Getting error about string argument without encoding when writing a CSV reader

[kszlim]

Was testing out develop branch due needing this change #558

chunks = tuple([dict(zip(self.header, line.split(bytes(self.separator)))) for line in process.split(b"\n")])  # type: ignore
TypeError: string argument without an encoding

This was specifically when using a CSVReader with chunked=False and as_dataframe=False

nautilus_trader/nautilus_trader/persistence/external/readers.py

Line 282 in dac7a28

chunks = tuple([dict(zip(self.header, line.split(bytes(self.separator)))) for line in process.split(b"\n")]) # type: ignore

This line breaks due to missing string encoding in the bytes function.

[cjdsellers]

Hi @kszlim thanks for reporting this!

We'll look into it as soon as we can, bandwidth dependent. Hopefully this week though.

[cjdsellers]

Fixed from #593
Many thanks to @limx0 🚀

[772435284]

I think it is not fixed, I still get the error when running the code in https://docs.nautilustrader.io/user_guide/loading_external_data.html. If I do not specify my decoding method in the code, the error still exists. Otherwise, this line should be changed to chunks = tuple([dict(zip(self.header, line.split(bytes(self.separator,encoding='utf8')))) for line in process.split(b"\n")]) # type: ignore