Question

我有以下工作python代码来进行impala连接/查询：

import pandas as pd


query = 'select my_c_instance_id, count(my_c_instance_id) as my_ins_id_count from ' + \
         '(select * from my_table where my_c_id like "%small%") as small_table' + \
         ' group by(my_c_instance_id)'
cursor = impala_con.cursor()
cursor.execute('USE my_db')
cursor.execute(query)
df_result = as_pandas(cursor)
df_result

代码工作正常，但我想知道是否可以将其分成两个更易读的部分，例如：

small_table = 'select * from my_table where my_c_id like "%small%"'

query = 'select my_c_instance_id, count(my_c_instance_id) as my_ins_id_count from small_table group by(my_c_instance_id)'

cursor = impala_con.cursor()
cursor.execute('USE my_db')
cursor.execute(query)
df_result = as_pandas(cursor)
df_result

如果可能的话，我如何让上述想法真正起作用？感谢。

Answer 1

除非我误解了某些内容，否则根本不需要子查询，只需将where条件移至主查询：

select my_c_instance_id, count(my_c_instance_id) as my_ins_id_count 
from my_table 
where my_c_id like '%small%'
group by my_c_instance_id

SQL / Impala将嵌套查询分解为更易读的格式

1 个答案: