I want to convert a sum result into percentage. When added it will result 100% in total. I already group it by remarks and already got the total from my query. But its still in number.
This is my query
SELECT
CASE
WHEN A.district = B.district THEN 'Delivered from store in same district'
WHEN A.city = B.city THEN 'Delivered from store in same city'
WHEN A.province = B.province THEN 'Delivered from store in same province'
ELSE 'Not delivered from nearest store'
END AS 'REMARK',
SUM(CASE
WHEN YEAR(T1.create_time) = '2019' AND MONTH(T1.create_time) = '01' THEN 1
ELSE 0
END) AS 'Jan-19',
SUM(CASE
WHEN YEAR(T1.create_time) = '2019' AND MONTH(T1.create_time) = '02' THEN 1
ELSE 0
END) AS 'Feb-19',
SUM(CASE
WHEN YEAR(T1.create_time) = '2019' AND MONTH(T1.create_time) = '03' THEN 1
ELSE 0
END) AS 'Mar-19',
SUM(CASE
WHEN YEAR(T1.create_time) = '2019' AND MONTH(T1.create_time) = '04' THEN 1
ELSE 0
END) AS 'Apr-19'
FROM
au_store_add A
INNER JOIN
(SELECT
A.id,
A.code,
A.create_time,
A.plat_create_time,
A.appoint_stcode,
A.status
FROM deli_order A
WHERE A.appoint_stcode IS NOT NULL
UNION
SELECT
B.id,
B.code,
B.create_time,
B.plat_create_time,
B.destination_code,
B.status
FROM deli_order B
WHERE B.destination_code IS NOT NULL) T1 ON T1.appoint_stcode = A.store_code
INNER JOIN
deli_order_delivery B ON B.order_id = T1.id
WHERE
YEAR(T1.plat_create_time)='2019' and T1.status = 6 and (month(T1.plat_create_time) in (1,2,3,4))
GROUP BY
CASE
WHEN A.district = B.district THEN 'Delivered from store in same district'
WHEN A.city = B.city THEN 'Delivered from store in same city'
WHEN A.province = B.province THEN 'Delivered from store in same province'
ELSE 'Not delivered from nearest store'
END
And the result is like this
REMARK | Jan-19| Feb-19| Mar-19| Apr-19|
-----------------------------------------------------------------------
Delivered from store in same city 252 198 308 283
Delivered from store in same district 114 110 163 138
Delivered from store in same province 1135 976 1387 1208
Not delivered from nearest store 3046 2518 3189 3123
I need to generate something like this for each month(without the 'Grand Total')
REMARK | Jan-19|
-----------------------------------------------
Delivered from store in same city 5.5%
Delivered from store in same district 2.5%
Delivered from store in same province 25%
Not delivered from nearest store 67%
Grand Total 4547 (100%)
What should i add to my query?
Given that you are using MySQL 8+, then COUNT() as an analytic function should be available to use. It comes in handy here:
SELECT
CASE
WHEN A.district = B.district THEN 'Delivered from store in same district'
WHEN A.city = B.city THEN 'Delivered from store in same city'
WHEN A.province = B.province THEN 'Delivered from store in same province'
ELSE 'Not delivered from nearest store'
END AS REMARK,
100.0 * COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-01' THEN 1 END)
/ COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-01'
THEN 1 END) OVER () `Jan-19`,
100.0 * COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-02' THEN 1 END)
/ COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-02'
THEN 1 END) OVER () `Feb-19`,
100.0 * COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-03' THEN 1 END)
/ COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-03'
THEN 1 END) OVER () `Mar-19`,
100.0 * COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-04' THEN 1 END)
/ COUNT(CASE WHEN DATE_FORMAT(create_time, '%Y-%m') = '2019-04'
THEN 1 END) OVER () `Apr-19`
FROM
au_store_add A
...
I used DATE_FORMAT above to make the check on the year and month a bit less verbose. I also used COUNT in place of SUM, which lets us use a slightly shorter conditional expression.
Related
I am trying to return multiple counts and averages from multiple tables sorting by gender and am getting incorrect data. I understand that the following is incorrect, but I am unsure of how to fix it. (Edit: Problem with group by gender. See below.)
