I try to create a query who select the contacts information (table invoice_contacts), and the adresses (table invoice_adresses) associate to the contact which is the most used in the (table invoice_compta)
For exemple I have two contact :
Mike
John
Mike have 2 adresses :
Paris
London
Mike have 1 invoice with Paris, and 5 invoice with London, so I want the adresse of London associate to Mike.
I have try this query with an subquery which count all adresses associate to the contact for an adresses (with NB_ADRESSES), and select only the biggest (with order by NB_ADRESSES desc and limit 1), it's seem wells but I have an error when I do where ia2.ID_CONTACT = ic.ID_CONTACT ic.ID_CONTACT is not found.. (and I need to associate the contact to the subquery).
select ic.*,
ia.*
from invoice_contacts ic
left join invoice_adresses ia on ia.ID_CONTACT = ic.ID_CONTACT
and ia.ID_ADRESSE in (
select ia3.ID_ADRESSE
from (
select ia2.ID_ADRESSE,
count(*) as NB_ADRESSES
from invoice_adresses ia2
left join invoice_comptas ico on ico.ID_ADRESSE_CONTACT = ia2.ID_ADRESSE
where ia2.ID_CONTACT = ic.ID_CONTACT
group by ia2.ID_ADRESSE
order by NB_ADRESSES desc
limit 1
) as ia3
)
group by ic.ID_CONTACT
order by CONTACT_TITRE asc
I also have try with "exist" or "inner join" instead of "in" but I doesn't find good results, so the best way seems it to be with this query for me, but I don't found the solution.
I hope you will help me :)
Thanks
UPDATE :
So finally I have found an solution with this query :
select ic.*,
ia.*
from invoice_contacts ic
left join invoice_adresses ia on ia.ID_CONTACT = ic.ID_CONTACT
and ia.ID_ADRESSE = (
select ia3.ID_ADRESSE
from (
select ia2.*,
count(*) as NB_ADRESSES
from invoice_adresses ia2
left join invoice_comptas ico on ico.ID_ADRESSE_CONTACT = ia2.ID_ADRESSE
group by ia2.ID_ADRESSE
) as ia3
where ia3.ID_CONTACT = ic.ID_CONTACT
order by NB_ADRESSES desc
limit 1
)
group by ic.ID_CONTACT
order by CONTACT_TITRE asc
Thanks
Let me rephrase the problem as finding the most common contact/address combination for a given invoice.
I find it hard to follow your query and your table naming. But this is the idea:
select contact, address
from (select contact, address, count(*) as cnt,
row_number() over (partition by contact order by count(*) desc) as seqnum
from invoices
group by contact, address
) ca
where seqnum = 1;
The subquery is counting the number of times a given address (or city if you prefer) occur for each contact. The row_number() enumerates these, so the most common one has a value of "1". The outer query then chooses the most common value.
Related
I have to say who is the scientist who have been the most in mission. I tried this code but it wasn't successful:
select name
from scientist, mission
where mission.nums = chercheur.nums
having count(*) = (select max(count(numis)) from mission, scientist where
mission.nums = chercheur.nums
group by name)
I have done several modifications for this request but I only obtain errors (ora-0095 and ora-0096 if I remember correctly).
Also, I create my tables with:
CREATE TABLE Scientist
(NUMS NUMBER(8),
NAME VARCHAR2 (15),
CONSTRAINT CP_CHER PRIMARY KEY (NUMS));
CREATE TABLE MISSION
(NUMIS NUMBER(8),
Country VARCHAR2 (15),
NUMS NUMBER(8),
CONSTRAINT CP_MIS PRIMARY KEY (NUMIS),
CONSTRAINT CE_MIS FOREIGN KEY (NUMS) REFERENCES SCIENTIST (NUMC));
You could count the missions each scientist participated in, and wrap that query in a query with a window function that will rank them according to their participation:
SELECT name
FROM (SELECT name, RANK() OVER (PARTITION BY name ORDER BY cnt DESC) AS rk
FROM (SELECT name, COUNT(*) AS cnt
FROM scientist s
JOIN mission m ON s.nums = m.nums
GROUP BY name) t
) q
WHERE rk = 1
Step 0 : Format your code :-) It would make it much easier to visualize
Step 1 : Get the count of Numis by Nums in the Mission table. This will tell you how many missions were done by each Nums
This is done in the cte block cnt_by_nums
Next to get the name of the scientist by joining cnt_by_nums with scientist table.
