Pandas nan 不被视为字符串

Question

I have a csv file containing multiuple crypto currencies like this. Column 2 to 4 is id, symbol, name.

6201,nano-dogecoin,indc,Nano Dogecoin
6202,nano-shiba-inu,NanoShiba,Nano Shiba Inu
6203,nantrade,nan,NanTrade
6204,naos-finance,naos,NAOS Finance
6205,napoleon-x,npx,Napoleon X

I have a function where i get ids by symbols like this:

def symbols_to_ids(self, symbols):
    ids = []
    df = pd.read_csv(os.getcwd() + "/Backtester/Results/Misc/allcoins.csv")
    for index, row in df.iterrows():
        for symbol in symbols:
            
            if str(row["symbol"].lower()) == str(symbol.lower()):
                
                ids.append(row["id"])
    return ids

However i get an error because the one of the symbols is nan. I am pretty sure it gets treated as a float, since this error gets thrown when the row symbol is nan:

 if str(row["symbol"].lower()) == str(symbol.lower()): AttributeError: 'float' object has no attribute 'lower'

I tried to convert it to string, but it does not work. I think this could be solved in pandas, but I dont know how.

Answer 1

You're calling the method before converting. Move .lower to outside conversion:

if str(row["symbol"]).lower() == str(symbol).lower():