Here is the query:
SELECT c.gender AS 'Gender',
COUNT(DISTINCT mr.mailing_recipient_id) AS 'Mailing Recipients',
(SELECT COUNT(DISTINCT CASE WHEN mrc.mailing_recipient_click_type_id = 2 THEN 1 ELSE 0 END) ) AS 'Open Total',
AVG(CASE WHEN mrc.mailing_recipient_click_type_id = 2 THEN 1 ELSE 0 END) AS 'Avg Open',
(SELECT COUNT(DISTINCT CASE WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1 ELSE 0 END) ) AS 'Click Total',
AVG(CASE WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1 ELSE 0 END) AS 'Avg Click',
COUNT(DISTINCT ca.cons_action_contribution_id) AS Donations,
AVG(ca.transaction_amt) AS 'Avg Donation Amt'
FROM ((mailing m
LEFT JOIN mailing_recipient mr ON m.mailing_id = mr.mailing_id)
LEFT JOIN mailing_recipient_click mrc ON mr.mailing_recipient_id = mrc.mailing_recipient_id
LEFT JOIN cons_action_contribution ca ON mr.cons_id = ca.cons_id
LEFT JOIN cons c ON c.cons_id = ca.cons_id)
WHERE m.mailing_id = 1
AND gender IS NOT NULL
GROUP BY c.gender;
Here is the table which would be correct if the totals in the fields were correct:
GENDER Mailing Recipient Open Total Avg Open Click Total Avg Click Donations Avg Amt
F 105 2 0.5000 2 0.5000 105 22.5000
M 98 2 0.5000 2 0.5000 98 18.8780
EDIT: Here is an example of what I am hoping to achieve. I am certain that the above values are being repeated. The below values are just examples of what I am expecting:
GENDER Mailing Recipient Open Total Avg Open Click Total Avg Click Donations Avg Amt
F 105 8 0.0761 4 0.0380 2 22.5000
M 98 2 0.0204 1 0.0102 1 18.8000
Edit:
After playing around a bit, I thought that I had discovered that the joining the cons table was what is giving me problematic returns, but the problem is with GROUP BY when using gender. To illustrate, this query (which is grouped by mailing name instead of gender) works beautifully.
select m.mailing_name AS 'mailing',
COUNT(DISTINCT mr.mailing_recipient_id) AS 'Mailing Recipients',
SUM(CASE
when mrc.mailing_recipient_click_type_id = 2 THEN 1
END)
AS 'Open Total',
AVG(CASE
WHEN mrc.mailing_recipient_click_type_id = 2 THEN 1
ELSE 0
END) AS 'Avg Open',
SUM(CASE
WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1
END)
AS 'Click Total',
AVG(CASE
WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1
ELSE 0
END) AS 'Avg Click',
COUNT(ca.cons_action_contribution_id) AS Donations,
AVG(ca.transaction_amt) AS 'Avg Donation Amt'
FROM
mailing m
LEFT JOIN mailing_recipient mr ON m.mailing_id = mr.mailing_id
LEFT JOIN mailing_recipient_click mrc
ON mr.mailing_recipient_id = mrc.mailing_recipient_id
LEFT JOIN cons_action_contribution ca ON mr.cons_id = ca.cons_id
LEFT JOIN cons c ON mr.cons_id = c.cons_id
WHERE m.mailing_id = 1
GROUP BY m.mailing_name;
The statement is identical with the exception of the first and last lines.
Try this:
I'm not sure what you mean by Avg Open and Avg Click.
SELECT c.gender AS 'Gender',
COUNT(DISTINCT mr.mailing_recipient_id) AS 'Mailing Recipients',
SUM(CASE WHEN mrc.mailing_recipient_click_type_id = 2 THEN 1 ELSE 0 END) AS 'Open Total',
AVG(CASE WHEN mrc.mailing_recipient_click_type_id = 2 THEN 1 ELSE 0 END) AS 'Avg Open',
SUM(CASE WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1 ELSE 0 END) AS 'Click Total',
AVG(CASE WHEN mrc.mailing_recipient_click_type_id = 1 THEN 1 ELSE 0 END) AS 'Avg Click',
COUNT(DISTINCT ca.cons_action_contribution_id) AS Donations,
AVG(ca.transaction_amt) AS 'Avg Donation Amt'
FROM mailing m
LEFT JOIN mailing_recipient mr ON m.mailing_id = mr.mailing_id
LEFT JOIN mailing_recipient_click mrc ON mr.mailing_recipient_id = mrc.mailing_recipient_id
LEFT JOIN cons_action_contribution ca ON mr.cons_id = ca.cons_id
LEFT JOIN cons c ON c.cons_id = ca.cons_id
WHERE m.mailing_id = 1
AND gender IS NOT NULL
GROUP BY c.gender;
I also think that mrc.mailing_recipient_click_type_id = 2 means open and mrc.mailing_recipient_click_type_id = 1 mean click seems strange to me. I would expect this data to be exclusive and stored in two different fields.
how can I achieve the desired result in mysql if my table looks like this.
result|year
1 |2011
2 |2011
1 |2011
0 |2011
1 |2012
2 |2012
1 = Won, 2 = lost, 0 = draw
Every year can have multiple values like this. Not sure how I can get the desired result like below.
