Reputation: 103
I have a table like this:
CITY QNT EXP RATE
LONDON 60 6 900
LONDON 35 8 337
LONDON 24 6 300
LONDON 22 6 266
BIRMINGHAM 22 6 266
NEWYORK 69 19 263
LONDON 21 6 250
ROME 24 7 242
BIRMINGHAM 24 7 242
BIRMINGHAM 24 7 242
LONDON 20 6 233
BIRMINGHAM 23 7 228
STUTTGART 29 9 222
LONDON 19 6 216
STUTTGART 25 8 212
PARIS 31 10 210
STUTTGART 34 11 209
STUTTGART 34 11 209
BIRMINGHAM 18 6 200
BIRMINGHAM 18 6 200
NEWYORK 18 6 200
BIRMINGHAM 17 6 183
LONDON 19 7 171
MUNICH 16 6 166
PARIS 21 8 162
STUTTGART 39 15 160
BARCELONA 18 7 157
LONDON 18 7 157
ROME 33 13 153
BARCELONA 15 6 150
PARIS 25 10 150
ROME 20 8 150
PARIS 25 10 150
ROME 20 8 150
LONDON 15 6 150
MUNICH 15 6 150
BIRMINGHAM 15 6 150
NEWYORK 15 6 150
LONDON 17 7 142
MUNICH 17 7 142
Here is my sql command:
select CITY, QNT, EXP, (QNT-EXP)*100/EXP as RATE
from tbl_city
order by RATE desc
I want to group by city these results. But I couldn't do it. I want the top5 line to change most.
Result should be like that:
LONDON 60 6 900
BIRMINGHAM 22 6 266
NEWYORK 69 19 263
ROME 24 7 242
STUTTGART 29 9 222
Upvotes: 2
Views: 1348
Reputation: 426
This is from MySQL 5.6.
select CITY, QNT, EXP, (QNT-EXP)*100/EXP as RATE
from tbl_city GROUP BY CITY
order by RATE desc LIMIT 5
Tested in this link: http://www.sqlfiddle.com/#!9/3f1ea1/1
This is from MS SQL Server 2017
select TOP 5 f.CITY, f.QNT, f.EXP, x.RATE
from (
select CITY, MAX((QNT-EXP)*100/EXP) as RATE
from tbl_city GROUP BY CITY
) as x inner join tbl_city as f on f.CITY = x.CITY
and ((f.QNT-f.EXP)*100/f.EXP) = x.RATE
ORDER BY RATE DESC;
Tested in this link: http://www.sqlfiddle.com/#!18/7b8da/61
Upvotes: 3
Reputation: 50163
I don't know what is logic behind the specified formula but this does what you want
select top 5 * from (
select top(1) with ties city, qnt, exp, (qnt-exp)*100/exp as rate
from tbl_city
order by row_number() over (partition by city order by (qnt-exp)*100/exp desc)
)t
order by rate desc
However, top(1) with ties with analytical functions are available in SQL Server.
Upvotes: 0
Reputation: 12318
Maybe you want something like this?
select top 5 CITY, QNT, EXP, RATE
from (
select *, row_number() over (partition by CITY order by RATE desc) AS RN
from (
select CITY, QNT, EXP, (QNT-EXP)*100/EXP as RATE
from tbl_city
) X
) Y
where RN = 1
order by RATE desc
I didn't test this, but it should take first the row for the city with biggest rate, and then take top 5 rows so that that the same city is not duplicated
Upvotes: 2