python数据处理

最新推荐文章于 2024-02-20 16:19:06 发布

zn505119020

最新推荐文章于 2024-02-20 16:19:06 发布

阅读量242

点赞数

分类专栏：工作相关

本文链接：https://blog.csdn.net/zn505119020/article/details/74616084

版权

工作相关专栏收录该内容

136 篇文章 0 订阅

订阅专栏

import pandas
import numpy as np
if __name__ == '__main__':
    a_p = "viomall_pitem_specific.csv"
    b_p = "ebay_specific_required.csv"
    a_df = pandas.DataFrame(
        pandas.read_csv(a_p, encoding="gbk", keep_default_na=True))
    b_df = pandas.DataFrame(
        pandas.read_csv(b_p, encoding="gbk", keep_default_na=True))
    skus = a_df.get("product_sku").drop_duplicates()
    reslut = pandas.DataFrame(columns=("product_sku", "specify_name_1", "specify_value", "specify_name_2", "state"))
    for sku in skus[0:3]:
        print(sku)
        A_df = pandas.DataFrame(data=a_df[a_df["product_sku"] == sku])
        ge_category1 = A_df.get("ge_category1").values[0]
        A_df = A_df[["product_sku",  "specify_name", "specify_value"]]
        B_df = pandas.DataFrame(data=b_df[b_df["ge_category1"] == ge_category1])
        B_df = B_df[["specify_name" , "state"]]

        for i in A_df.values:
            specify_name = i[1]
            tm_pd = pandas.DataFrame(data=(B_df[B_df["specify_name"] == specify_name]))
            state = np.nan
            if len(tm_pd.get("state").values) > 0 :
                state = tm_pd.get("state").values[0]
            if len(tm_pd.values) == 1 :
                reslut = reslut.append( pandas.DataFrame(
                    {"product_sku":i[0] , "specify_name_1":specify_name , "specify_value":i[2] , "specify_name_2":specify_name , "state":state} , index=["0"]) ,
                    ignore_index = True)
            else :
                reslut = reslut.append(pandas.DataFrame(
                    {"product_sku": i[0], "specify_name_1": specify_name, "specify_value": i[2],
                     "specify_name_2": np.nan, "state": state}, index=["0"]),
                    ignore_index=True)

        for i in B_df.values:
            specify_name = i[0]
            tm_pd = pandas.DataFrame(data=(A_df[A_df["specify_name"] == specify_name]))
            if len(tm_pd.values) == 0 :
                reslut = reslut.append(pandas.DataFrame(
                    {"product_sku": sku, "specify_name_1": np.nan, "specify_value": np.nan,
                     "specify_name_2": specify_name , "state": i[1]}, index=["0"]),
                    ignore_index=True)

        reslut.to_csv('sum.csv')