我需要在hive中执行此查询：
SELECT
waybill_no,
vehicle_no,
WAYBILL_TYPE,
etim_no,
trip_no,
route_no,
sum(full_ticket_count) as full_ticket_count,
sum(half_ticket_count) as half_ticket_count,
sum(BOT_ticket_count) as BOT_ticket_count,
sum(manual_ticket_count) as manual_ticket_count,
sum(window_booking_count) as window_booking_count,
sum(pass_ticket_count) as pass_ticket_count,
sum(luggage_ticket_count) as luggage_ticket_count,
format((sum(base_fare)/100),2) as base_fare,
format((sum(asc_amt-ysy_amt)/100),2) as asc_amt,
format((sum(ysy_amt)/100),2) as ysy_amt,
format((sum(bsc)/100),2) as bsc,
format((sum(toll)/100),2) as toll,
format((sum(it)/100),2) as it,
format((sum(sleeper)/100),2) as sleeper,
format((sum(toll_fee)/100),2) as toll_fee,
format((sum(bus_stand_fee)/100),2) as bus_stand_fee,
sum(insp_ticket_count) as insp_ticket_count,
sum(insp_ticket_amt) as insp_ticket_amt,
sum(dhaba_collection) as dhaba_collection,
sum( window_booking) as window_booking,
format((sum(penalty_amt)/100),2) as penalty_amt,
format((sum(luggage_amt)/100),2) as luggage_amt,
format((sum(total_amount)/100),2) as total_amount,
format((sum(total_expenses)/100),2) as total_expenses
FROM (
SELECT
td.waybill_no,
td.vehicle_no,
wb.WAYBILL_TYPE,
td.etim_no,
td.trip_no,
route_no,
sum( if (ticket_code IN('TO','BS','FX','CR'),0,full_ticket_count)) as full_ticket_count,
sum( if (ticket_code IN('TO','BS','FX','CR'),0,half_ticket_count)) as half_ticket_count,
sum( if (ticket_code IN('FX','CR'),0,1)) as BOT_ticket_count,
'0' as manual_ticket_count,
sum( if (ticket_code='WI',1,0)) as window_booking_count,
sum( if (ticket_code='PS',(full_ticket_count+half_ticket_count),0)) as pass_ticket_count,
sum(luggage_ticket_count) as luggage_ticket_count,
sum( if (td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*adult_basic_fare)+(half_ticket_count*child_basic_fare)))) as base_fare,
sum( if (td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*td.asc_amt)+(half_ticket_count*td.asc_amt)))) as asc_amt,
sum(
if (
ticket_code IN('TO','BS'),0,((
if(ticket_code IN('WI'),1,full_ticket_count) * state_code_10
)+(
if(ticket_code IN('WI'),0,half_ticket_count)*state_code_10
))
)
) as ysy_amt,
sum( if(td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*td.bsc_amt)+(half_ticket_count*td.bsc_amt)))) as bsc,
sum( if(td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*td.toll_amt)+(half_ticket_count*td.toll_amt)))) as toll,
sum( if(td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*td.it_amt)+(half_ticket_count*td.it_amt)))) as it,
sum( if(td.ticket_code IN('TO','BS','FX','CR'),0,((full_ticket_count*td.sleeper_amt)+(half_ticket_count*td.sleeper_amt)))) as sleeper,
sum( if(td.ticket_code='TO',(td.total_amount),0)) as toll_fee,
sum( if(td.ticket_code='BS',(td.total_amount),0)) as bus_stand_fee,
sum( if(td.ticket_code='IP',full_ticket_count,0)) as insp_ticket_count,
sum( if(td.ticket_code='IP',(td.total_amount),0)) as insp_ticket_amt,
sum( if(td.ticket_code='DC',(td.total_amount),0)) as dhaba_collection,
sum( if(td.ticket_code='WI',(td.total_amount),0)) as window_booking,
sum( if(td.ticket_code='PN',(td.total_amount),0)) as penalty_amt,
sum( if(td.ticket_code='LG',(td.total_amount),0)) as luggage_amt,
sum(
if(
td.ticket_code IN('TO','BS','FX','CR'),0,td.total_amount
) + if(
ticket_code='PA',adult_basic_fare,0
) - if(
ticket_code='DC',if(docket_no='SBICARD',(total_amount),0),0
)
) as total_amount,
sum(
if(
td.ticket_code NOT IN('TO','BS','WI'),0,td.total_amount
)+if(
td.ticket_code='PA',((full_ticket_count*adult_basic_fare)+(half_ticket_count*child_basic_fare)),0
)+if(
e_purse_flag='1',total_amount,0
)
) as total_expenses,
td.ticket_code
FROM ticket_data as td
INNER JOIN waybillprogramming wb on wb.waybill_no=td.waybill_no
WHERE wb.collection_tm between str_to_date('2016-07-11 00:00:00','%Y-%m-%d %H:%i:%s') and str_to_date('2016-07-11 23:59:59','%Y-%m-%d %H:%i:%s') AND td.`ticket_type` != '51'
