使用pandas对两个dataframe进行join的实例

#coding=utf-8
from pandas import Series,DataFrame
import pandas as pd
#reader1 = pd.read_csv('Dm_Mobile.txt',iterator=True,encoding="gb2312")
#df1 = reader1.get_chunk(10)
#reader2 = pd.read_csv('201604.csv',iterator=True,encoding="gb2312")
#df2 = reader.get_chunk(10)

#读取两个csv文件,生成dataframe
df1 = pd.read_csv('Dm_Mobile.csv',encoding="gb2312")
df2 = pd.read_csv('201604.csv',encoding="gb2312")

#截取手机号前七位,作为新列添加到dataframe
df2['p7s'] = Series([str(x)[:7] for x in Series(df2[u'手机号'])])
df2['p7i'] = df2['p7s'].astype("int64")

#在两个dataframe的手机前七位列上创建索引
index_df1 = df1.set_index('MobileNumber')
index_df2 = df2.set_index(['p7i'])

#以手机号前七位列为join列，对两个dataframe进行join
result = pd.concat([index_df1, index_df2], axis=1, join='inner')

#选取需要显示的列，重新生成result
result = result.reindex(columns=[u'积分商城订单号', u'手机号',u'产品编码',u'商品名称',u'商品价格',u'数量',u'虚拟码',u'消费时间',u'时间',u'兑换渠道商',u'MobileArea'])

#写入到excel文件中
writer = pd.ExcelWriter('pandas_simple.xlsx')
result.to_excel(writer, sheet_name=u'设计院',index=False)
writer.save()