So i have this relational model for hospital (not made by me).
Patient (has an adress and an id), hospital (has id and address), and also there's a table for relationship representing placement in the hospital (hospital.id, patient.id) (also there's other tables, but they don't matter in this query);
The purpose of the query is to find hospitals where is no placed patients from from other cities than hospital's one (on condition that address only contains city).
The problem that i have is theoretical, i don't really know if to use full outer join with a or b null, or something else in the query that finds hospitals containing "foreign" patients, (like join hospital with its placement and then full outer join with a or b table record null, but that leads to a question will i get results in the query? Because i need cities that don't match but all the explanations of that join are about .
Thanks to all who embraced my utterly imperfect english and understood it.
Upd.
Patient:
id=1, city =A;
id=2, city =B;
id=3, city =B;
id=4, city =A;
id=5, city =C;
Hospital:
Id =1, city=A
id =2, city=B;
Placement:
h.id p.id
1 1
1 4
2 2
2 3
2 5
Expected results is "1", id of the first hospital (where's no patients from other city) and others with that "feature"
my query is like
select id from hospital where id not in
(select id,address from hospital inner join placement on h.id=placement.h.id as b inner join patient on placement.p.id=p.id where hospital.address<>patient.address )
Sorry for the delay
Is shawn's query correct?
Can i use h.id instead 1? Idk if our teacher would accept that, because he's never showed us something like that and in 10 years he hasn't managed to create an example of that database for students to test queries on.
select * from hospitals h
where not exists (
select 1 -- dummy value, use h.id if you prefer
from patients p inner join placement pl on pl.pid = p.id
where pl.hid = h.id and p.city <> h.city
)
or
select h.id
from hospitals h
left outer join
placement pl inner join patients p on p.id = pl.pid
on pl.hid = h.id
group by h.id
-- this won't count nulls resulting from zero placements for that hospital
-- as long as standard sql null comparisons are active
having count(case when h.city <> p.city then 1 end) = 0
Looks like it works to me: http://rextester.com/BTJB59061
Related
This is a vote system, where candidate can be voted from different(limited) places. and I want the number of vote per place of each candidate.
I have 3 tables
TABLE candidate
------------------
id
name
TABLE place
------------------
id
label
TABLE vote
------------------
id
id_candidate
id_vote
no_votes // represents the amount of votes in this place for that particular candidate
Suppose I have 10 candidates and 15 different places, I'm trying to make a query that will return 10*15 = 150 rows even if there is no votes, keeping NULL value for ids that are not present in the relational table(which i can replace by 0).
But i'm not making the correct query
Here is the query i made so far (i've tried many modification, inner, outer joins... but nothing worked)
SELECT *
FROM votes
RIGHT JOIN candidate ON candidate.id = candidate_id
LEFT JOIN palce ON place.id = place_id
First, if you want the number of votes per candidate, then you should be thinking "aggregation".
Second, don't mix left and right joins in a query. It is just confusing. Start with the table where you want to keep all the rows, and then just use left join.
So, something like this:
SELECT c.*,
SUM(p.place_name = 'place1') as place1,
SUM(p.place_name = 'place2') as place2,
SUM(p.place_name = 'place3') as place3
FROM candidate c LEFT JOIN
votes v
ON c.id = v.candidate_id LEFT JOIN
place p
ON p.id = v.place_id
GROUP BY c.id;
Considering:
TABLE vote
------------------
id PK
id_candidate FK to candidate
id_vote FK to place
no_votes
-
SELECT CA.name,
PL.label,
SUM(VO.no_votes) as votes
FROM candidate CA
LEFT JOIN vote VO ON CA.id = VO.id_candidate
LEFT JOIN place PL ON PL.id = VO.id_vote
GROUP BY CA.id, PL.id
I'm sure this has been asked before but can't find the answer.
I have 3 tables OWNER, CAR, HOUSE
OWNER has 2 columns id and name
CAR has 3 columns id, ownerId and cartype
HOUSE has 4 columns id, ownerId, address, country
I want to write a SQL query that gets the owners name, cartypes, and addresses that are in Sweden
Here comes the tricky part. I want all the owners names and cartypes in the result table even if they don't own a house in Sweden. Can I get all that in 1 query or do I need to use 2? How would that query look?
You should be able to accomplish this with a simple left join:
SELECT O.name, C.cartype, H.address, H.country
FROM OWNER AS O
JOIN CAR AS C ON O.id = C.ownerid
LEFT JOIN HOUSE AS H ON O.id = H.ownerid AND Ucase(H.country) = "SWEDEN"
This will always give you a list of owners and their car types, in addition, it will give you a list of those that also happen to have a house address in sweden.
First you need to join the table then add new column in query by using CASE to check
SELECT o.* , c.* ,h.*,
(CASE WHEN h.county ='sweden' THEN h.county ELSE NULL END) AS HasCountry
FROM OWNER o
JOIN CAR c ON (c.ownerId =o.id)
JOIN HOUSE h ON (h.ownerId =o.id)
I need to write a SQL query to get the patients that have stayed in ALL the hospitals of the city where they live. In one city there may be several hospitals of course.
