Inner join 3 tables, retrieve if conditions - mysql

I have three tables
Person
- id
- other_fields
PersonTypes
- id
- person_id
- type_id
PersonEmails
- id
- person_id
- email_address
How to get only these email addresses where PersonTypes.type_id is different than value ?
I'm using MariaDB v.10, I tried to use left outer joins with in all tables and then return only these rows where type_id is different. But I'm not sure about the results.
SELECT pe.email_address
FROM Person p
LEFT OUTER JOIN PersonTypes pt ON p.id = pt.person_id
LEFT OUTER JOIN PersonEmails pe ON p.id = pe.person_id
WHERE pt.type_id != 14;

select P1.id,
P2.type_id, -- just for the sake of this output
P2.email_address
from Person P1
inner join
(
select PT.person_id, PT.type_id, PE.email_address
from PersonTypes PT
inner join PersonEmails PE
on PE.person_id = PT.person_id
where PT.type_id <> 14
) P2
on P2.person_id = P1.id
or
select pe.email_address
from PersonEmails pe
where person_id not in
(
select person_id
from PersonTypes
where Type_id = 14
)

The query selects all persons that don't have type_id 14 and joins Person and PersonsEmails tables data
SELECT pe.email_address
FROM PersonTypes pt
INNER JOIN Person p ON pt.person_id = p.id
INNER JOIN PersonEmails pe ON pt.person_id = pe.person_id
WHERE pt.type_id != 14;

Related

Get only those rows where column a has as many duplicate entries, as many there is distinctive values in column b

My current query:
select users.id as user_id, opportunities.id as op_id, opportunities.title, certificates.id as cert_id from opportunities
join opportunity_certificates on opportunities.id=opportunity_certificates.opportunity_id
join certificates on opportunity_certificates.certificate_id=certificates.id
join user_certificates on certificates.id=user_certificates.certificate_id
join users on user_certificates.user_id=users.id
where opportunity_certificates.is_required = 1 and
opportunities.id = 1
This produces the table on the picture below.
cert_id column can have values from 1 to 7, depends on the opportunities.id. In the table below, I want the query to return only the rows which have the same user_id but different cert_id, 1 and 2.
If the table had 3 different cert_id, I would want it to return only the rows which have same user_id but different cert_id, 1,2 and 3.
when the cert_id has only one value, query should return all the records with that one value in cert_id. Basically, it should show all users who have all required certificates.
The query has to be in the current format. I experimented with
group by users.id
having count(*) >
but I don't know how to make that comparison dynamic, relative to the count of distinctive values in the cert_id column.
Compare counts with a having condition.
select u.id as user_id --, o.id as op_id, o.title
from opportunities o
join opportunity_certificates oc on o.id=oc.opportunity_id
join certificates c on oc.certificate_id=c.id
join user_certificates uc on c.id=uc.certificate_id
join users u on uc.user_id=u.id
where oc.is_required = 1 and o.id = 1
group by u.id --,o.id,o.title
having count(distinct c.id)=(select count(distinct id) from certificates)
Useful?
with data as (
select users.id as user_id, o.title, c.id as cert_id
from opportunities o
inner join opportunity_certificates oc on oc.opportunity_id = o.id
inner join certificates c on c.id = oc.certificate_id
inner join user_certificates uc on uc.certificate_id = c.id
inner join users u on u.id = uc.user_id
where oc.is_required = 1 and o.id = 1
)
select user_id, min(title) as title, max(cert_id) as num_certs
from data
group by user_id
having count(cert_id) = (select max(cert_id) from data);
I'm assuming that cert_id values start and 1 and run sequentially. You could also use count(distinct ...) in the having clause but it guess it's debatable which ones expresses you intent more clearly.
If your version of MySQL doesn't support CTEs then you should be able to just drop that whole subquery into the having clause as well.
select u.id as user_id, min(o.title) as title, max(c.cert_id) as num_certs
from opportunities o
inner join opportunity_certificates oc on oc.opportunity_id = o.id
inner join certificates c on c.id = oc.certificate_id
inner join user_certificates uc on uc.certificate_id = c.id
inner join users u on u.id = uc.user_id
where oc.is_required = 1 and o.id = 1
group by u.id
having count(c.cert_id) = (
select max(c.cert_id)
from opportunities o
inner join opportunity_certificates oc on oc.opportunity_id = o.id
inner join certificates c on c.id = oc.certificate_id
inner join user_certificates uc on uc.certificate_id = c.id
inner join users u on u.id = uc.user_id
where oc.is_required = 1 and o.id = 1
);
Here's another one that might work if you have window functions available. (It might work with Laravel better?):
select *
from (
select users.id as user_id, o.title,
count(distinct c.id) over (partition by u.id) as user_certs,
max(c.id) over () as total_certs
from opportunities o
inner join opportunity_certificates oc on oc.opportunity_id = o.id
inner join certificates c on c.id = oc.certificate_id
inner join user_certificates uc on uc.certificate_id = c.id
inner join users u on u.id = uc.user_id
where oc.is_required = 1 and o.id = 1
) t
where user_certs = total_certs;

