I am having problems trying to create a query that allow me to group by months and by a column that has different values.
The following is a small representation of the table and columns I need to query.
Table name is requests and has two columns date and status. Status may have the values pending, attended, absent and canceled.
I want to get a query that looks like this
Right now i am trying a subquery on status for each possible value. It works but is a very slow query. It takes arround 48s for 8000 rows.
SELECT
MONTHNAME(date)
(SELECT count(status) FROM requests WHERE status = "pending"),
(SELECT count(status) FROM requests WHERE status = "attended"),
(SELECT count(status) FROM requests WHERE status = "absent"),
(SELECT count(status) FROM requests WHERE status = "canceled")
FROM request
GROUP BY 1;
Any recommendations on how to get the result efficiently? Thank you very much
You could use case whene on status instead of several subselect
select
MONTHNAME(date)
, sum( case when status = "pending" then 1 else 0 end) pending
, sum( case when status = "attended" then 1 else 0 end) attended
, sum( case when status = "absent" then 1 else 0 end) absent
, sum( case when status = "canceled" then 1 else 0 end) canceled
FROM request
GROUP BY MONTHNAME(date) ;
Related
How do i formulate a sql query where i can display the number of counts of success and errors and warning for the tables for latest day . So I can create a visualization in dashboard. The ultimate goal is for a user to know on a current day how many healthy pipelines are there and how many unhealthy pipelines are there.
i tried something like this, but i complicated it.
SELECT status, COUNT(*) as Total, cast(cast(time as Timestamp) as date) as time,
SUM(CASE WHEN Status = 'Success' THEN 1 ELSE 0 END) as Success,
SUM(CASE WHEN Status = 'Warning' THEN 1 ELSE 0 END) as Warning,
SUM(CASE WHEN Status = 'Error' THEN 1 ELSE 0 END) as Error
FROM table-name
where (cast(cast(time as Timestamp) as date) in ({{ Time }}) or {{ Time }} = 'All')
GROUP BY 1, cast(cast(time as Timestamp) as date)
order by cast(cast(time as Timestamp) as date) desc
lets say this is the given table
How do i formulate a sql query where i can display the number of counts of success and errors and warning for the tables for latest day .
You can use something like this:
SELECT DATE(time) as date, COUNT(*) as Total, ,
SUM(Status = 'Success') as Success,
SUM(Status = 'Warning') as Warning,
SUM(Status = 'Error') as Error
FROM table-name
WHERE date(time) = (SELECT MAX(date(time)) FROM table-name)
GROUP BY 1;
I have a table named Order with schema as
user_id, state amount
11 success 100
11 FAILED 10
11 FAILED 10
11 success 17
state can have two values (Success/Failed).
I want to fetch sum(amount) when state = "SUCCESS" - sum(amount) when state = "FAILED"
means difference total amount when success - total amount when failed.
I can solve this problem in 2 queries.
A = select id, sum(amount) when state = "SUCCESS"
B = select id, sum(amount) when state = "FAILED"
And solution will be A-B.
Is there any way I can achieve this in single sql query?
use case when
select user_id,sum(case when state = 'SUCCESS' then amount else 0 end)-sum(case when state = 'FAILED' then amount else 0 end)
from table group by user_id
Use conditional aggregation:
select id,
sum(case when state = 'SUCCESS' then amount else - amount end) as total
from t
where state in ('SUCCESS', 'FAILED')
group by id;
I assume that you want this sum per id and not overall in the table.
select sum(case when state = "SUCCESS" then amount else o end) -
sum(case when state = "FAILED" then amount else o end)
from tbl
group by userid
I have the following query that I am trying to understand and convert
it to a db2 format:
TRANSFORM Sum(Cases) AS SumOfCases
SELECT Process, Sum(Cases) AS total
FROM tbl
GROUP BY Process
PIVOT tbl.STATUS;
Table has data like:
Process Status Cases
a Cancelled 14
a Closed 179
b Cancelled 20
b Closed 30
b Pending 10
How can I write that query to db2?
I tried the following query:
SELECT Process
, MAX(CASE WHEN STATUS = 'Cancelled' THEN CASES END) "Cancelled"
, MAX(CASE WHEN STATUS = 'Closed' THEN CASES END) "Closed"
, MAX(CASE WHEN STATUS = 'Pending' THEN CASES END) "Pending"
FROM tbl
GROUP BY Process;
Since I do not have MS Access hence I am not confident that if I had done the right thing in db2 or not.
Would appreciate if I could get some advice on this.
Your DB2 query works correctly in replicating Access's crosstab query except you missed the Total column. By the way any aggregate function would work for your CASE/WHEN statements: MIN(), MAX(), MEDIAN(), AVG(), even SUM():
SELECT Process
, SUM(CASES) AS "Total"
, MAX(CASE WHEN STATUS = 'Cancelled' THEN CASES END) AS "Cancelled"
, MAX(CASE WHEN STATUS = 'Closed' THEN CASES END) AS "Closed"
, MAX(CASE WHEN STATUS = 'Pending' THEN CASES END) AS "Pending"
FROM tbl
GROUP BY Process;
I'm having trouble writing a SQL query to show the status of a row as it's own column as shown in the picture.
I was thinking of using an alias for Status as:
SELECT 'Table A.Date', 'Table A.Status' as ... FROM Table A;
But this doesn't resolve the issue on how to display each status type as their own column and number value.
Can someone point out how to do this?
Try this one. I used CASE statement to conditionally count the status as one depending on the status type given.
SELECT
Date,
SUM(CASE WHEN Status='Pending' THEN 1 ELSE 0 END)Pending,
SUM(CASE WHEN Status='Completed' THEN 1 ELSE 0 END)Completed,
SUM(CASE WHEN Status='Cancelled' THEN 1 ELSE 0 END)Cancelled
FROM Table A
WHERE Date='2014-01-01'
GROUP BY Date
Try this:
select A.date,
count(
case
when A.Status='Pending'
Then 1
Else NULL
End
) as Pending,
count(
case
when A.Status='Completed'
Then 1
Else NULL
End
) as Completed,
count(
case
when A.Status='Cancelled'
Then 1
Else NULL
End
) as Cancelled
From A
group by A.date
I have a table which looks something like this...
supplier_name status count
supplier_a Unavailable 1
supplier_a Refunded 5
supplier_a Shipped 10
supplier_z Refunded 2
supplier_z Shipped 4
I know exactly what columns I want and I would like to use the values from the first table to return values structured like the following...
supplier_name unavailable refunded shipped
supplier_a 1 5 10
supplier_z 0 2 4
(note: in cases where there is no value for a specific column it should be set to 0)
Would this be possible to achieve in a query?
I think something like this should work:
SELECT Supplier_Name,
MAX(CASE WHEN STATUS = 'Unavailable' THEN Count ELSE 0 END) as Unavailable,
MAX(CASE WHEN STATUS = 'Refunded' THEN Count ELSE 0 END) as refunded,
MAX(CASE WHEN STATUS = 'Shipped' THEN Count ELSE 0 END) as shipped
FROM YourTable
GROUP BY Supplier_Name
And here is the SQL Fiddle.
Good luck.
Try this:
SELECT
supplier_name,
SUM(COALESCE(CASE WHEN status = 'unavailable' THEN `count` END, 0)) unavailable,
SUM(COALESCE(CASE WHEN status = 'refunded' THEN `count` END, 0)) refunded,
SUM(COALESCE(CASE WHEN status = 'shipped' THEN `count` END, 0)) shipped
FROM tbl
GROUP BY supplier_name
SQL FIDDLE DEMO