SQL - Get two users' names - mysql

I'm trying to get names of users from a voting table, I can't quite figure out how to get both the voter's name and the votee's name.
Simplified Votes Table
ID voteeID voterID
1 1 2
2 1 3
3 1 4
4 2 1
5 2 3
6 2 4
Simplified User Info Table
ID username
1 Bob
2 Sam
3 Ed
4 Mark
Where I'm at:
SELECT votes.voteeID, votes.voterID, userinfo.username
FROM votes
JOIN ???
I was playing around with the join, but I'm kind of stuck. If I join on the votee's id to the userinfo id, how would I also get the voter's name or vice versa?
Database: MySQL 5
Any help or even a hint would be wonderful, thanks in advance.

try this one,
SELECT b.`username` as Votee_Name,
c.`username` as Voter_Name
FROM Votes a
INNER JOIN UserInfo b
ON a.voteeID = b.Id
INNER JOIN UserInfo c
on a.voterID = c.id
SQLFiddle Demo

How about something like
SELECT v.ID,
v.voteeID,
u1.username Votee,
v.voterID,
u2.username Voter
FROM Votes v INNER JOIN
User u1 ON v.voteeID = u1.ID INNER JOIN
User u2 ON v.voterID = u2.ID

You need to take INNER JOIN twice to get usernames for both votee and voter:
SELECT a.voteeID, a.voterID,
b.username AS votee_name, c.username AS voter_name
FROM votes a
INNER JOIN userinfo b
ON a.voteeID = b.ID
INNER JOIN userinfo c
ON a.voterID = c.ID;

SELECT V.ID as ID B.username as Votee_Name, C.username as Voter_Name
FROM Votes V
INNER JOIN UserInfo B ON V.voteeID = B.Id
INNER JOIN UserInfo C on V.voterID = C.id;

Related

MySQL query with join + count for a specific user in DB

I want to get the sum of several entities from several tables using a query.
Specifically, I want the devices and messages count of a user. In addition, the timestamp of the last received message for that user.
I have 3 tables:
Users
id name (other fields)
1 Mike ...
2 John ...
3 Yay ...
4 Jim ...
Devices
id user_id (other fields)
1 1 ...
2 1 ...
3 1 ...
4 2 ...
Messages
id device_id message time (other fields)
1 1 Hi 2019-04-07 12:06:44 ...
2 1 Hey 2019-04-06 12:06:44 ...
3 2 Sup 2019-04-05 12:06:44 ...
4 3 Ok 2019-04-04 12:06:44 ...
5 4 Yay 2019-04-08 12:06:44 ...
... and, for example, for user Mike I want to end up with:
Result
nDevices nMessages time
3 4 2019-04-07 12:06:44
Any ideas?
Thanks in advance.
Join the 3 tables and count distinct values of the columns:
select
count(distinct d.id) devicescounter,
count(distinct m.id) messagescounter,
max(m.time) lastmessagetime
from users u
left join devices d on u.id = d.user_id
left join nessages m on m.device_id = d.id
where u.name = 'Mike'
If you want the results for all users:
select
u.id, u.name,
count(distinct d.id) devicescounter,
count(distinct m.id) messagescounter,
max(m.time) lastmessagetime
from users u
left join devices d on u.id = d.user_id
left join nessages m on m.device_id = d.id
group by u.id, u.name
The easiest way to do this would be to get the number of messages and most recent message time for each device in a subquery.
SELECT u.id,
COUNT(d.id) AS num_devices,
SUM(messages) AS num_messages,
MAX(most_recent) AS most_recent_message
FROM users u
JOIN devices d ON d.user_id = u.id
JOIN (SELECT device_id,
COUNT() AS messages,
MAX(TIME) AS most_recent_message
FROM messages
GROUP BY device_id) m ON m.device_id = d.id
GROUP BY u.id
You could try using a join between user and device and a inner join with the subquery for message and device
select a.name, a.id, count(d.user_id), t.nDevice, t.max_time
from Users a
inner join device d on a.id = d.user_id
inner join (
select m.device_id, count(m.device_id) nDevice, max(m.time) max_time
from Messages m
group by m.device_id
) t on t.device_id = d.id
group by a.name, a.id

