I have a table that has user_seq_id and no of days a user was active in the program. I want to understand the drop-off funnel. Like how many users were active on day 0 (100%) and on day 1, 2 and so on.
Input table :
create table test (
user_seq_id int ,
NoOfDaysUserWasActive int
);
insert into test (user_seq_id , NoOfDaysUserWasActive)
values (13451, 2), (76453, 1), (22342, 3), (11654, 0),
(54659, 2), (64420, 1), (48906, 5);
I want Day, ActiveUsers, and % Distribution of these users.
One method doesn't use window functions at all. Just a list of days and aggregation:
select v.day, count(t.user_seq_id),
count(t.user_seq_id) / c.cnt as ratio
from (select 0 as day union all select 1 union all select 2 union all select 3 union all select 4 union all select 5
) v(day) left join
test t
on v.day <= t.NoOfDaysUserWasActive cross join
(select count(*) as cnt from test) c
group by v.day, c.cnt
order by v.day asc;
Here is a db<>fiddle.
The mention of window function suggests that you are thinking:
select NoOfDaysUserWasActive,
sum(count(*)) over (order by NoOfDaysUserWasActive desc) as cnt,
sum(count(*)) over (order by NoOfDaysUserWasActive desc) / sum(count(*)) over () as ratio
from test
group by NoOfDaysUserWasActive
order by NoOfDaysUserWasActive
The problem is that this does not "fill in" the days that are not explicitly in the original data. If that is not an issue, then this should have better performance.
I'm trying to create a query using SELECT CASE that would return the accumulative results.
Here is my query, it works but doesnt return the correct result.
SELECT total, count(*) as count FROM
(
SELECT case
WHEN ( X ) < 1 THEN '1 km'
WHEN ( X ) < 3 THEN '3 km'
WHEN ( X ) < 5 THEN '5 km'
WHEN ( X ) < 10 THEN '10 km'
WHEN ( X ) < 15 THEN '15 km'
WHEN ( X ) < 25 THEN '25 km'
WHEN ( X ) < 50 THEN '50 km'
WHEN ( X ) < 100 THEN '100 km'
WHEN ( X ) > 0 THEN '-1'
else '-2'
end AS `total`
FROM `store` AS d WHERE d.pending!='1'
) AS someRandomAliasHere
GROUP BY `total`
X is a formula i'm using to calculate radius from a lat and lang. total is NOT a column in my database table, just a result to calculations of X
The query above gives me this..
1 km (4)
3 km (19)
5 km (103)
25 km (540)
50 km (61)
....
4,19,103,540,62 are the total matches found.
The total count for 3 should be 19+4=23.
5 should be 103+19+4=122 etc. And WHEN ( X ) > 0 THEN '-1' should show the total count. of matches
I tried using BETWEEN 0 AND 1, BETWEEN 0 AND 3 etc but it still didn't give me the correct results
Any ideas?
Another approach is to calculate the results independently then union them:
SELECT 1 AS total, COUNT(*) AS cnt FROM `store` WHERE store.pending != 1 AND ( X ) < 1
UNION ALL
SELECT 3 AS total, COUNT(*) AS cnt FROM `store` WHERE store.pending != 1 AND ( X ) < 3
UNION ALL
SELECT 5 AS total, COUNT(*) AS cnt FROM `store` WHERE store.pending != 1 AND ( X ) < 5
UNION ALL
/** ... **/
SELECT 100 AS total, COUNT(*) AS cnt FROM `store` WHERE store.pending != 1 AND ( X ) < 100
In addition to the accumulation, you also want a total value at the end with -1. This is a bit of a pain, but it can be accomplished.