GROUP BY wb.waybill_no
UNION
SELECT
wb.waybill_no,
'' as vehicle_no,
wb.WAYBILL_TYPE,
wb.etim_no ,
'0' as trip_no,
'0' as route_no,
'0' as full_ticket_count,
'0' as half_ticket_count,sum(NO_OF_PASSENGER) as BOT_ticket_count,
'0' as manual_ticket_count,
'0' as window_booking_count,
'0' as pass_ticket_count,
'0' as luggage_ticket_count,
'0' as base_fare,
'0' as asc_amt,
'0' as ysy_amt,
'0' as bsc,
'0' as toll,
'0' as it,
'0' as sleeper,
'0' as toll_fee,
'0' as bus_stand_fee,
'0' as insp_ticket_count,
'0' as insp_ticket_amt,
'0' as dhaba_collection,
'0' as window_booking,
'0' as penalty_amt,
'0' as luggage_amt,
( sum( TICKET_SALE_AMOUNT + RECOVERY_AMT + DHABA_COLLECTION ) * 100 ) as total_amount,
( sum( mtc.BRIDGE_FEE + mtc.BUS_STAND_FEE + mtc.REFUND_AMT + mtc.DIESEL_PERCHASED + mtc.OTHER_EXPENSES + mtc.OTHER_BOOKING + mtc.PART_CASH ) * 100 ) as total_expenses,
'NN' as ticket_code
FROM manual_ticket_collection mtc
INNER JOIN waybillprogramming wb on mtc.WAYBILL_NO = wb.WAYBILL_NO
WHERE collection_tm between str_to_date('2016-07-11 00:00:00','%Y-%m-%d %H:%i:%s') and str_to_date('2016-07-11 23:59:59','%Y-%m-%d %H:%i:%s')
GROUP BY wb.waybill_no
UNION
SELECT
wb.waybill_no,
'' as vehicle_no,
wb.WAYBILL_TYPE,
wb.etim_no,
'0' as trip_no,
'0' as route_no,
'0' as full_ticket_count,
'0' as half_ticket_count,
'0' as BOT_ticket_count,
count(ticket_code) as manual_ticket_count,
sum(if(ticket_code='WI',(full_ticket+half_ticket),0)) as window_booking_count,
sum(if(ticket_code='PS',(full_ticket+half_ticket),0)) as pass_ticket_count,
sum(lugg_ticket) as luggage_ticket_count,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*full_base_fare)+(half_ticket*half_base_fare)))*100) as base_fare,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*mtd.asc_amt)+(half_ticket*mtd.asc_amt)))*100) as asc_amt,
'0' as ysy_amt,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*mtd.bsc_amt)+(half_ticket*mtd.bsc_amt)))*100) as bsc,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*mtd.toll_amt)+(half_ticket*mtd.toll_amt)))*100) as toll,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*mtd.it_amt)+(half_ticket*mtd.it_amt)))*100) as it,
sum(if(mtd.ticket_code IN('TO','BS'),0,((full_ticket*mtd.sleeper_amt)+(half_ticket*mtd.sleeper_amt)))*100) as sleeper,
sum(if(mtd.ticket_code='TO',(mtd.total_amt),0)) as toll_fee,
sum(if(mtd.ticket_code='BS',(mtd.total_amt),0)*100) as bus_stand_fee,
sum(if(mtd.ticket_code='IP',full_ticket,0)) as insp_ticket_count,
sum(if(mtd.ticket_code='IP',(mtd.total_amt),0)*100) as insp_ticket_amt,
sum(if(mtd.ticket_code='DC',(mtd.total_amt),0)) as dhaba_collection,
sum(if(mtd.ticket_code='WI',(mtd.total_amt),0)) as window_booking,
sum(if(mtd.ticket_code='PN',(mtd.total_amt),0)*100) as penalty_amt,
sum(if(mtd.ticket_code='LG',(mtd.total_amt),0)*100) as luggage_amt,
sum(total_amt*100) as total_amount,
sum((if(ticket_code='TO', total_amt,0)+if(ticket_code='BS', total_amt,0))*100) as total_expenses,
mtd.ticket_code
FROM manual_ticket_details mtd
INNER JOIN waybillprogramming wb on mtd.WAYBILL_NO = wb.WAYBILL_NO
WHERE collection_tm between str_to_date('2016-07-11 00:00:00','%Y-%m-%d %H:%i:%s') and str_to_date('2016-07-11 23:59:59','%Y-%m-%d %H:%i:%s')
GROUP BY wb.waybill_no
)abc
GROUP BY waybill_no;
但是我在执行查询时遇到此错误
FAILED：SemanticException [错误10025]：第35行：2表达式不在 GROUP BY键'waybill_no'）
答案 0 :(得分：1)
您正在选择td.waybill_no并在GROUP BY wb.waybill_no上进行分组。由于它是内部联接，请尝试选择wb.waybill_no
答案 1 :(得分：1)
在Hive上从Apache wiki报价
但是，下面的查询不起作用：
SELECT
a,
b
FROM
t1
GROUP BY
a;
这是因为select子句还有一个附加列（b），它不包含在group by子句中（并且它也不是聚合函数）。这是因为，如果表t1看起来像：
| a b |
| 100 | 1 |
| 100 | 2 |
| 100 | 3 |
由于仅对a进行分组，Hive应该为a = 100的组显示b的哪个值？有人可以说它应该是第一个值或最低值，但是我们都同意有多种可能的选择。 Hive通过使select子句中未包含在group by子句中的列变为无效SQL（准确地说是HQL）来消除这种猜测。
在MapReduce范例中将其视为reduce语句。必须指定聚合函数，否则必须将列包含在GROUP BY子句中。