year won lost draw totalPlayed
2011 2 1 1 3
2012 1 1 0 2
I have tried the following query but does not get the desired result
select year,
league_types.league_name,
sum(if(result = 1,1,0)) as won,
sum(if(result = 0,1,0)) as draw,
sum(if(result = 4,1,0)) as noResult,
sum(if(result = 2,1,0)) as lost,
sum(if(result = 3,1,0)) as tied,
sum(if(result > 0 and result < 4,1,0)) as played
from match_score_card
inner join fixtures on match_score_card.match_id = fixtures.match_id
inner join league_types on fixtures.league_id = league_types.league_id
where
team_id = 1 group by year order by year desc
Here is the SQL Fiddle that demonstrates the following query:
SELECT m.year,
SUM(CASE WHEN m.result = 1 THEN 1 ELSE 0 END) AS 'Won',
SUM(CASE WHEN m.result = 2 THEN 1 ELSE 0 END) AS 'Lost',
SUM(CASE WHEN m.result = 0 THEN 1 ELSE 0 END) AS 'Draw',
COUNT(*) AS 'TotalPlayed'
FROM MyTable AS m
GROUP BY m.year
I'm not familiar with that IF function in mySQL, but this standard SQL should work:
select year
, league_types.league_name
, sum(CASE WHEN result = 1 THEN 1 ELSE 0 END) as won
, sum(CASE WHEN result = 2 THEN 1 ELSE 0 END) as lost
, sum(CASE WHEN result = 3 THEN 1 ELSE 0 END) as draw
, sum(CASE WHEN result = 4 THEN 1 ELSE 0 END) as noResult
, sum(CASE WHEN result = 1
or result = 2 THEN 1 ELSE 0 END) as played
from match_score_card
inner join fixtures
on match_score_card.match_id = fixtures.match_id
inner join league_types
on fixtures.league_id = league_types.league_id
where team_id = 1
group by year, league_types.league_name
order by year desc, league_types.league_name
I'm guessing that you only want to count wins and losses as "played".
I am having a mysql problem I am trying to find both the most current payment value, and, for a particular month (in this query I'm using April). Link to the sqlfillde is here
(case when max(py.pay_date)and month(py.pay_date)= 4 then amount else 0 end) max_pay_april,
This is what I have but it doesn't seem to be working. The most current payment value is: (5, '2012-04-20', 90,) therefore it would be 90 I would really appreciate some help please.
How about this:
select p.name,
v.v_name,
sum(case when Month(py.pay_date) = 4 then amount end) april_amount,
max(case
when month(py.pay_date)= 4
and py.pay_date = (select max(pay_date)
from payment
where month(pay_date) =4 )
then amount else 0 end) max_pay_april,
sum(case
when Month(py.pay_date) = Month(curdate())
then amount end) current_month_amount,
sum(case
when Month(py.pay_date) = Month(curdate())-1
then amount end) previous_month_amount
from persons p
left join vehicle v
on p.id = v.person_veh
left join payment py
on p.id = py.person_id
group by p.name,
v.v_name
see SQL Fiddle with demo
See the SQL query below, it work fine. It calculate the number of Yes, NOT, Other and the number of matching mobile number [Sales field] (D.MobileNo = S.mobile)
SELECT D.Username,
SUM(CASE WHEN D.type = 'Yes' THEN 1 ELSE 0 END) as Yes,
SUM(CASE WHEN D.type = 'Not' THEN 1 ELSE 0 END) as Not,
SUM(CASE WHEN D.type = '' THEN 1 ELSE 0 END) as Other,
SUM(CASE WHEN S.mobile IS NULL THEN 0 ELSE 1 END) as Sales,
COUNT(*) as TOTAL
FROM dairy as D
LEFT JOIN (SELECT DISTINCT mobile FROM sales) as S on D.MobileNo = S.mobile
WHERE source = 'Network'
AND UNIX_TIMESTAMP(CheckDate) >= 1309474800
AND UNIX_TIMESTAMP(CheckDate) <= 1311894000
GROUP BY D.Username
ORDER BY TOTAL DESC
I want to exclude WHERE for the Sales field - it should not be part from CheckDate. Meaning it should check any record in the dairy table without CheckDate for the Sales field.
How can that be done?
If you really want those results in only one query, this might do the trick.