After that you want to get only those scientists who have the cnt_by_missions as the max available value from cnt_by_num
with cnt_by_nums
as (select Nums,count(Numis) as cnt_missions
from mission
group by Nums
)
select a.Nums,max(b.Name) as name
from cnt_by_nums a
join scientist b
on a.Nums=b.Nums
group by a.Nums
having count(a.cnt_missions)=(select max(a1.cnt_missions) from cnt_by_nums a1)
I'd write a query like this:
SELECT NAME, COUNTER
FROM
(SELECT NAME, COUNT(*) AS COUNTER
FROM SCIENTIST S
LEFT JOIN MISSION M
ON S.NUMS=M.NUMS
GROUP BY NAME) NUM
INNER JOIN
(SELECT MAX(COUNTER) AS MAX_COUNTER FROM
(SELECT NAME, COUNT(*) AS COUNTER
FROM SCIENTIST S
LEFT JOIN MISSION M
ON S.NUMS=M.NUMS
GROUP BY NAME) C) MAX
ON NUM.COUNTER=MAX.MAX_COUNTER;
(it works on MYSQL, I hope it's the same in Oracle)
As you don't select the name of your scientist (only count their missions) you don't need to join those tables within the subquery. Grouping over the foreign key would be sufficient:
select count(numis) from mission group by nums
You column names are a bit weird but that's your choice ;-)
Selecting only the scientist with the most mission references could be achieved in two ways. One way would be your approach where you may get multiple scientists if they have the same max missions.
The first problem you have in your query is that you are checking an aggregation (HAVING COUNT(*) = ) without grouping. You are only grouping your subselect.
Second, you could not aggregate an aggregation (MAX(COUNT)) but you may select only the first row of that subselect ordered by it's size or select the max of it by subquerying the subquery.
Approach with only one line:
select s.name from scientist s, mission m
where m.nums = s.nums
group by name
having count(*) =
(select count(numis) from mission
group by nums
order by 1 desc
fetch first 1 row only)
Approach with double subquery:
select s.name from scientist s, mission m
where m.nums = s.nums
group by name having count(*) =
(select max(numis) from
(select count(numis) numis from mission group by nums)
)
Second approach would be doing the FETCH FIRST on yur final result but this would give you exactly 1 scientist even if there are multiple with the same max missions:
select s.name from scientist s, mission m
where m.nums = s.nums
group by name
order by count(*) desc
fetch first 1 row only
Doing a cartisian product is not state of the art but the optimizer would make it a "good join" with the given reference in the where clause.
Made these with IBM Db2 but should also work on Oracle.
If you want one row, then in Oracle 12+, you can do:
SELECT name, COUNT(*) AS cnt
FROM scientist s JOIN
mission m
ON s.nums = m.nums
GROUP BY name
ORDER BY COUNT(*) DESC
FETCH FIRST 1 ROW ONLY;
In earlier versions, you would generally use a subquery:
SELECT s.*
FROM (SELECT name, COUNT(*) AS cnt
FROM scientist s JOIN
mission m
ON s.nums = m.nums
GROUP BY name
ORDER BY COUNT(*) DESC
) sm
WHERE rownum = 1;
If you want ties, then generally window functions would be a simple solution:
SELECT s.*
FROM (SELECT name, COUNT(*) AS cnt,
RANK() OVER (ORDER BY COUNT(*) DESC) as seqnum
FROM scientist s JOIN
mission m
ON s.nums = m.nums
GROUP BY name
ORDER BY COUNT(*) DESC
) sm
WHERE seqnum = 1;
I've been trying to remove duplicates using HAVING count(*) > 1, group by, distinct and sub queries but can't get any of these to work..
SELECT UserID, BuildingNo
FROM Staff INNER JOIN TblBuildings ON Staff.StaffID =
TblBuildingsStaffID
GROUP BY TblStaff.User_Code, BuildingNo
What I get is..
