PyODPS是MaxCompute的Python版本的SDK,提供简单方便的Python编程接口。PyODPS支持类似Pandas的快速、灵活和富有表现力的数据结构。您可以通过PyODPS提供的DataFrame API使用Pandas的数据结果处理功能。本文用于帮助您快速开始使用PyODPS,并且能够用于实际项目。
Pyodps提供了两种执行SQL语句的方法,execute_sql与run_sql,前者会阻塞调起SQL实例,而后者是不会阻塞的,可实现并行
from odps import ODPS
import sys
reload(sys)
#修改系统默认编码 查询出的中文乱吗解决
sys.setdefaultencoding("utf-8")
# print ('ds=' + args['bizdate'])
ds = args['bizdate']
sql = 'select c.cid,l.name from xgj.classes_out as c ,xgj.locations_deal as l where c.ds ='+ds+' and l.ds = '+ds+ ' and c.cid = l.cls_id'
instance = o.execute_sql(sql)
with instance.open_reader() as reader:
# print (reader)
# for record in reader 遍历这2万条数据,这里通过切片只取10条。
#切片只取1条
for record in reader[:1]:
print(record)
name = record.name
cid = record.cid
u_sql = '''UPDATE xgj.classes_out SET school_name = '%s' where ds='%s' and cid = '%s' '''%(name,ds,cid)
print(u_sql)
# print(name)
#odps.execute_sql(u_sql) # 阻塞方法
odps.run_sql(u_sql) # 异步方法
异常错误
Please add put { "odps.sql.submit.mode" : "script"} for multi-statement query in settings
想实现拼接多条sql一起执行,报以上的问题,解决方式 参考
sql = '''
UPDATE xgj.classes_out SET school_name = '韩店镇中心小学' where ds='20211110' and cid = '612b34c2996163490ab8adb5' ;
UPDATE xgj.classes_out SET school_name = '山东省泰安市岱岳区开元中学' where ds='20211110' and cid = '5e905a37a499a70e9c5b30f0' ;
UPDATE xgj.classes_out SET school_name = '蔡亭小学' where ds='20211110' and cid = '6147ca898f2bd2cd28739529' ;
'''
odps.execute_sql(sql,hints={"odps.sql.submit.mode":"script"})