SELECT D.Username,
SUM(CASE WHEN D.type = 'Yes' AND UNIX_TIMESTAMP(CheckDate) >= 1309474800 AND UNIX_TIMESTAMP(CheckDate) <= 1311894000 THEN 1 ELSE 0 END) as Yes,
SUM(CASE WHEN D.type = 'Not' AND UNIX_TIMESTAMP(CheckDate) >= 1309474800 AND UNIX_TIMESTAMP(CheckDate) <= 1311894000 THEN 1 ELSE 0 END) as Not,
SUM(CASE WHEN D.type = '' AND UNIX_TIMESTAMP(CheckDate) >= 1309474800 AND UNIX_TIMESTAMP(CheckDate) <= 1311894000 THEN 1 ELSE 0 END) as Other,
SUM(CASE WHEN S.mobile IS NULL THEN 0 ELSE 1 END) as Sales,
COUNT(*) as TOTAL,
SUM(CASE WHEN UNIX_TIMESTAMP(CheckDate) >= 1309474800 AND UNIX_TIMESTAMP(CheckDate) <= 1311894000 THEN 1 ELSE 0 END) AS TOTALINCHECKDATE
FROM dairy as D
LEFT JOIN (SELECT DISTINCT mobile FROM sales) as S on D.MobileNo = S.mobile
WHERE source = 'Network'
GROUP BY D.Username
ORDER BY TOTAL DESC
Note that "TOTAL" will count all rows (including those who where not within your CheckDate range), TOTALINCHECKDATE return the same value as in your previous query.
Obviously, this can still be optimized.
Assuming username exists also in SALES table
SELECT Username,SUM(Yes) As Yes, SUM(`Not`) As `Not`
, SUM(Other) As Other, SUM(sales) Sales, SUM(total)
FROM (
-- get diary data
SELECT username,mobileNo As mobile,
CASE WHEN D.type = 'Yes' THEN 1 ELSE 0 END as Yes,
CASE WHEN D.type = 'Not' THEN 1 ELSE 0 END as `Not`,
CASE WHEN D.type = '' THEN 1 ELSE 0 END as Other,
0 As sales, 1 as total
FROM dairy as D
WHERE source = 'Network'
AND UNIX_TIMESTAMP(CheckDate) >= 1309474800
AND UNIX_TIMESTAMP(CheckDate) <= 1311894000
UNION ALL
-- get all sales
SELECT DISTINCT username,mobile, 0 as Yes, 0 as `Not`, 0 As Other, 1 As sales, 0 As total
FROM sales
WHERE UNIX_TIMESTAMP(CheckDate) >= 1309474800
AND UNIX_TIMESTAMP(CheckDate) <= 1311894000
) AS a
GROUP BY Username
Also try your original query with date addition
SELECT D.Username,
SUM(CASE WHEN D.type = 'Yes' THEN 1 ELSE 0 END) as Yes,
SUM(CASE WHEN D.type = 'Not' THEN 1 ELSE 0 END) as Not,
SUM(CASE WHEN D.type = '' THEN 1 ELSE 0 END) as Other,
SUM(CASE WHEN S.mobile IS NULL THEN 0 ELSE 1 END) as Sales,
COUNT(*) as TOTAL
FROM dairy as D
LEFT JOIN (SELECT DISTINCT mobile FROM sales WHERE UNIX_TIMESTAMP(CheckDate) >= 1309474800 AND UNIX_TIMESTAMP(CheckDate) <= 1311894000 ) as S on D.MobileNo = S.mobile
WHERE source = 'Network'
AND UNIX_TIMESTAMP(CheckDate) >= 1309474800
AND UNIX_TIMESTAMP(CheckDate) <= 1311894000
GROUP BY D.Username
ORDER BY TOTAL DESC
I am using SUM ( CASE WHEN ) to count number of Yes and No, it work fine.
I am havin problem counting number of matching mobile number from two tables. It dont seem to be counting correctly.
There is a MobileNO field in dairy table and mobile field in the sales table
SELECT
D.Username,
SUM(CASE WHEN D.type = 'Yes' THEN 1 ELSE 0 END) as Yes,
SUM(CASE WHEN D.type = 'No' THEN 1 ELSE 0 END) as No,
SUM(CASE WHEN D.type = '' THEN 1 ELSE 0 END) as Other,
(SELECT SUM(CASE WHEN D.MobileNo = S.mobile THEN 1 ELSE 0 END) from sales as S) as Sales,
COUNT(*) as TOTAL FROM dairy as D
WHERE source = 'Company' AND UNIX_TIMESTAMP(CheckDate) >= 1293840000 AND UNIX_TIMESTAMP(CheckDate) <= 1322697600
group by D.Username order by TOTAL DESC
SELECT
D.Username,
SUM(CASE WHEN D.type = 'Yes' THEN 1 ELSE 0 END) as Yes,
SUM(CASE WHEN D.type = 'No' THEN 1 ELSE 0 END) as No,
SUM(CASE WHEN D.type = '' THEN 1 ELSE 0 END) as Other,
SUM(CASE WHEN S.mobile IS NULL THEN 0 ELSE 1 END) as Sales,
COUNT(*) as TOTAL
FROM dairy as D
LEFT JOIN (SELECT DISTINCT mobile FROM sales) as S ON D.MobileNo = S.mobile
WHERE source = 'Company' AND UNIX_TIMESTAMP(CheckDate) >= 1293840000 AND UNIX_TIMESTAMP(CheckDate) <= 1322697600
group by D.Username order by TOTAL DESC