pandas使用pipe整合流程

pandas使用pipe的官方文档：https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.pipe.html

import pandas as pd


def drop_duplicates(df, column_name):
    """删除"""
    return df.drop_duplicates(subset=column_name)


def add_new_column(dataframe, new_column={}):
	"""添加新的数据"""
    for key, value in new_column.items():
        dataframe[key] = value
    return dataframe


def main():
    dataframe = pd.DataFrame({
        "a": [1, 2, 3, 4, 4],
        'b': [1, None, None, 4, 4],
    })
    df_processed = (
        dataframe
            .pipe(drop_duplicates, column_name=['a'])
            .pipe(add_new_column, new_column={"c": [1, 2, 3, 4]})
    )
    print(df_processed)


if __name__ == '__main__':
    main()

得到的结果如下：

   a    b  c
0  1  1.0  1
1  2  NaN  2
2  3  NaN  3
3  4  4.0  4