I have the following database structure:
FieldID|Year|Value
a|2011|sugar
a|2012|salt
a|2013|pepper
b|2011|pepper
b|2012|pepper
b|2013|pepper
c|2011|sugar
c|2012|salt
c|2013|salt
now I would like to run a query that counts the number of fields for every item in the particular year looking something like this:
value|2011|2012|2013
sugar|2|0|0
salt |0|2|1
pepper|1|1|2
I used multiple tables for every year before. However the distinct values for 2011,2012 and 2013 might be different (e.g. sugar would only be present in 2011)
For individual years I used:
SELECT `Value`, COUNT( `FieldID` ) FROM `Table` WHERE `Year`=2011 GROUP BY `Value`
A1ex07's answer is fine. However, in MySQL, I prefer this formulation:
SELECT Value,
sum(`Year` = 2011) AS cnt2011,
sum(`Year` = 2012) AS cnt2012,
sum(`Year` = 2013) AS cnt2013
FROM t
GROUP BY value;
The use of count( . . . ) produces the correct answer, but only because the else clause is missing. The default value is NULL and that doesn't get counted. To me, this is a construct that is prone to error.
If you want the above in standard SQL, I go for:
SELECT Value,
sum(case when `Year` = 2011 then 1 else 0 end) AS cnt2011,
sum(case when `Year` = 2012 then 1 else 0 end) AS cnt2012,
sum(case when `Year` = 2013 then 1 else 0 end) AS cnt2013
FROM t
GROUP BY value;
You can do pivoting :
SELECT `Value`,
COUNT(CASE WHEN `Year` = 2011 THEN FieldID END) AS cnt2011,
COUNT(CASE WHEN `Year` = 2012 THEN FieldID END) AS cnt2012,
COUNT(CASE WHEN `Year` = 2013 THEN FieldID END) AS cnt2013
FROM `Table`
GROUP BY `Value`
It is called Pivot Table, achieve with a chain of CASE statements which apply a 1 or 0 for each condition, then SUM() up the ones and zeros to retrieve a count.
SELECT
Value,
SUM(CASE WHEN Year = 2011 THEN 1 ELSE 0 END) AS 2012,
SUM(CASE WHEN Year = 2012 THEN 1 ELSE 0 END) AS 2012,
SUM(CASE WHEN Year = 2013 THEN 1 ELSE 0 END) AS 2013
FROM Table
GROUP BY Value
Related
SELECT
COUNT(CASE WHEN VALUE = 1 THEN 1 END) AS score_1,
COUNT(CASE WHEN VALUE = 2 THEN 1 END) AS score_2,
COUNT(CASE WHEN VALUE = 3 THEN 1 END) AS score_3,
COUNT(CASE WHEN VALUE = 4 THEN 1 END) AS score_4,
COUNT(CASE WHEN VALUE = 5 THEN 1 END) AS score_5,
COUNT(CASE WHEN VALUE = 6 THEN 1 END) AS score_6,
COUNT(CASE WHEN VALUE = 7 THEN 1 END) AS score_7,
COUNT(CASE WHEN VALUE = 8 THEN 1 END) AS score_8,
COUNT(CASE WHEN VALUE = 9 THEN 1 END) AS score_9,
COUNT(CASE WHEN VALUE = 10 THEN 1 END) AS score_10
FROM
`answers`
WHERE
`created_at` BETWEEN '2017-01-01 00:00:00' AND '2019-11-30 23:59:59'
Is there a way to optimize this query, because I have 4 million answer records in my DB, and it runs very slowly?
Try running this one time to create an index:
CREATE INDEX ix_ca on answers(created_at)
That should speed your query up. If you are curious about why, see here:
What is an index in SQL?
You could try add a redundant composite index
create idx1 on table answers(created_at, value)
using redudance in index the query should be result without accessing to table data just using the index content
Want it to be 10 times as fast? Use the Data Warehousing technique of buiding and maintaining a "Summary table". In this example the summary table might be
CREATE TABLE subtotals (
dy DATE NOT NULL,
`value` ... NOT NULL, -- TINYINT UNSIGNED ?
ct SMALLINT UNSIGNED NOT NULL, -- this is 2 bytes, max 65K; change if might be bigger
PRIMARY KEY(value, dy) -- or perhaps the opposite order
) ENGINE=InnoDB
Each night you summarize the day's data and build 10 new rows in subtotals.
Then the "report" query becomes
SELECT
SUM(CASE WHEN VALUE = 1 THEN ct END) AS score_1,
SUM(CASE WHEN VALUE = 2 THEN ct END) AS score_2,
SUM(CASE WHEN VALUE = 3 THEN ct END) AS score_3,
SUM(CASE WHEN VALUE = 4 THEN ct END) AS score_4,
SUM(CASE WHEN VALUE = 5 THEN ct END) AS score_5,
SUM(CASE WHEN VALUE = 6 THEN ct END) AS score_6,
SUM(CASE WHEN VALUE = 7 THEN ct END) AS score_7,
SUM(CASE WHEN VALUE = 8 THEN ct END) AS score_8,
SUM(CASE WHEN VALUE = 9 THEN ct END) AS score_9,
SUM(CASE WHEN VALUE = 10 THEN ct END) AS score_10
FROM
`subtotals`
WHERE `created_at` >= '2017-01-01'
AND `created_at` < '2019-12-01'
Based on what you have provided, there will be about 10K rows in subtotals; that's a lot less to wade through than 4M rows. It might run more than 10 times as fast.