StaffID1 | BuildingNo1
StaffID1 | BuildingNo2
StaffID2 | BuildingNo2
StaffID3 | BuildingNo1
StaffID3 | BuildingNo2
I'm trying to get it so it just displays staff with one building number (if they have two regardless of which it shows) like:
StaffID1 | BuildingNo1
StaffID2 | BuildingNo2
StaffID3 | BuildingNo1
It can't be too hard.. I've tried CTE's left joining the building to the staff table, these come up NULL for some reason when I try this
Any help would be great!
Don't group by BuildingNo, then you can use having to filter out the groups you want.
SELECT s.UserID, min(b.BuildingNo) as buildingno
FROM Staff s
JOIN TblBuildings ON s.StaffID = b.TblBuildingsStaffID
GROUP BY s.UserID
having count(distinct b.BuildingNo) = 1;
The min() aggregate is required because buildingno is not part of the group by clause. But as the having() clause only returns those with one building, it doesn't change anything.
If you want to display all staff members, and simply pick one (arbitrary) building, then simply leave out the having condition.
If you want to include staff members without a building you need a left join:
SELECT s.UserID, min(b.BuildingNo) as buildingno
FROM Staff s
LEFT JOIN TblBuildings b ON s.StaffID = t.TblBuildingsStaffID
GROUP BY t.UserID;
Use row partition keyword in your query to avoid duplicacy
WITH CTE AS( SELECT ROW_NUMBER() OVER(PARTITION BY UserID ORDER BY UserID ) AS 'Num',UserID, BuildingNo
FROM Staff INNER JOIN TblBuildings ON Staff.StaffID =
TblBuildingsStaffID
GROUP BY TblStaff.User_Code, BuildingNo)
SELECT * FROM CTE
WHERE Num =1
try this -
SELECT distinct UserID, BuildingNo
FROM Staff INNER JOIN TblBuildings ON Staff.StaffID =
TblBuildingsStaffID
This is my query:
SELECT f.name, COUNT(*) as num_books
from author f
JOIN book b on b.tittle = f.book
Group by f.name
Which gives me this table:
NAME NUM_BOOKS
-------------------------------------------------- ----------
Dyremann 2
Nam mann 1
Thomas 1
Asgeir 1
Tullemann 5
Plantemann 1
Beste forfatter 1
Fagmann 5
Lars 1
Hans 1
Svein Arne 1
How could I easly alter the query to only display the author with the highest amount of released books? (While keeping in mind I'm rather new to sql)
Oracle, and as far as I know - only Oracle, allows you to nest two aggregate functions.
SELECT max (f.name) keep (dense_rank last order by count (*)) as name
from author f
JOIN book b on b.tittle = f.book
Group by f.name
In order to get ALL top authors:
select name
from (SELECT f.name,rank () over (order by count(*) desc) as rnk
from author f
JOIN book b on b.tittle = f.book
Group by f.name
)
where rnk = 1
Since Oracle 12c:
SELECT f.name
from author f
JOIN book b on b.tittle = f.book
Group by f.name
order by count (*) desc
fetch first row /* with ties (optional, in order to get all top authors) */
The best way to do is to use:
SELECT f.name, COUNT(*) as num_books
from author f
JOIN book b on b.tittle = f.book
Group by f.name
Order by num_books DESC
FETCH FIRST ROW ONLY
This will order the results from biggest to smallest and return the first result.
1) Oracle Specific : ( Using ROWNUM, For Postgres/MySql use limit )
select * from
(SELECT f.name, COUNT(*) as num_books
from author f
JOIN book b on b.tittle = f.book
Group by f.name order by num_books desc )
where ROWNUM = 1
2) General Query for all databases :
select f.name,count(*) as max_num_books from author f
JOIN book b on b.tittle = f.book
Group by f.name
having count(*) =
(select max(num_books)
from
(SELECT f.name, COUNT(*) as num_books
from author f
JOIN book b on b.tittle = f.book
Group by f.name)
);
I am not sure why you need a join in the first place. It appears that the author table has a column book - why is it not enough to count(book) from that table, grouping by name? This arrangement is very strange - the author table should only have author properties, the author name should be in the title table, but you do join on author.book = book.title which seems to suggest that you do, in fact, have that strange arrangement (and therefore you don't need a join). Also, having a table and a column (in another table) share the same name, book, is a practice best to be avoided.
