将数据框行值插入 json 字符串

Question

如果值不是 nan，是否可以将数据框中的行值添加到 json 字符串？

df = pd.DataFrame({"col1": ["foo", None], "col2": [20, 19]})
string = """{
        "foo": {
            "bar": ""
        },
        "baz": "qux"
    }
    """
processed_string = json.loads(string)

if df['col1'] !='nan':
    processed_string["foo"]["bar"]["col1"] = df["col1"]
if df['col2'] !='nan':
    processed_string["foo"]["bar"]["col2"] = df["col2"]

预期的输出字符串将是 {"foo": {"bar": {"col1": "foo", "col2": 20}}} {"foo": {"bar": {"col2": 19}}}

Answer 1

可能会改变

        "foo": {
            "bar": ""
        },
        "baz": "qux"
    }
    """

至

        "foo": {
            "bar": {}
        },
    "baz": "qux"
    }
    """

由于您正在尝试将元素插入字典 baz