I have the following mariadb table. The data is added 3 times per day. I am looking to write a SQL query that would give me the average amount for the day. This way I can say on May 13 'serender' averaged x amt, 'shilta' averaged x amt and 'snowq' averaged x amt. On May 14th the averages were... and so on for each date.
key | timestamp | card | amt |
-------------------------------------------
| 126 | 1620837006 | serender | 8040 |
| 127 | 1620837006 | shilta | 752 |
| 128 | 1620837006 | snowq | 308 |
| 132 | 1620862207 | serender | 846 |
| 133 | 1620862207 | shilta | 803 |
| 134 | 1620862207 | snowq | 759 |
| 139 | 1620894616 | serender | 845 |
| 140 | 1620894616 | shilta | 805 |
| 141 | 1620894616 | snowq | 759 |
| 146 | 1620923404 | serender | 869 |
| 147 | 1620923404 | shilta | 804 |
| 148 | 1620923404 | snowq | 759 |
| 153 | 1620948607 | serender | 755 |
| 154 | 1620948607 | shilta | 650 |
| 155 | 1620948607 | snowq | 530 |
If you want to see the date then convert it from a Unix timestamp to a date:
select date(from_unixtime(timstamp)) as dte, card, avg(amt)
from t
group by dte, card;
Related
I'm a bit new in sql.
I have the following table:
+-----+---------+------------------------+
| ID | ID_TEST | FILE_PATH |
+-----+---------+------------------------+
| 575 | 3 | Landscapes_001_h_A.jpg |
| 576 | 3 | Landscapes_001_h_B.jpg |
| 577 | 3 | Landscapes_001_h_C.jpg |
| 578 | 3 | Landscapes_001_h_D.jpg |
| 579 | 3 | Landscapes_001_h_E.jpg |
| 580 | 3 | Landscapes_002_h_A.jpg |
| 581 | 3 | Landscapes_002_h_B.jpg |
| 582 | 3 | Landscapes_002_h_C.jpg |
| 583 | 3 | Landscapes_002_h_D.jpg |
| 584 | 3 | Landscapes_002_h_E.jpg |
+-----+---------+------------------------+
The pattern for picture is Landscapes_XXX_h_Y.jpg
where
XXX is number from 1 to 185 and Y is quality version from A to E
I wanna select each image name with different quality.
The output should be
+-----+---------+------------------------+
| ID | ID_TEST | FILE_PATH |
+-----+---------+------------------------+
| 575 | 3 | Landscapes_001_h_A.jpg |
| 576 | 3 | Landscapes_002_h_E.jpg |
| 577 | 3 | Landscapes_003_h_C.jpg |
| 578 | 3 | Landscapes_004_h_B.jpg |
| 579 | 3 | Landscapes_005_h_D.jpg |
| 580 | 3 | Landscapes_006_h_A.jpg |
| 581 | 3 | Landscapes_007_h_E.jpg |
| 582 | 3 | Landscapes_008_h_C.jpg |
| 583 | 3 | Landscapes_009_h_B.jpg |
| 584 | 3 | Landscapes_010_h_E.jpg |
+-----+---------+------------------------+
but of course for 185 elements.
I'm using 5.5.60-MariaDB.
How to write SELECT statement? Using REGEXP?
My data looks like this -
+-----------+-----------+-----------+----------+
| FLIGHT_NO | FL_DATE | SERIAL_NO | PILOT_NO |
+-----------+-----------+-----------+----------+
| 501 | 15-OCT-19 | 456710 | 345 |
| 521 | 16-OCT-19 | 562911 | 345 |
| 534 | 17-OCT-19 | 877694 | 345 |
| 577 | 17-OCT-19 | 338157 | 345 |
| 501 | 14-OCT-19 | 921225 | 346 |
| 534 | 15-OCT-19 | 877694 | 346 |
| 534 | 14-OCT-19 | 338157 | 347 |
| 590 | 16-OCT-19 | 650012 | 347 |
| 531 | 14-OCT-19 | 562911 | 348 |
| 531 | 15-OCT-19 | 562911 | 348 |
| 501 | 16-OCT-19 | 220989 | 349 |
| 521 | 18-OCT-19 | 650012 | 349 |
| 590 | 14-OCT-19 | 562911 | 351 |
| 577 | 18-OCT-19 | 877694 | 351 |
| 590 | 18-OCT-19 | 456710 | 346 |
+-----------+-----------+-----------+----------+
My aim is to return the total number of flights flying and not flying on 18-oct-19.
I'm doing it with dual but that doesn't seem to be the correct/best method.
Can anyone help me do it the correct way?
SELECT
(SELECT COUNT(FLIGHT_NO) NO_FLY FROM schd_flight WHERE fl_date = '18-OCT-19') AS FLY,
(SELECT COUNT(FLIGHT_NO) NO_FLY FROM schd_flight WHERE fl_date <> '18-OCT-19') AS NO_FLY
FROM dual;
My output -
+-----+--------+
| fly | no_fly |
+-----+--------+
| 3 | 12 |
+-----+--------+
Simply use sum with case statement
Select
sum(case when fl_date = '18-OCT-19' then 1 end) fly,
sum(case when fl_date <> '18-OCT-19' then 1 end) no_fly
From schd_flight;
Cheers!!