The most elementary solution (not the most efficient though), in this case, is
select name, count(book) as max_num_books
from author
group by name
having count(book) = (select max(count(book) from author group by name);
The subquery groups by name, and then it selects the max over all group counts. The outer query selects the names that have a book count equal to this maximum. The subquery returns a single row in a single column - a single value. Such a query is called a "scalar" subquery and can be used wherever a single value is needed, such as the HAVING clause of the outer query. (It's in the HAVING clause and not a WHERE clause, since it refers to group properties - count(book) - and not to individual row properties).
The more efficient solution is as Dudu showed:
select name, ct as max_num_books
from ( select name, count(*) as ct, rank() over (order by count(*) desc) rnk
from author
group by name
)
where rnk = 1;
I have 3 tables:
Person (with a column PersonKey)
Telephone (with columns Tel_NumberKey, Tel_Number, Tel_NumberType e.g. 1=home, 2=mobile)
xref_Person+Telephone (columns PersonKey, Tel_NumberKey, CreatedDate, ModifiedDate)
I'm looking to get the most recent (e.g. the highest Tel_NumberKey) from the xref_Person+Telephone for each Person and use that Tel_NumberKey to get the actual Tel_Number from the Telephone table.
The problem I am having is that I keep getting duplicates for the same Tel_NumberKey. I also need to be sure I get both the home and mobile from the Telephone table, which I've been looking to do via 2 individual joins for each Tel_NumberType - again getting duplicates.
Been trying the following but to no avail:
-- For HOME
SELECT
p.PersonKey, pn.Phone_Number, pn.Tel_NumberKey
FROM
Persons AS p
INNER JOIN
xref_Person+Telephone AS x ON p.PersonKey = x.PersonKey
INNER JOIN
Telephone AS pn ON x.Tel_NumberKey = pn.Tel_NumberKey
WHERE
pn.Tel_NumberType = 1 -- e.g. Home phone number
AND pn.Tel_NumberKey = (SELECT MAX(pn1.Tel_NumberKey) AS Tel_NumberKey
FROM Person AS p1
INNER JOIN xref_Person+Telephone AS x1 ON p1.PersonKey = x1.PersonKey
INNER JOIN Telephone AS pn1 ON x1.Tel_NumberKey = pn1.Tel_NumberKey
WHERE pn1.Tel_NumberType = 1
AND p1.PersonKey = p.PersonKey
AND pn1.Tel_Number = pn.Tel_Number)
ORDER BY
p.PersonKey
And have been looking over the following links but again keep getting duplicates.
SQL select max(date) and corresponding value
How can I SELECT rows with MAX(Column value), DISTINCT by another column in SQL?
SQL Server: SELECT only the rows with MAX(DATE)
Am sure this must be possible but been at this a couple of days and can't believe its that difficult to get the most recent / highest value when referencing 3 tables. Any help greatly appreciated.
select *
from
( SELECT p.PersonKey, pn.Phone_Number, pn.Tel_NumberKey
, row_number() over (partition by p.PersonKey, pn.Phone_Number order by pn.Tel_NumberKey desc) rn
FROM
Persons AS p
INNER JOIN
xref_Person+Telephone AS x ON p.PersonKey = x.PersonKey
INNER JOIN
Telephone AS pn ON x.Tel_NumberKey = pn.Tel_NumberKey
WHERE
pn.Tel_NumberType = 1
) tt
where tt.rn = 1
ORDER BY
tt.PersonKey
you have to use max() function and then you have to order by rownum in descending order like.
select f.empno
from(select max(empno) empno from emp e
group by rownum)f
order by rownum desc
It will give you all employees having highest employee number to lowest employee number. Now implement it with your case then let me know.
I have 4 tables with the following structure:
Table artist:
artistID lastname firstname nationality dateofbirth datedcease
Table work:
workId title copy medium description artist ID
Table Trans:
TransactionID Date Acquired Acquistionprice datesold askingprice salesprice customerID workID
Table Customer:
customerID lastname Firstname street city state zippostalcode country areacode phonenumber email
First question is which artist has the most works of artsold and how many of the artist works have been sold.