The simplest way to do cumulative sums in MySQL is using variables. The basic idea is this:
SELECT total, cnt, (#cnt := #cnt + cnt) as count
FROM (SELECT (case WHEN ( X ) < 1 THEN '1'
WHEN ( X ) < 3 THEN '3'
WHEN ( X ) < 5 THEN '5'
WHEN ( X ) < 10 THEN '10'
WHEN ( X ) < 15 THEN '15'
WHEN ( X ) < 25 THEN '25'
WHEN ( X ) < 50 THEN '50'
WHEN ( X ) < 100 THEN '100'
WHEN ( X ) > 0 THEN '-1'
else '-2'
end) AS total, COUNT(*) as cnt
FROM store s
WHERE s.pending <> '1'
GROUP BY total
) t CROSS JOIN
(SELECT #cnt := 0) vars
ORDER BY total;
The issue with this is that you will not get an overall total of the non-negative values. Let me assume that you have no negative values. This requires adding another row into the total line:
SELECT total, cnt, (#cnt := #cnt + cnt) as count
FROM (SELECT (case WHEN ( X ) < 1 THEN '1'
WHEN ( X ) < 3 THEN '3'
WHEN ( X ) < 5 THEN '5'
WHEN ( X ) < 10 THEN '10'
WHEN ( X ) < 15 THEN '15'
WHEN ( X ) < 25 THEN '25'
WHEN ( X ) < 50 THEN '50'
WHEN ( X ) < 100 THEN '100'
WHEN ( X ) > 0 THEN '-1'
else '-2'
end) AS total, COUNT(*) as cnt
FROM store s
WHERE s.pending <> '1'
GROUP BY `total`
UNION ALL
SELECT -1, 0
) t CROSS JOIN
(SELECT #cnt := 0) vars
ORDER BY (total >= 0) desc, total;
I've changed the order by as well. Note that the value -2 is probably meaningless, because X < 1 and X > 0 cover all possible values of X (except for NULL). If you actually have values 100 or greater, there are some small changes to refine the query. You do not describe what to do with those values, so clarification on the question would be helpful.
Not sure if this works since I don't have a database to test this on. Also not exactly in the format you want.
select sum(if(X<1,1,0)) as C1,
sum(if(X<3,1,0)) as C3,
sum(if(X<5,1,0)) as C5,
sum(if(X<10,1,0)) as C10,
sum(if(X<15,1,0)) as C15,
sum(if(X<25,1,0)) as C25,
sum(if(X<50,1,0)) as C50,
sum(if(X<100,1,0)) as C100,
sum(if(X>=100,1,0)) as C100P
from store
where store.pending != '1'
Unfortunatelly MySQL do not have analytical functions and windowed functions, but in this case, you can achieve your goal using a variable and a nested subquery:
SELECT
total,
cnt,
#rollupCount:=#rollupCount+cnt AS rollupCount
FROM
(
SELECT
total,
count(*) AS cnt
FROM
(
SELECT
CASE
WHEN ( X ) < 1 THEN '1'
WHEN ( X ) < 3 THEN '3'
WHEN ( X ) < 5 THEN '5'
WHEN ( X ) < 10 THEN '10'
WHEN ( X ) < 15 THEN '15'
WHEN ( X ) < 25 THEN '25'
WHEN ( X ) < 50 THEN '50'
WHEN ( X ) < 100 THEN '100'
WHEN ( X ) > 0 THEN '-1'
ELSE '-2'
END AS `total`
FROM
`store` AS d
WHERE
d.pending != '1'
) AS someRandomAliasHere
GROUP BY
`total`
) AS anotherRandomAliasHere
, (SELECT #rollupCount:=0) AS RC
ORDER BY
total ASC
This is the same as when you want to calculate the row number for each record:
SELECT
#rowNumber:=#rowNumber+1 AS rowNumber,
sourceColumns
FROM
sourceTable, (SELECT #rowNumber:=0) AS t
ORDER BY
orderColumn;
A column to give cumulative totals from an existing query can be created relatively simply:
SELECT X,
total,
(SELECT SUM(total)
FROM (<<<your_current_query>>>) ycq2
WHERE ycq2.X <= ycq1.X) `cumulative_total`
FROM (<<<your_current_query>>>) ycq1
Of course this will expand quite a lot when pasting your current query in the two marked places.
See SQL fiddle demo.
try this:
/sql server version/
DECLARE #GROUPS TABLE (TOTAL INT)
INSERT INTO #GROUPS
VALUES (1),
(3),
(5),
(10),
(15),
(25),
(50),
(100),
(-1)
SELECT a.TOTAL, z.[COUNT] FROM #GROUPS a
CROSS APPLY
(SELECT COUNT(*) as [COUNT] FROM store x WHERE CASE WHEN a.TOTAL = -1 THEN -x.X ELSE x.X END < REPLACE(a.TOTAL,-1,0)
AND pending != 1) z
/mysql version/
CREATE TEMPORARY TABLE GROUPS TABLE (TOTAL INT)
INSERT INTO GROUPS
VALUES (1),
(3),
(5),
(10),
(15),
(25),
(50),
(100),
(-1)
SELECT a.TOTAL, z.[COUNT] FROM GROUPS a
CROSS APPLY
(SELECT COUNT(*) as [COUNT] FROM store x WHERE CASE WHEN a.TOTAL = -1 THEN -x.X ELSE x.X END < REPLACE(a.TOTAL,-1,0)
AND pending != 1) z
try this one :
1.) value after "THEN" must be numeric
2.) make temporary table with running id (example for SQL SERVER)
SELECT identity(int, 1, 1) as id,
case
WHEN ( X ) < 1 THEN 1
WHEN ( X ) < 3 THEN 3
WHEN ( X ) < 5 THEN 5
WHEN ( X ) < 10 THEN 10
WHEN ( X ) < 15 THEN 15
WHEN ( X ) < 25 THEN 25
WHEN ( X ) < 50 THEN 50
WHEN ( X ) < 100 THEN 100
WHEN ( X ) > 0 THEN -1
else '-2' end AS total
Into storeID
FROM store AS d
WHERE d.pending!='1'
Order BY total
3.) join table with same table, with criteria
Select a.*, sum(b.total) as NewTotal
From storeID a
Left Join storeID b
On b.id <= a.id
Group By a.id, a.total
4.) i think, "NewTotal" is what you are looking for
Your question is a bit hard to understand. Is this what you want?
