I have a POST table, a CATEGORY table, a ACTION table and ACTION_TYPE table, I explain the ACTION table contains all the actions that were made, and the table ACTION_TYPE contains the actions details for example the ACTION whose ID = 4 has ACTION_TYPE_ID = 1 for POST_ID 6, which mean an action was made for post number 50, we can have many actions for one post_id
The POST table
id title content category_id
---------- ---------- ---------- ------------
1 title1 Text... 1
2 title2 Text... 1
3 title3 Text... 1
4 title4 Text... 3
5 title5 Text... 2
6 title6 Text... 1
The CATEGORY table
id name
---------- ----------
1 category_1
2 category_2
3 category_3
The ACTION_TYPE table
id name
---------- ----------
1 updated
2 deleted
3 restored
4 hided
The ACTION table
id post_id action_type_id date
---------- ---------- -------------- -----
1 1 1 2017-01-01
2 1 1 2017-02-15
3 1 3 2018-06-10
4 6 1 2019-08-01
5 5 2 2019-12-09
6 2 3 2020-04-27
7 2 1 2020-07-29
8 3 2 2021-03-13
Now I explain the case, I actually have two queries a query to count the posts for each category and another to count the actions performed on each post by category which work perfectly.
Here is my first query
select categories, count(*) as cnt_posts_per_cat
from(
select
case
when p.category_id is not null then c.name
end as categories
from post p
left join category c on p.category _id = c.id
)
group by categories
;
Which brings this result
categories cnt_posts_per_cat
---------- -------------------
category_1 4
category_2 1
category_3 1
Ans here is my second query
select categories, count(*) as cnt_actions_per_cat
from(
select distinct ac.post_id AS action_post_id, max(ac.date) over (partition by ac.post_id) as max_date,
case
when ac.action_type_id is not null then act.name
end as actions,
case
when p.category_id is not null then c.name
else 'na'
end as categories
from action ac
left join post p on ac.post_id = p.id
left join category c on p.category _id = c.id
left join action_type act on ac.action_type_id = act.id
where act.name in ('restored','deleted','updated')
)
group by categories
;
Which brings this correct result because it's seclect the last action for each action_type
categories cnt_actions_per_cat
---------- -------------------
category_1 3
category_2 1
category_3 na
But I would like to have a single result table for both queries at the same time as follow :
Here the result expected to be
categories cnt_posts_per_cat cnt_actions_per_cat
---------- ----------------- -------------------
category_1 4 3
category_2 1 1
category_3 1 na
i was trying union and union all but it's not correct it return following result
categories cnt_posts_per_cat
---------- -----------------
category_1 7
category_2 2
category_3 1
Best regards
How about correlated subqueries?
Sample data:
SQL> with
2 post (id, category_id) as
3 (select 1, 1 from dual union all
4 select 2, 1 from dual union all
5 select 3, 1 from dual union all
6 select 4, 3 from dual union all
7 select 5, 2 from dual union all
8 select 6, 1 from dual
9 ),
10 category (id, name) as
11 (select 1, 'category_1' from dual union all
12 select 2, 'category_2' from dual union all
13 select 3, 'category_3' from dual
14 ),
15 action_type (id, name) as
16 (select 1, 'updated' from dual union all
17 select 2, 'deleted' from dual union all
18 select 3, 'restored' from dual union all
19 select 4, 'hided' from dual
20 ),
21 action (id, post_id, action_type_id) as
22 (select 1, 1, 1 from dual union all
23 select 2, 1, 1 from dual union all
24 select 3, 1, 3 from dual union all
25 select 4, 6, 1 from dual union all
26 select 5, 5, 2 from dual union all
27 select 6, 2, 3 from dual union all
28 select 7, 2, 1 from dual union all
29 select 8, 3, 2 from dual
30 )
Query begins here:
31 select c.name,
32 --
33 (select count(*)
34 from post p
35 where p.category_id = c.id
36 ) cnt_posts_per_cat,
37 --
38 (select count(*)
39 from action a join post p on p.id = a.post_id
40 join action_type t on t.id = a.id
41 where p.category_id = c.id
42 and t.name in ('restored', 'deleted', 'updated')
43 ) cnt_actions_per_cat
44 from category c
45 order by c.name;
NAME CNT_POSTS_PER_CAT CNT_ACTIONS_PER_CAT
---------- ----------------- -------------------
category_1 4 3
category_2 1 0
category_3 1 0
SQL>
Related
I'm trying to get the unique number of invoices a company has received and sent out using 2 count() functions. In invoices table there are two columns that are references to the same company id (one is id of a company that is sending an invoice and the other one is id of a company that is receiving an invoice)
This is the code I tried using:
SELECT K.ID,K.NAME,K.CITY, COUNT(*) AS NUM_OF_INVOICES_SENT, COUNT(*) AS NUM_OF_INVOICES_RECEIVED
FROM COMPANY K LEFT JOIN INVOICE F ON F.COMP_SNEDING = K.ID
GROUP BY K.NAME,K.ID,K.CITY
This is for a school project so I am in no means well versed in sql/oracle
actual data invoices:
actual data company:
desired outcome with given actual data:
Here's one option; it doesn't use count, but sum with case expression.