My SQL query is this:
SELECT * From dtoohey.artist A1
INNER JOIN
(
SELECT COUNT(W1.ArtistID) AS COUNTER, artistID FROM dtoohey.trans T1
INNER JOIN dtoohey.work W1
ON W1.workid = T1.Workid
GROUP BY W1.artistID
) TEMP1
ON TEMP1.artistID = A1.artistID
WHERE A1.artistID = TEMP1.artistId
ORDER BY COUNTER desc;
I am to get the whole table but I only want show only the first row which is the highest count how do I do that??
I have tried inserting WHERE ROWNUM <=1 but it shows artist ID with 1
qns 2 is sales of which artist's work have resulted in the highest average profit (i.e) the average of the profits made on each sale of worksby an artist), and what is that amount.
My SQL query is:
SELECT A1.artistid, A1.firstname FROM
(
SELECT
(salesPrice - AcquisitionPrice) as profit,
w1.artistid as ArtistID
FROM dtoohey.trans T1
INNER JOIN dtoohey.WORK W1
on W1.workid = T1.workid
) TEMP1
INNER JOIN dtoohey.artist A1
ON A1.artistID = TEMP1.artistID
GROUP BY A1.artistid
HAVING MAX(PROFIT) = AVG(PROFIT);
I'm not able to execute it
I have tried query below but still not able to get it keep getting the error missing right parenthesis
SELECT A1.artistid, A1.firstname, TEMP1.avgProfit
FROM
(
SELECT
AVG(salesPrice - AcquisitionPrice) as avgProfit,
W1.artistid as artistid
FROM dtoohey.trans T1
INNER JOIN dtoohey.WORK W1
ON W1.workid = T1.workid
GROUP BY artistid
ORDER BY avgProfit DESC
LIMIT 1
) TEMP1
INNER JOIN dtoohey.artist A1
ON A1.artisid = TEMP1.artistid
Sometimes ORA-00907: missing right parenthesis means exactly that: we have a left bracket without a matching right one. But it can also be thrown by a syntax error in a part of a statement bounded by parentheses.
It's that second cause here: LIMIT is a Mysql command which Oracle does not recognise. You can use an analytic function here:
SELECT A1.artistid, A1.firstname, TEMP1.avgProfit
FROM
(
select artistid
, avgProfit
, rank() over (order by avgProfit desc) as rnk
from (
SELECT
AVG(salesPrice - AcquisitionPrice) as avgProfit,
W1.artistid as artistid
FROM dtoohey.trans T1
INNER JOIN dtoohey.WORK W1
ON W1.workid = T1.workid
GROUP BY artistid
)
) TEMP1
INNER JOIN dtoohey.artist A1
ON A1.artisid = TEMP1.artistid
where TEMP1.rnk = 1
This uses the RANK() function which will return more than one row if several artists achieve the same average profit. You might want to use ROW_NUMBER() instead. Analytic functions can be very powerful. Find out more.
You can apply ROWN_NUMBER(), RANK() and DENSE_RANK() to any top-n problem. You can use one of them to solve your first problem too.
"however the avg profit is null."
That's probably a data issue. If one of the numbers in (salesPrice - AcquisitionPrice) is null the result will be null, and won't be included in the average. If all the rows for an artist are null the AVG() will be null.
As it happens the sort order will put NULL last. But as the PARTITION BY clause sorts by AvgProfit desc that puts the NULL results at rank 1. The solution is to use the NULLS LAST in the windowing clause:
, rank() over (order by avgProfit desc nulls last) as rnk
This will guarantee you a non-null result at the top (providing at least one of your artists has values in both columns).
1st question - Oracle does not guarantee the order by which rows are retrieved. Hence you must first order and then limit the ordered set.
SELECT * from (
SELECT A1.* From dtoohey.artist A1
INNER JOIN
(
SELECT COUNT(W1.ArtistID) AS COUNTER, artistID FROM dtoohey.trans T1
INNER JOIN dtoohey.work W1
ON W1.workid = T1.Workid
GROUP BY W1.artistID
) TEMP1
ON TEMP1.artistID = A1.artistID
WHERE A1.artistID = TEMP1.artistId
ORDER BY COUNTER desc
) WHERE ROWNUM = 1
2nd question: I believe (haven't tested) that you have that LIMIT 1 wrong. That keyword is for use with Bulk collecting.