将表格压缩为一行 进行统计 以供匹配

class Class_test:
    def __init__(self):
        self.fun_5_1()
    
    # 若无则报错AttributeError: 'Class_test' object has no attribute 'fun_zip'
    def fun_zip(
        self,
        df_arg=pd.DataFrame(),
        bool_arg=False
    ):
        # 每列元素打包
        zipped = zip(*[df_arg.iloc[i] for i in range(df_arg.shape[0])])
        # 是否包含标题
        if bool_arg:
            list_zip = [(col_name, *row) for col_name, row in zip(df_arg.columns, zipped)]
        else:
            list_zip = [tuple(row) for row in zipped]
        return list_zip
    
    def fun_columns(
        self,
        list_columns=[str(i)+'_重量' if int(i[3])//2 else i for i in ['col'+str(j) for j in range(3)]],
        str_keyword='重量',
        str_suffix='合计'
    ):
        dict_columns={
            key: value for key, value in zip(
                list_columns,
                [
                    str(z) + str(str_suffix)
                    if str(str_keyword) in str(z)
                    else None
                    for z in list_columns
                ]
            )
        }
        return dict_columns
    
    def fun_5_1(self):
        self.dict_columns = self.fun_columns()
        
    def fun_sum(
        self,
        df_arg=pd.DataFrame(columns=['col0', 'col1', 'col2_重量']),
        int_arg=0,
        bool_on=False,
        list_on=['column_to_merge_on'],
        list_LookupValue=['lookup_value']
    ):
        # 生成1行表格
        # 增加合计列,填入合计重量
        list_values = [value for value in self.dict_columns.values() if value is not None]
        df_zip_arg = pd.DataFrame(
            data={
                key: str(value) for key, value in zip(df_arg.columns, self.fun_zip(df_arg))
            },
            index=[int_arg]
        )
        df_zip_arg = pd.concat(
            objs=[
                df_zip_arg, 
                pd.DataFrame(columns=list_values)
            ],
            axis=1
        )
        if df_arg.shape[0] > 0:
            for value in list_values:
                df_zip_arg.loc[int_arg, value] = df_arg.loc[
                    :, [k for k, v in self.dict_columns.items() if v == value][0]
                ].sum()
        
        # 如果需要新增str_on列以供匹配
        if bool_on:
            df_zip_arg = pd.concat(
                objs=[
                    pd.DataFrame(columns=list_on),
                    df_zip_arg
                ],
                axis=1
            )
            if df_arg.shape[0] > 0:
                dict_on = {k: v for k, v in zip(list_on, list_LookupValue)}
                for key in dict_on.keys():
                    df_zip_arg.loc[int_arg, key] = [v for k, v in dict_on.items() if k == key][0]
        
        return df_zip_arg
obj_test = Class_test()  # 若无括号则报错TypeError: Class_test.fun_sum() missing 1 required positional argument: 'self'
obj_test.fun_sum(pd.DataFrame([[1,1,1],[2,2,2],[3,3,3]],columns=['col0', 'col1', 'col2_重量']))
obj_test.fun_sum(pd.DataFrame(columns=['col0', 'col1', 'col2_重量']))
obj_test.fun_sum(pd.DataFrame([[1,1,1],[2,2,2],[3,3,3]],columns=['col0', 'col1', 'col2_重量']), bool_on=True)
obj_test.fun_sum(pd.DataFrame(columns=['col0', 'col1', 'col2_重量']), bool_on=True)