all the contracts that are not erc20 or erc721.There are multiple contracts that are not erc20 neither erc721, but they expose the
symbol()
andname()
methods, so we could still populate the tokens.csv with this information.Currently, in the repository https://github.com/blockchain-etl/ethereum-etl-airflow, we keep the following csv. This csv is used to add manually tokens that this ETL is not able to process.
But if we take the tokens address from this csv and run
We will get the following
And just a few errors
The cause of this errors are two things.
0x9f8F72aA9304c8B593d555F12eF6589cC3A579A2
and0x89d24A6b4CcB1B6fAA2625fE562bDD9a23260359
are using byte32 instead of string in thename
andsymbol
methods. This can be handle easily by just doing:0x629A673A8242c2AC4B7B8C5D8735fbeac21A6205
and0xD1E5b0FF1287aA9f9A268759062E4Ab08b9Dacbe
are just NFT using the ERC165. Not having decimals or total supply makes sense, so we could just throw errors and keep the column emptyWhat do you think? I think it could be worth it
Thanks, Alejandro