I think the second query is not necessary, no_fly = total - fly.
So I came up with my solution, may improve the query time :
SELECT sub.FLY as FLY, (SELECT count(*) from schd_flight) - sub.FLY as NO_FLY
FROM (
SELECT COUNT(CASE when fl_date = '18-OCT-19' then 1 end) AS FLY
from schd_flight
) sub;
Not tested yet though.
input:
| a.user_id | a_stream_length | b_stream_length | subtract_inactive |
-----------------------------------------------------------------------------
| a | 11 | 1686 | 22 |
| a | 1686 | 328 | 12 |
| a | 328 | 732 | 22 |
| a | 732 | 11 | 1699 |
| a | 11 | 2123 | 18 |
| a | 2123 | 160 | 2 |
| a | 160 | 1358 | 0 |
| a | 1358 | 129 | 1 |
| a | 129 | 4042 | 109334 |
output:
| a | (1686+11+328+732) (if subtract_inactive < 1000) |
| a | 732(a_stream_length) if subtract_inactive > 1000) |
I have my dataset in the given format
It's a month level data along with salary for each month.
I need to calculate cumulative salary for each month end. How can I do this
+----------+-------+--------+---------------+
| Account | Month | Salary | Running Total |
+----------+-------+--------+---------------+
| a | 1 | 586 | 586 |
| a | 2 | 928 | 1514 |
| a | 3 | 726 | 2240 |
| a | 4 | 538 | 538 |
| b | 1 | 956 | 1494 |
| b | 3 | 667 | 2161 |
| b | 4 | 841 | 3002 |
| c | 1 | 826 | 826 |
| c | 2 | 558 | 1384 |
| c | 3 | 558 | 1972 |
| c | 4 | 735 | 2707 |
| c | 5 | 691 | 3398 |
| d | 1 | 670 | 670 |
| d | 4 | 838 | 1508 |
| d | 5 | 1000 | 2508 |
+----------+-------+--------+---------------+
I need to calculate running total column which is cumulative column. How can I do efficiently in SQL?
You can use SUM with ORDER BY clause inside the OVER clause:
SELECT Account, Month, Salary,
SUM(Salary) OVER (PARTITION BY Account ORDER BY Month) AS RunningTotal
FROM mytable
I have a query that I am trying to filter for a report. Each addressID can have multiple jobs and each job can have multiple elements to it.
Basically I am trying to get the maximum jobID for each addressID, but I want to get each element of the job.
The current Query results are:
+-----------+-------+--------+
| AddressID | JobID | Cost |
+-----------+-------+--------+
| 326 | 328 | £52.50 |
| 327 | 329 | £55.13 |
| 328 | 330 | £57.88 |
| 329 | 331 | £60.78 |
| 329 | 331 | £63.81 |
| 330 | 332 | £67.00 |
| 330 | 332 | £70.36 |
| 330 | 332 | £73.87 |
| 330 | 332 | £77.57 |
| 330 | 333 | £57.75 |
| 330 | 333 | £60.64 |
| 330 | 333 | £63.67 |
| 330 | 333 | £66.85 |
| 331 | 334 | £70.20 |
| 331 | 334 | £73.71 |
| 331 | 335 | £77.39 |
| 331 | 336 | £81.26 |
| 331 | 336 | £85.32 |
| 331 | 336 | £89.59 |
+-----------+-------+--------+
And I am trying to get:
+-----------+-------+--------+
| AddressID | JobID | Cost |
+-----------+-------+--------+
| 326 | 328 | £52.50 |
| 327 | 329 | £55.13 |
| 328 | 330 | £57.88 |
| 329 | 331 | £60.78 |
| 329 | 331 | £63.81 |
| 330 | 333 | £57.75 |
| 330 | 333 | £60.64 |
| 330 | 333 | £63.67 |
| 330 | 333 | £66.85 |
| 331 | 336 | £81.26 |
| 331 | 336 | £85.32 |
| 331 | 336 | £89.59 |
+-----------+-------+--------+
I had been looking at SELECT TOP 1 to isolate the MAX JobID, but ended up limiting the query to just 1 entry.
Currently tweaking this subquery, but still not sure I'm on the right track:
(SELECT Max(vusearch.JobID) FROM vuSearch AS T WHERE PAID = vuSearch.AddressID GROUP BY AddressID)
Can anyone advise?
Here is one method:
select v.*
from vusearch as v
where v.JobId = (select max(v2.JobId)
from vusearch as v2
where v2.AddressId = v.AddressId
);
Managed to get it fixed - I probably hadn't provided enough information as I was trying to keep my explanation simple.
Many thanks for your help Gordon
((vuSearch.PDID) IN ( (SELECT Max(v2.PDID) FROM vuSearch AS v2 GROUP BY v2.PAID)))