CREATE TABLE Totals (X INT, SUM INT);
INSERT INTO Totals VALUES
(1, 4),
(3, 19),
(5, 103),
(25, 540),
(50, 61)
SELECT first.X
, Sum(second.SUM)
FROM Totals first
JOIN Totals second
ON first.x >= second.x
GROUP BY first.X
UNION
SELECT 0, SUM(sum) * 2
FROM Totals
http://sqlfiddle.com/#!3/65665/12
I suppose that you X function returns a floating point number. If I understand your logic correctly, you want to group together values where X is >=0 and <1, where X >=0 and <3, >=0 and <5 and so on, and you want to return -1 when the value is >=0, and -2 when the value is a negative <0 number.
I would use an intervals table, defined like this:
CREATE TABLE intervals (
i_begin INT,
i_end INT,
i_value INT
);
INSERT INTO intervals VALUES
(0, 1, 1),
(0, 3, 3),
(0, 5, 5),
(0, 10, 10),
(0, 15, 15),
(0, 25, 25),
(0, 50, 50),
(0, 100, 100),
(0, null, -1),
(null, 0, -2);
or you can play with values in this table to make it suite your needs.
then you can just use an INNER JOIN and a GROUP BY query:
SELECT
i_value, COUNT(*)
FROM
store INNER JOIN intervals
ON ((i_begin IS NULL OR X>=i_begin) AND (i_end IS NULL OR X<i_end))
WHERE
store.pending<>1
GROUP BY
i_value
Please see an example here.
Here is sqlfiddle that i made with mysql query
http://sqlfiddle.com/#!2/f2794/4
It count 10 consecutive days when present = 0, but i need to add second condition to count where present is > 10.
For example
11
22
0
0
0
0
0
0
0
0
0
0
0
0
1
should count 14
here is that query
select sum(count) total from (
SELECT COUNT(present) as count FROM (
SELECT
IF((q.present != 0), #rownum:=#rownum+1, #rownum:=#rownum) AS rownumber, #prevDate:=q.date, q.*
FROM (
SELECT
name
, date
, present
FROM
teacher, (SELECT #rownum:=0, #prevDate:='') vars
WHERE date BETWEEN '2013-07-01' AND '2013-07-31'
ORDER BY date, present
) q
) sq
GROUP BY present, rownumber
HAVING COUNT(*) >= 10
) d
So if U can help me, pls do it :)
best regards
m.
I dont really understand your query overly well, but I think simply changing (q.Present != 0) to incorporate the additional test should solve your problem:
SELECT sum(count) total from (
SELECT COUNT(present) as count FROM (
SELECT
IF((q.present != 0 AND q.present <= 10), #rownum:=#rownum+1, #rownum:=#rownum) AS rownumber, #prevDate:=q.date, q.*
FROM (
SELECT
name
, date
, present
FROM
teacher, (SELECT #rownum:=0, #prevDate:='') vars
WHERE date BETWEEN '2013-07-01' AND '2013-07-31'
ORDER BY date, present
) q
) sq
GROUP BY present, rownumber
HAVING COUNT(*) >= 10
) d
I have this query:
SELECT COUNT(*) AS invoice_count, IFNULL(SUM(qa_invoices.invoice_total), 0)
AS invoice_total, IFNULL(SUM(qa_invoices.invoice_discount) ,0) AS invoice_discount
FROM qa_invoices
WHERE (DATE(qa_invoices.invoice_date) BETWEEN '12/06/25' AND '12/06/25')
AND qa_invoices.status_code IN (5, 8)
UNION
SELECT IFNULL(SUM(qa_returns.client_credit), 0)
FROM qa_returns
WHERE (DATE(qa_returns.returnlog_date) BETWEEN '12/06/25' AND '12/06/25');
I get the error:
The used SELECT statements have a different number of columns.