More discussion: http://mysql.rjweb.org/doc.php/summarytables
new to MySQL..so pls help me out with this basic code..
i have a query something like this...
select weekofyear(id_time),
(id),
#Tat1:=exp1,
#Tat2:=exp2,
#check1:=exp3,
#check2:=exp4,
(case when #check2=0 then
(case when (#Tat1>(#Tat2+30) or (#check1=1 and (#Tat1>#Tat2+10))) then 1 else 0 end)
else
(case when (#Tat1>(#Tat2+30) or (#check1=1 and (#Tat1>#Tat2+20))) then 1 else 0 end)
end) as BO
from datb
where cid=18
and id_time between '2019-11-01 06:00:00' and '2019-11-25 06:00:00'
and it gives correct results as--here
however i want to use sum after case when statement so that I can get total values where BO=1 and group by week of year , so i made following changes-
select weekofyear(id_time),
count(id),
#Tat1:=exp1,
#Tat2:=exp2,
#check1:=exp3,
#check2:=exp4,
sum(case when #check2=0 then
(case when (#Tat1>(#Tat2+30) or (#check1=1 and (#Tat1>#Tat2+10))) then 1 else 0 end)
else
(case when (#Tat1>(#Tat2+30) or (#check1=1 and (#Tat1>#Tat2+20))) then 1 else 0 end)
end) as BO
from datb
where cid=18
and id_time between '2019-11-01 06:00:00' and '2019-11-25 06:00:00'
group by weekofyear(id_time)
but it always returns 0 as output.
Output --here 2
Please help , I don't know what am I doing wrong here.
Thanx !
As others have already said, session variables can be unpredictable (especially when aggregation gets mixed in). That said, it doesn't look like you're using the session variables to carry over values from one row to the next (as is often done), but to just make aliases of sorts for calculations you don't want to repeat.
A better way to handle that is just through subqueries.
SELECT woy, id, Tat1, Tat2, check1, check2
, CASE
WHEN check2=0 THEN (
CASE
WHEN (Tat1>(Tat2+30) OR (check1=1 AND (Tat1>Tat2+10))) THEN 1
ELSE 0
END
)
ELSE (
CASE WHEN (Tat1>(Tat2+30) OR (check1=1 AND (Tat1>Tat2+20))) THEN 1
ELSE 0
END
)
END AS BO
FROM (
SELECT WEEKOFYEAR(id_time) AS woy
, id
, exp1 AS Tat1
, exp2 AS Tat2
, exp3 AS check1
, exp4 AS check2
FROM datb
WHERE cid=18
AND id_time BETWEEN '2019-11-01 06:00:00' AND '2019-11-25 06:00:00'
) AS subQ
;
You can then tweak the above query for aggregation, or use it as a subquery for an aggregating outer query.
I have 70 different types of accounts. And I am fetching the data as per the account type.
The query like this,
$mainData = "SELECT
count(*) AS totalRows,
sum(pay) as totalPay
sum(case when account_type = 1 then 1 else 0 end) AS account_1_Total,
sum(case when account_type = 1 then pay else 0 end) AS account_1_Pay,
sum(case when account_type = 2 then 1 else 0 end) AS account_2_Total,
sum(case when account_type = 2 then pay else 0 end) AS account_2_Pay,
{all_account_types_here}
FROM account_table";
In the end, those sum() are about more than 140.
So the question is, how many sum() is recommended in a single query?
Thanks!
EDITED:
The GROUP BY is the solution of it.
Locked. There are disputes about this question’s content being resolved at this time. It is not currently accepting new answers or interactions.
I am new to SQL and would like to know how to approach writing a query for this question.
Lets say we have these fields:
date_created date_unsubscribed subscriberid
How to write a SQL query that lists, by month, how many people subscribed to the list, unsubscribed from the list, and how many net subscribers there were (new subscribers minus unsubscribers).
All in a single query...
Here's one option using conditional aggregation and union all:
select month(dt),
count(case when subscribe = 1 then 1 end) subscribecount,
count(case when subscribe = -1 then 1 end) unsubscribecountt,
sum(subscribe) overallcount
from (
select date_created as dt, 1 as subscribe
from yourtable
union all
select date_unsubscribed, -1
from yourtable
where date_unsubscribed is not null
) t
group by month(dt)
The subquery creates a list of dates with a flag for subscribe or unsubscribe. Then you can use count with case to determine the appropriate number of subscribers/unsubscribers.