Sample data:
SQL> with
2 invoice (id, amount, comp_sending, comp_receiving) as
3 (select 1, 2000 , 1, 2 from dual union all
4 select 2, 28250, 3, 2 from dual union all
5 select 3, 8700 , 4, 1 from dual union all
6 select 4, 20200, 5, 3 from dual union all
7 select 5, 21500, 3, 4 from dual
8 ),
9 company (id, name, city, state) as
10 (select 1, 'Microsoft', 'Redmond' , 'Washington' from dual union all
11 select 2, 'Ubisoft' , 'Paris' , 'France' from dual union all
12 select 4, 'Starbucks', 'Seattle' , 'Washington' from dual union all
13 select 5, 'Apple' , 'Cupertino', 'California' from dual union all
14 select 3, 'Nvidia' , 'Cupertino', 'California' from dual
15 )
Query begins here:
16 select c.id, c.name,
17 sum(case when c.id = i.comp_sending then 1 else 0 end) cnt_sent,
18 sum(case when c.id = i.comp_receiving then 1 else 0 end) cnt_received
19 from company c left join invoice i on c.id in (i.comp_sending, i.comp_receiving)
20 group by c.id, c.name
21 order by c.id;
ID NAME CNT_SENT CNT_RECEIVED
---------- --------- ---------- ------------
1 Microsoft 1 1
2 Ubisoft 0 2
3 Nvidia 2 1
4 Starbucks 1 1
5 Apple 1 0
SQL>
You can use COUNT if you replace the 0 in the CASE expressions with NULL. So #Littlefoot's query becomes
select c.id, c.name,
COUNT(case when c.id = i.comp_sending then 1 else NULL end) cnt_sent,
COUNT(case when c.id = i.comp_receiving then 1 else NULL end) cnt_received
from company c left join invoice i on c.id in (i.comp_sending, i.comp_receiving)
group by c.id, c.name
order by c.id;
This works because COUNT counts only those rows which have a non-NULL value in the expression which is being counted.
db<>fiddle here
In image I have given table structure and sample data and I need output result as mentioned
With sample data you provided (lines #1 - 8), this returns desired result. Will it work for all other cases, I have no idea as the question lacks in quite a lot of information so YMMV.
SQL> with employee (id, name, type, visit_date) as
2 (select 1, 'Mohan', '01', date '2010-09-09' from dual union all
3 select 1, 'Mohan', '02', date '2010-09-10' from dual union all
4 --
5 select 1, 'Gani' , '01', date '2010-09-01' from dual union all
6 select 1, 'Gani' , '01', date '2010-09-02' from dual union all
7 select 1, 'Gani' , '01', date '2010-09-03' from dual
8 ),
9 --
10 type1 as
11 (select id, name, visit_date
12 from employee
13 where type = '01'
14 ),
15 type2 as
16 (select id, name, visit_date
17 from employee
18 where type = '02'
19 )
20 select
21 a.id,
22 a.name,
23 a.visit_date type1date,
24 b.visit_date type2date
25 from type1 a left join type2 b on a.id = b.id and a.name = b.name
26 order by a.id, a.name desc, a.visit_date;
ID NAME TYPE1DATE TYPE2DATE
---------- ----- ---------- ----------
1 Mohan 09/09/2010 10/09/2010
1 Gani 01/09/2010
1 Gani 02/09/2010
1 Gani 03/09/2010
SQL>
table "team1" :
id country
1 India
2 Pakistan
3 srilanka
4 England
table "team2" :
id name name2
1 2 4
2 1 3
i have to combine two tables
another table when retrieve the data that time in place of 2 , 4 Pakistan,England
It is about the self join of team1 table (lines #15 and 16):
SQL> with
2 team1 (id, country) as
3 (select 1, 'India' from dual union all
4 select 2, 'Pakistan' from dual union all
5 select 3, 'Sri Lanka' from dual union all
6 select 4, 'England' from dual
7 ),
8 team2 (id, name, name2) as
9 (select 1, 2, 4 from dual union all
10 select 2, 1, 3 from dual
11 )
12 select b.id,
13 t1.country,
14 t2.country
15 from team2 b join team1 t1 on t1.id = b.name
16 join team1 t2 on t2.id = b.name2
17 order by b.id;
ID COUNTRY COUNTRY
---------- --------- ---------
1 Pakistan England
2 India Sri Lanka
SQL>
Just showing another way of writing the same query with aggregate functions and grouping.
with
team1 (id, country) as
(select 1, 'India' from dual union all
select 2, 'Pakistan' from dual union all
select 3, 'Sri Lanka' from dual union all
select 4, 'England' from dual
),
team2 (id, name, name2) as
(select 1, 2, 4 from dual union all
select 2, 1, 3 from dual
)
SELECT
T2.ID,
MAX(CASE
WHEN T2.NAME = T1.ID THEN T1.COUNTRY
END) AS TEAM1,
MAX(CASE
WHEN T2.NAME2 = T1.ID THEN T1.COUNTRY
END) AS TEAM2
FROM
TEAM2 T2
JOIN TEAM1 T1 ON T1.ID IN (
T2.NAME,
T2.NAME2
)
GROUP BY
T2.ID
ORDER BY
T2.ID;
Output:
ID TEAM1 TEAM2
---------- --------- ---------
1 Pakistan England
2 India Sri Lanka
Cheers!!
