Search code examples
pythonpandasdataframebinance

Transform DataFrame of Dataframe into Single DataFrame Selecting only Some Columns Python

I have the next Json File:

exchangeInfo = {
    "timezone": "UTC",
    "serverTime": 1565246363776,
    "rateLimits": [],
    "exchangeFilters": [],
    "symbols": [
        {
            "symbol": "ETHBTC",
            "status": "TRADING",
            "baseAsset": "ETH",
            "baseAssetPrecision": 8,
            "quoteAsset": "BTC",
            "quotePrecision": 8,
            "quoteAssetPrecision": 8,
            "baseCommissionPrecision": 8,
            "quoteCommissionPrecision": 8,
            "filters": [
                {
                    "filterType": "PRICE_FILTER",
                    "minPrice": "0.00000100",
                    "maxPrice": "100000.00000000",
                    "tickSize": "0.00000100",
                },
                {
                    "filterType": "PERCENT_PRICE",
                    "multiplierUp": "1.3000",
                    "multiplierDown": "0.7000",
                    "avgPriceMins": 5,
                },
                {
                    "filterType": "LOT_SIZE",
                    "minQty": "0.00100000",
                    "maxQty": "100000.00000000",
                    "stepSize": "0.00100000",
                },
            ],
        },
    ],
}

And applying the next code, I transform the column "filters" into columns.

df = pd.json_normalize(exchangeInfo["symbols"])
df = pd.concat(
    [
        df,
        df.pop("filters")
        .apply(lambda x: dict(i for d in x for i in d.items()))
        .apply(pd.Series),
    ],
    axis=1,
).drop(columns="filterType")
print(df)

Prints:

symbol   status baseAsset  baseAssetPrecision quoteAsset  quotePrecision  quoteAssetPrecision  baseCommissionPrecision  quoteCommissionPrecision    minPrice         maxPrice    tickSize multiplierUp multiplierDown  avgPriceMins      minQty           maxQty    stepSize
0  ETHBTC  TRADING       ETH                   8        BTC               8                    8                        8                         8  0.00000100  100000.00000000  0.00000100       1.3000         0.7000             5  0.00100000  100000.00000000  0.00100000

But, I would like to select only 2 of those filters, by filterType name, I would like to have "PRICE_FILTER" and "LOT_SIZE"

Solution

  • To get columns from only "PRICE_FILTER" and "LOT_SIZE" filters, try:

    df = pd.json_normalize(exchangeInfo["symbols"])
df = pd.concat(
    [
        df,
        df.pop("filters")
        .apply(
            lambda x: dict(
                i
                for d in x
                for i in d.items()
                if d["filterType"] in {"PRICE_FILTER", "LOT_SIZE"}
            )
        )
        .apply(pd.Series),
    ],
    axis=1,
).drop(columns="filterType")
print(df)

    Prints:

       symbol   status baseAsset  baseAssetPrecision quoteAsset  quotePrecision  quoteAssetPrecision  baseCommissionPrecision  quoteCommissionPrecision    minPrice         maxPrice    tickSize      minQty           maxQty    stepSize
0  ETHBTC  TRADING       ETH                   8        BTC               8                    8                        8                         8  0.00000100  100000.00000000  0.00000100  0.00100000  100000.00000000  0.00100000