How to optimize query with two inner join

Using this query to get the products with words that fulfill all three required word terms (lenovo, laptop, computer):
SELECT t1.id, t1.name, t1.price FROM
(SELECT p.id AS productid, name, price
FROM products p JOIN productwords pw ON p.id = pw.productid
JOIN words w ON pw.wordid = w.id WHERE word.term = 'lenovo') t1
INNER JOIN
(SELECT p.id AS productid, name, price
FROM products p JOIN productwords pw ON p.id = pw.productid
JOIN words w ON pw.wordid = w.id WHERE word.term = 'laptop') t2
INNER JOIN
(SELECT p.id AS productid, name, price
FROM products p JOIN productwords pw ON p.id = pw.productid
JOIN words w ON pw.wordid = w.id WHERE word.term = 'computer') t3
ON
t1.productid = t2.productid
AND
t1.productid = t3.productid
ORDER BY t1.name
As far as I can see, the query considers the whole words table for each term (the tables have indexes. Database is MySql).
Can the query be rewritten in a better way, so it will become faster? (the tables contain millions of rows)
For example with subsets, so the 'laptop' search only considers the rows matching 'lenovo' - and the 'computer' search only considers the rows matching first 'lenovo' and then 'laptop'.
Thanks!
You can use the HAVING clause :
SELECT p.id AS productid, name, price
FROM products p
JOIN productwords pw ON p.id = pw.productid
JOIN words w ON pw.wordid = w.id
WHERE word.term in ('lenovo','computer','laptop')
GROUP BY p.id , name, price
HAVING COUNT(DISTINCT word.term) = 3
That is if I understood the question, it looks like product -> words is 1:n relation , and if no column from the word table is selected, that should work perfectly.
This might be a quicker way of doing it:
SELECT p.id, name, price
FROM products p
where
EXISTS (select null
from productwords pw1
JOIN words w1 ON pw1.wordid = w1.id
where w1.term = 'lenovo'
and p.id = pw1.productid )
and EXISTS (select null
productwords pw2
JOIN words w2 ON pw2.wordid = w2.id
where w2.term = 'laptop'
and and p.id = pw2.productid )
and EXISTS (select null
productwords pw3 ON p.id = pw3.productid
JOIN words w3
where w3.term = 'computer'
and p.id = pw3.productid )
ORDER BY name;