I'm trying to join this 2 selects with an UNION command, if we look returnlog_date and invoice_date have the same data condition, if there is any way to perform both queries into one would be better.
Use a subselect:
SELECT
COUNT(*) AS invoice_count,
IFNULL(SUM(invoice_total), 0) AS invoice_total,
IFNULL(SUM(invoice_discount), 0) AS invoice_discount,
(
SELECT IFNULL(SUM(qa_returns.client_credit), 0)
FROM qa_returns
WHERE qa_returns.returnlog_date >= '2012-06-25'
AND qa_returns.returnlog_date < '2012-06-26'
) AS client_credit
FROM qa_invoices
WHERE invoice_date >= '2012-06-25'
AND invoice_date < '2012-06-26'
AND status_code IN (5, 8)
The error is telling you exactly what the problem is, for a UNION you have to have the same number of columns in each query.
I am not sure which column in your second query corresponds to your first query, but you can insert a zero in your second query.
Something like this:
SELECT COUNT(*) AS invoice_count
, IFNULL(SUM(qa_invoices.invoice_total), 0) AS invoice_total
, IFNULL(SUM(qa_invoices.invoice_discount) ,0) AS invoice_discount
FROM qa_invoices
WHERE (DATE(qa_invoices.invoice_date) BETWEEN '12/06/25' AND '12/06/25')
AND qa_invoices.status_code IN (5, 8)
UNION
SELECT 0
, IFNULL(SUM(qa_returns.client_credit), 0)
, 0
FROM qa_returns
WHERE (DATE(qa_returns.returnlog_date) BETWEEN '12/06/25' AND '12/06/25');
Result set you union together have to have the exact same columns.
Well in order to do a UNION u need to have same number of columns
I'm trying to calculate a percentage in my SQL query.
This is what I have right now:
SELECT
DATE(processed_at) AS day,
(
SELECT
COUNT(*) FROM return_items
WHERE return_id IN (SELECT id FROM returns WHERE DATE(processed_at) = day)
) as products_returned,
COUNT(*) as return_count,
(
SELECT
COUNT(*) as co_returns
FROM returns
WHERE return_method = 'mondial_relais'
AND DATE(processed_at) = day
) as return_rate_mr
FROM returns
WHERE MONTH(processed_at) = 10
AND YEAR(processed_at) = 2011
GROUP BY day;
Basically I need the return_rate_mr to be a percentage value.
I tried doing something like return_rate_mr * 100 / return_count as perc_value but this doesn't work. (I don't actually need the current return_rate_mr value, just the percentage.
Any ideas?
Assuming your original query returns the desired results, you can wrap it as a subquery:
SELECT
day,
return_rate_mr * 100 / return_count as perc_value,
... any other columns ...
FROM
( ... your original query here ...) as myalias;
Basically, the subquery creates a result set where the columns are renamed. Then, the outer query is free to use those new column names.
Are you looking for this?
SELECT
`day`,
`products_returned`,
(`return_rate_mr` * 100) / `return_count` AS `percentage`
FROM (
SELECT
DATE(processed_at) AS day,
(
SELECT
COUNT(*) FROM return_items
WHERE return_id IN (SELECT id FROM returns WHERE DATE(processed_at) = day)
) as products_returned,
COUNT(*) as return_count,
(
SELECT
COUNT(*) as co_returns
FROM returns
WHERE return_method = 'mondial_relais'
AND DATE(processed_at) = day
) as return_rate_mr
FROM returns
WHERE MONTH(processed_at) = 10
AND YEAR(processed_at) = 2011
GROUP BY day) AS `ss`
Did you try something like:
SELECT (`return_rate_mr` * 100 ) / `return_count` as "yourValue", OthersFields
FROM SELECT
DATE(processed_at) AS day,
(
SELECT
COUNT(*) FROM return_items
WHERE return_id IN (SELECT id FROM returns WHERE DATE(processed_at) = day)
) as products_returned,
COUNT(*) as return_count,
(
SELECT
COUNT(*) as co_returns
FROM returns
WHERE return_method = 'mondial_relais'
AND DATE(processed_at) = day
) as return_rate_mr
FROM returns
WHERE MONTH(processed_at) = 10
AND YEAR(processed_at) = 2011
GROUP BY day;
Hope this helps
Try:
SELECT DATE(processed_at) AS day,
count(distinct id) as products_returned,
COUNT(*) as return_count,
100* sum(case return_method when 'mondial_relais' then 1 end) / COUNT(*)
as return_perc_mr
FROM returns
WHERE MONTH(processed_at) = 10
AND YEAR(processed_at) = 2011
GROUP BY day;
I suspect that products_returned should be counting distinct item_id values (or something similar), but this should duplicate the logic in the original query.