I would like to pivot a select statement.
Columns "Country", "Store" and "Sales" are given.
Now I would like to have an output like:
Store1 Store2 Store3
Country1 2342 2342 5675
Country2 5753 3274 7326
Country3 1543 4367 3367
So basically I need the salescount for every Store, for every Country.
The Input comes from (example):
Country: StoreNr: ProductSold:
Belgium 23 Car
Belgium 23 House
Netherland 23 Car
Output would be:
Store23
Belgium 2
Netherlands 1
If the number of stores is finite, you could use one of these approaches:
Using count() aggregate function combined with case expression:
-- sample of data. just for the sake of demonstration
SQL> with t1(Country, StoreNr, ProductSold) as(
2 select 'Belgium' , 23, 'Car' from dual union all
3 select 'Belgium' , 23, 'House' from dual union all
4 select 'Netherland', 23, 'Car' from dual union all
5 select 'Belgium' , 25, 'House' from dual
6 )
7 select country
8 , count(case
9 when StoreNr = 23
10 then 1
11 end) as storeNr_23
12 , count(case
13 when StoreNr = 25
14 then 1
15 end) as storeNr_25
16 from t1
17 group by country
18 ;
Result:
COUNTRY STORENR_23 STORENR_25
---------- ---------- ----------
Belgium 2 1
Netherland 1 0
Starting from Oracle 11g and up, the pivot operator as follows:
select *
from (Select country as country
, country as country_cnt
, StoreNr
from t1)
pivot( -- list all store numbers here
count(country_cnt) for storenr in ( 23 as StoreNr_23
, 25 as StoreNr_25)
)
Result:
COUNTRY STORENR_23 STORENR_25
---------- ---------- ----------
Belgium 2 1
Netherland 1 0
I am working on a time attendance system. I have the employees' transactions stored in the following table:
I want to get the earliest and the latest transactions for each employee including their date and type.
I am able to get the dates using grouping and aggregation. However, I am not able to figure out how to get types with them.
Would you please help me in it.
Thank you.
That's what the FIRST and LAST aggregate functions are designed for.
Here is a link to the documentation:
FIRST: http://download.oracle.com/docs/cd/E11882_01/server.112/e17118/functions065.htm#SQLRF00641
LAST: http://download.oracle.com/docs/cd/E11882_01/server.112/e17118/functions083.htm#sthref1206
And here is an example:
SQL> create table my_transactions (id,employee_id,action_date,type)
2 as
3 select 1, 1, sysdate, 'A' from dual union all
4 select 2, 1, sysdate-1, 'B' from dual union all
5 select 3, 1, sysdate-2, 'C' from dual union all
6 select 4, 1, sysdate-3, 'D' from dual union all
7 select 5, 2, sysdate-11, 'E' from dual union all
8 select 6, 2, sysdate-12, 'F' from dual union all
9 select 7, 2, sysdate-13, 'G' from dual
10 /
Table created.
SQL> select *
2 from my_transactions
3 order by id
4 /
ID EMPLOYEE_ID ACTION_DATE T
---------- ----------- ------------------- -
1 1 04-07-2011 10:15:07 A
2 1 03-07-2011 10:15:07 B
3 1 02-07-2011 10:15:07 C
4 1 01-07-2011 10:15:07 D
5 2 23-06-2011 10:15:07 E
6 2 22-06-2011 10:15:07 F
7 2 21-06-2011 10:15:07 G
7 rows selected.
SQL> select employee_id
2 , min(action_date) min_date
3 , max(type) keep (dense_rank first order by action_date) min_date_type
4 , max(action_date) max_date
5 , max(type) keep (dense_rank last order by action_date) max_date_type
6 from my_transactions
7 group by employee_id
8 /
EMPLOYEE_ID MIN_DATE M MAX_DATE M
----------- ------------------- - ------------------- -
1 01-07-2011 10:15:07 D 04-07-2011 10:15:07 A
2 21-06-2011 10:15:07 G 23-06-2011 10:15:07 E
2 rows selected.
Regards,
Rob.
You could try to use analytical(or windowing functions)
select *
from
(select id, employee_id, action_date,type,
max(action_date) over (partition by employee_id) max_action_date,
min(action_date) over (partition by employee_id) min_action_date
from transaction)
where action_date in (max_action_date, min_action_date)