MySQL - #1066 - Not unique table/alias: 'components' with multiple inner joins

I have this query and I am getting error #1066 - Not unique table/alias: 'components'. What seems to be the issue?
SELECT COUNT(*) FROM `products`, `components`, `tradeNames`
INNER JOIN `componentsMap` ON componentsMap.product_id = product.id
INNER JOIN `components` ON componentsMap.component_id = components.id
INNER JOIN `tradeNamesMap` ON .tradeNamesMap.product_id = products.id
INNER JOIN `tradeNames` ON tradeNamesMap.tradeName_id = tradeNames.id
WHERE (((((LOWER(inci) LIKE '%abies%')
OR (trade_name.LOWER(name) LIKE '%abies%'))
OR (components.LOWER(no_cas)='abies'))
OR (components.LOWER(no_einecs)='abies'))
OR (components.LOWER(name)='abies'))
AND (`published`=1)
ORDER BY `trade_name`.`name` DESC
You don't need to list the tables before the INNER JOINs. In fact, simply don't ever use commas in the FROM clause. So:
SELECT COUNT(*)
FROM `products`
INNER JOIN `componentsMap` ON componentsMap.product_id = product.id
INNER JOIN `components` ON componentsMap.component_id = components.id
INNER JOIN `tradeNamesMap` ON tradeNamesMap.product_id = products.id
INNER JOIN `tradeNames` ON tradeNamesMap.tradeName_id = tradeNames.id
WHERE (((((LOWER(inci) LIKE '%abies%')
OR (trade_name.LOWER(name) LIKE '%abies%'))
OR (components.LOWER(no_cas)='abies'))
OR (components.LOWER(no_einecs)='abies'))
OR (components.LOWER(name)='abies'))
AND (`published`=1)
ORDER BY `trade_name`.`name` DESC;
The above query only returns one row because of the COUNT(). The order by suggests that you actually want this information for each trade_name.name. If so, you need a GROUP BY:
SELECT tn.name, COUNT(*)
FROM `products` p INNER JOIN
`componentsMap cm
ON cm.product_id = p.id INNER JOIN
`components` c
ON cm.component_id = c.id INNER JOIN
`tradeNamesMap` tnm
ON tnm.product_id = p.id INNER JOIN
`tradeNames` tn
ON tnm.tradeName_id = tn.id
WHERE ((LOWER(inci) LIKE '%abies%') OR
(tn.LOWER(name) LIKE '%abies%') OR
(c.LOWER(no_cas)='abies') OR
(c.LOWER(no_einecs)='abies') OR
(c.LOWER(name)='abies')
) AND
(`published` = 1)
GROUP BY tn.name
ORDER BY tn.`name` DESC
INNER JOIN `[components]` ON componentsMap.component_id = components.id
AND
SELECT COUNT(*) FROM `products`, [`components`], `tradeNames`
Two components are there.
Just guessing, and untested, but I suspect that something like this would do what you're after...
SELECT n.name
, COUNT(*)
FROM products p
JOIN componentsMap pc
ON pc.product_id = p.id
JOIN components c
ON c.id = pc.component_id
JOIN tradeNamesMap pn
ON pn.product_id = p.id
JOIN tradeNames n
ON n.id = pn.tradeName_id
WHERE
( inci LIKE '%abies%'
OR n.name LIKE '%abies%'
OR 'abies' IN (c.no_cas,c.no_einecs,c.name)
)
AND published = 1
GROUP
BY n.name
ORDER
BY n.name DESC

Multiple foreign keys on same column

I have following tables:
Table: user
Columns
- id
- username
- full_name
Table: pet
Columns
- id
- pet_name
Table: results
Columns
- id
- user_id_1
- user_id_2
- user_id_3
- pet_id
- date
- some_text
Can I make an SQL query witch will give me one row with: id, full_name, full_name, full_name, pet_name, date, some_text where the results.id is 3?
select u1.full_name
, u2.full_name
, u3.full_name
, p.pet_name
, r.date
, r.some_text
from Results r
join User u1
on u1.id = r.user_id_1
join User u2
on u2.id = r.user_id_2
join User u3
on u3.id = r.user_id_3
join pet p
on p.id = r.pet_id
where r.id = 3
Try Following query :
SELECT A.id, B.full_name, C.full_name, D.full_name, E.pet_name, A.date, A.some_text
FROM RESULTS AS A
LEFT OUTER JOIN USER AS B ON A.USER_ID1 = B.ID
LEFT OUTER JOIN USER AS C ON A.USER_ID2 = C.ID
LEFT OUTER JOIN USER AS D ON A.USER_ID3 = D.ID
LEFT OUTER JOIN PET AS E ON A.PET_ID = E.ID
WHERE A.id = 3

Select a product from filters

I have a problem with a query:
I have 3 tables:
products (id, name)
settings (id, name)
product_setting (product_id, setting_id)
for example: I would like to select only the products you have selected filters!
I do this:
SELECT p. *, s.id as setting
FROM Products p
INNER JOIN product_setting p2 ON (p.id = p2.product_id)
INNER JOIN settings s ON (s.id = p2.setting_id)
WHERE s.id IN (1,2)
but I get all products that have the 'setting' id = 1 OR id = 2.
How to get only those products that have those 'setting' (AND)?
thanks!!
SELECT p.*, s.id as setting
FROM Products p
INNER JOIN product_setting p2 ON (p.id = p2.product_id)
INNER JOIN settings s ON (s.id = p2.setting_id)
WHERE s.id IN (1,2)
GROUP BY p.id
HAVING COUNT(*)=2; // size of IN()
This seems like over kill but...
SELECT p. *, s.id as setting
FROM Products p
INNER JOIN product_setting p2 ON (p.id = p2.product_id)
INNER JOIN settings s ON (s.id = p2.setting_id)
INNER JOIN settings s2 ON (s.id = p2.setting_id)
WHERE
s.id = 1
AND s2.id = 2