So for example, if the patient 'xxx' who lives in Washington has been in a hospital, I need to list him only if he's been in all the hospitals of Washington and no less.
This is the structure of the tables:
table patient
patientID
patientCity
table hospital
hospitalCode
hospitalCity
table hospital_stay
hospitalCode
patientID
cityStay
What's the most efficient way to do this for MySQL? Thank you!
Unfortunately, MySQL can't order before grouping, so I had to use subquery to order the result correctly before grouping it.
Have fun :)
SELECT * FROM (
SELECT
p.patientID,
hs.hospitalCode
FROM
patient p
INNER JOIN hospital h ON (p.patientCity = h.hospitalCity)
LEFT JOIN hospital_stay hs ON (p.patientID = hs.patientID AND h.hospitalCode = hs.hospitalCode)
ORDER BY 2
) AS tmp_table
GROUP BY 1
HAVING NOT ISNULL(hospitalCode)
This query should work :
Select p.patientID
, p.patientCity
from patient p
inner join hospital h on h.hospitalCity = p.patientCity
inner join hospital_stay hs on hs.hospitalCode = h.hospitalCode
--where hs.cityStay = 1
group by p.patientID, p.patientCity
having count(*) = (select count(*) from hospital
where hospitalCity = p.patientCity);
Remove the comment if cityStay is kind of a flag that says that the patient went to the hospital.
I have a hard time figuring out how to write a query that selects all rows that matches certain conditions from one table and extends the rows with data from another table if there is data that matches another set of conditions.
Table: books
id
school
isbn
name
Table: orders
id
school
department
isbn
quantity
The query I have is:
SELECT orders.*, books.name FROM orders
LEFT JOIN books ON orders.isbn = books.isbn
WHERE orders.school = 1 AND orders.department = 2
AND books.school = 1
Now, the problem is that if a school hasn't added their books, isbn and names into the books table I'd still like to have the orders.* data returned with books.name set to null or something similar. Now I get zero rows instead. Is there a way to do this with one query?
Move
and books.school = 1
from the where clause to the from clause.
left join books on orders.isbn = books.isbn
and books.school = 1
Edit starts here
for raheel who says it doesn't matter. Using the world database that came with the MySQL I downloaded, this query
select c.name, city.name cityname
from country c left join city on c.code = city.countrycode
AND city.name = 'toronto'
order by cityname;
returns 239 rows. This query:
select c.name, city.name cityname
from country c left join city on c.code = city.countrycode
WHERE city.name = 'toronto'
order by cityname;
returns 1 row. I think it matters
I have three tables persons, jobs, jobs_persons
One person can have multiple jobs.
Person 1 is Technical Support AND Manager
Person 2 is Technical Support
Person 3 Is Manager
Job 1 Technical Support
Job 2 Manager
I need to find a query give me the result for the person who currently is Technical Support AND Manager
The answer would be only Person 1
SELECT persons.*
FROM persons INNER JOIN jobs_persons ON persons.id = jobs_persons.person_id
INNER JOIN jobs ON jobs.id = jobs_persons.job_id
WHERE job.id IN (1,2)
Returns 3 rows
SELECT persons.*
FROM persons INNER JOIN jobs_persons ON persons.id = jobs_persons.person_id
INNER JOIN jobs ON jobs.id = jobs_persons.job_id
WHERE job.id = 1 AND job.id = 2
Returns 0 rows.
I'm currently working on Ruby on Rails.
Somebody can help?
You want to use an OR operator. Using job.id = 1 AND job.id = 2 will only return elements where id equals 1 and at the same time 2. No element can do that. You want elemets where th id is 1 or where the id is 2.
to make it more obvious:
SELECT * FROM table WHERE lastname = 'Smith' AND firstname = 'James';
when executing this you obviously don't want everybody who is called Smith or James. ;-)
EDIT:
Misread the question. what you need is a second join to join the jobs table two times in and join them with the different jobs. It is a bit hard as you didn't show the schema, but this might work:
SELECT persons.*
FROM persons
INNER JOIN jobs_persons jp1 ON persons.id = jp1.person_id
INNER JOIN jobs_persons jp2 ON persons.id = jp2.person_id
INNER JOIN jobs j1 ON j1.id = jp1.job_id
INNER JOIN jobs j2 ON j2.id = jp2.job_id
WHERE j1.id = 1 AND j2.id = 2
Try this to get all the person who has got a job of 1 AND 2 in your associative entity table. No need to hit the job table.
SELECT p.*
FROM persons p
WHERE id in (
SELECT person_id FROM jobs_persons
WHERE job_id IN (1,2)
GROUP BY person_id
HAVING COUNT(*) = 2
);
With your recent comments, it seems you're having performance problems. That's really outside of the scope of this question and answer.
You need to make sure your indexes are in place on the appropriate columns:
jobs_person.job_id
persons.id