pandas 基础操作

pandas数组中一行中有一个列表，把列表拆分

df0 =      asin_name                                 page_bad_review_id
0  B08L5X9SZX                                  ["RE6WNBG0BB94C"]
1  B08VNCJ3D6  ["R3K64G2XVC9CZN", "R2FMNEUEP7IYFG", "RWE7E817...
2  B08VHW531Y  ["R2I32O1ADN9IID", "R3RCR9VCB2NV2E", "R18AJHV5...
3  B08L5VMX7Q                                  ["RJN0J8J5QBWEI"]
4  B08VHWJZKP  ["R2I32O1ADN9IID", "R3RCR9VCB2NV2E", "R18AJHV5...
5  B08PBRNFY6                                                 []


df1 =      asin_name                                          review_id
0  B08L5X9SZX                                      RE6WNBG0BB94C
1  B08VNCJ3D6  R3K64G2XVC9CZN,R2FMNEUEP7IYFG,RWE7E817GGYS7,RO...
2  B08VHW531Y  R2I32O1ADN9IID,R3RCR9VCB2NV2E,R18AJHV57DWZ1A,R...
3  B08L5VMX7Q                                      RJN0J8J5QBWEI
4  B08VHWJZKP  R2I32O1ADN9IID,R3RCR9VCB2NV2E,R18AJHV57DWZ1A,R...
5  B08PBRNFY6                                                   

df2 =      asin_name       review_id
0  B08L5X9SZX   RE6WNBG0BB94C
1  B08VNCJ3D6  R3K64G2XVC9CZN
1  B08VNCJ3D6  R2FMNEUEP7IYFG
1  B08VNCJ3D6   RWE7E817GGYS7
1  B08VNCJ3D6   RO0N62VDEAQOZ
1  B08VNCJ3D6  R27XSTM4EBHSXN
1  B08VNCJ3D6   RBZ3U6LE7KU4W


def json_dump(self,df):
        if '[' in str(df['page_bad_review_id']):
            return str(df['page_bad_review_id']).replace('[','').replace(']','').replace('"','').replace(' ','')
        else:
            return ''


df['review_id'] = df.apply(lambda r: self.json_dump(r), axis=1)
print('df1 = ',df[['asin_name','review_id']])
df = df.drop('review_id', axis=1).join(df['review_id'].str.split(',', expand=True).stack().reset_index(level=1, drop=True).rename('review_id'))

pandas连接mysql

import pymysql
import pandas as pd

#连接数据库
conn = pymysql.connect(host=setting.Host, port=setting.Port, user=setting.User, passwd=setting.Passwd, db=setting.DB, charset='utf8')

sql = """select * from test"""

df = pd.read_sql(sql, con=conn)