SQL Fiddle Demo
You could write a sum(case) (a sum with conditions) to aggregate - assuming the date_created column is never null. For instance:
ORACLE:
SELECT
TO_CHAR(DATE_CREATED,'MM-YYYY') CREATE_MONTH
,SUM(CASE WHEN date_unsubscribed is not null then 1 else 0 end) unsubscribed
,SUM(CASE WHEN date_unsubscribed is null then 1 else 0 end) subscribed
,COUNT(SUBSCRIBER_ID)
FROM
--YOURTABLENAME
--WHERE
--WHATEVER OTHER CONDITIONS YOU HAVE APPLY
GROUP BY TO_CHAR(DATE_CREATED,'MM-YYYY')
MYSQL:
SELECT
DATE_FORMAT(DATE_CREATED,'%m-%Y') CREATE_MONTH
,SUM(CASE WHEN date_unsubscribed is not null then 1 else 0 end) unsubscribed
,SUM(CASE WHEN date_unsubscribed is null then 1 else 0 end) subscribed
,COUNT(SUBSCRIBER_ID)
FROM
--YOURTABLENAME
--WHERE
--WHATEVER OTHER CONDITIONS YOU HAVE APPLY
GROUP BY DATE_FORMAT(DATE_CREATED,'%m-%Y')
Oracle solution
Here is a query using the PIVOT operator, which was created exactly for this kind of work, and ROLLUP to get the net number. This is just for illustration; I assume the year is a user or application input (bind variable :year, set to 2015 for the output), and I show the summary for January through June.
with
test_data ( date_created, date_unsubscribed, subscriber_id ) as (
select date '2015-05-10', null , 330053448 from dual union all
select date '2015-04-28', null , 330053457 from dual union all
select date '2015-05-10', null , 330053466 from dual union all
select date '2015-04-28', null , 220053475 from dual union all
select date '2015-04-28', date '2015-05-10', 330053484 from dual
),
prep ( type, val, mth ) as (
select 'Subscribed' , 1, extract(month from date_created) from test_data
where extract(year from date_created) = :year
union all
select 'Unsubscribed', -1, extract(month from date_unsubscribed) from test_data
where extract(year from date_unsubscribed) = :year
)
select nvl(type, 'Net Subscr') as description,
nvl(sum(jan), 0) as jan, nvl(sum(feb), 0) as feb, nvl(sum(mar), 0) as mar,
nvl(sum(apr), 0) as apr, nvl(sum(may), 0) as may, nvl(sum(jun), 0) as jun
from prep
pivot (
sum(val)
for mth in (1 as jan, 2 as feb, 3 as mar, 4 as apr, 5 as may, 6 as jun)
)
group by rollup(type)
order by case type when 'Subscribed' then 1 when 'Unsubscribed' then 2 else 3 end
;
DESCRIPTION JAN FEB MAR APR MAY JUN
------------ ---------- ---------- ---------- ---------- ---------- ----------
Subscribed 0 0 0 3 2 0
Unsubscribed 0 0 0 0 -1 0
Net Subscr 0 0 0 3 1 0
3 rows selected.
I am trying to get the count of females and males in the gender field of a table.
Is there a way to get the count of each in one query?
Something like:
select * from table count(where gender = 'm') as total_males, count(where gender = 'f') as total_females;
or will it require two queries?
select count(*) from table where gender = 'm';
select count(*) from table where gender = 'f';
This is basically a PIVOT. MySQL does not have a pivot so you can use an aggregate function with a CASE statement to perform this:
select
sum(case when gender = 'm' then 1 else 0 end) Total_Male,
sum(case when gender = 'f' then 1 else 0 end) Total_Female
from yourtable
See SQL Fiddle with Demo
Or using COUNT:
select
count(case when gender = 'm' then 1 else null end) Total_Male,
count(case when gender = 'f' then 1 else null end) Total_Female
from yourtable;
See SQL Fiddle with Demo
Something like this will work:
SELECT SUM(IF(t.gender='m',1,0)) AS total_males
, SUM(IF(t.gender='f',1,0)) AS total_females
FROM mytable t
The "trick" here is that we are using a conditional test to return either a 0 or a 1 for each row, and then adding up the 0's and 1's. To make this a little more clear, I am using the SUM aggregate function rather than COUNT, although COUNT could be used just as easily, though we'd need to return a NULL in place of the zero.
SELECT COUNT(IF(t.gender='m',1,NULL)) AS total_males
, COUNT(IF(t.gender='f',1,NULL)) AS total_females
FROM mytable t
Consider that the two expressions in the SELECT list of this query:
SELECT COUNT(1)
, SUM(1)
FROM mytable t
Will return the same value.
If you want to avoid the MySQL IF function, this can also be done using the ANSI SQL CASE expression:
SELECT SUM( CASE WHEN t.gender = 'm' THEN 1 ELSE 0 END )) AS total_males
, SUM( CASE WHEN t.gender = 'f' THEN 1 ELSE 0 END )) AS total_females
FROM mytable t
select sum(case when gender='m' then 1 else null end) as total_males, sum(case when gender='f' then 1 else null end) as total_females from ...
Should work just fine!
If your only issue is to avoid two queries, you can always write two queries as subselects of one query.
Select (select 1 from dual) as one, (select 2 from dual) as two from dual
This would work for your scenario, too.