struggling with a sql query join

I am struggling with writing a query join in mysql
I have two table
Challenges
challenge_ID(int) |to_user(int)|from_user(int)|timestamp|gameID=>nullable
Users
iduser(int)|first_name(string)
I want get the first name of to_user and form_user when I have the challengeID
for instance if
Challenges
challenge_ID(int) |to_user(int)|from_user(int)|timestamp|gameID
1 9 10 sometimestamp
Users
iduser(int)|first_name(string)
9 Tom
10 Chris
11 Patrick
I would like to get 'Tom' and 'Chris' for challenge id 1
Thanks for your help.
It may be something like this:
SELECT first_name
FROM Users
WHERE iduser IN (SELECT to_user
FROM challenges
WHERE Challenge_Id = 1
UNION
SELECT from_user
FROM challenges
WHERE Challenge_Id = 1)
this is how you do this
select c.* ,u.first_name as to_name , u2.first_name as from_name
FROM challenges c
join users u on c.to_user = u.id
join users u2 on c.from_user = u2.id
where c.challenge_ID = 1
It looks like you need a UNION and then a JOIN:
SELECT users.first_name
FROM (
SELECT from_user AS usr FROM challenges WHERE challenge_id=1
UNION
SELECT to_user FROM challenges WHERE challenge_id=1
) u INNER JOIN users ON u.usr = users.id_user
UNION will remove duplicates on the subquery, if there are no duplicates you can use UNION ALL which is faster.
You can try this query
SELECT u_to.first_name, u_from.first_name
FROM challenges c
INNER JOIN users u_to ON u_to.iduser = c.to_user
INNER JOIN users u_from ON u_from.iduser = c.from_user
WHERE c.challange_ID = 1

mysql group by and take specific record from right join table

I have database with following data structure with sample data. Each company have multiple members. The relationship is in the company_member table. Please note only required fields I have given below.
company
id title
1 company-1
2 company-2
company_member
companyid memberid
1 1
1 2
1 3
2 4
2 5
2 6
member
id firstname member_type_id
1 Name-1 2
2 Name-2 3
3 Name-3 3
4 Name-4 3
5 Name-5 2
6 Name-6 1
member_type
id user_level
1 0
2 1
3 2
I want list of unique companies with one member from each. But the member should be the lowest user_level within the company. i.e, following result should come;
result
companyid company_title memberid member_name user_level
1 company-1 1 Name-1 1
2 company-2 6 Name-6 0
I want to know how to get one member with lowest user level among the same company.
This is a bit complicated one, however this is one way of doing it using not exists, for bigger tables its wise to use not exits since using pivot tables it will not use index.
select
c.id,
c.title,
m.id as member_id,
m.firstname,
mt.user_level
from company_member_map cmp
join company c on c.id = cmp.companyid
join member m on m.id = cmp.memberid
join member_type mt on mt.id = m.member_type_id
where not exists
(
select 1 from company_member_map t1
join member t2 on t2.id = t1.memberid
join member_type t3 on t3.id = t2.member_type_id
where
t1.companyid = cmp.companyid
and t3.user_level < mt.user_level
)
DEMO
Finally I found solution:
select c.id companyid, c.title company_title, m.firstname member_name, mt.user_level
from company c
inner join company_member cm on cm.companyid = c.id
inner join member m on m.id = cm.memberid
inner join member_type mt on mt.id = m.member_type_id
inner join
(select c1.id companyid, mt1.user_level
from company c1
join company_member cm1 on cm1.companyid = c1.id
join member m1 on m1.id = cm1.memberid
join member_type mt1 on mt1.id = m1.member_type_id
group by c1.id,m1.id
order by user_level asc
) sq on c.id = sq.companyid and sq.user_level = mt.user_level
group by c.id;
Correct this, if anyone have better solution or simplified solution.
Check this SQL Fiddle

User CASE or subquery?

I have 3 tables: tickets, tickets_users and users. My problem is that in the users table I have 2 types of users: requesters and solvers: type 1 and 2.
I want to select the ticket, requester (if any, can be multiple) and solver (if any, can be multiple).
I'm thinking something in the lines of:
SELECT
t.id, t.description,
(u.id where u.type = 1) AS requester,
(u.id where u.type = 2) AS solver
FROM
tickets t
INNER JOIN
tickets_users tu ON t.id = tu.ticket_id
INNER JOIN
users u ON tu.user_id = u.id
Obviously this does not work.
The tables look like this:
Tickets:
ID Description
1 Description 1
2 Description 2
Tickets_users
ID Ticket_ID User_id Type
1 3 4 1
2 5 8 2
Users
ID Name
1 John
2 Mary
Thanks,
In the meantime I think I found a solution using a sub-query in the join clause, but to me it looks rudimentary:
SELECT
t.id, t.name AS ticket_name, type1.users_id AS requester,
type2.users_id AS solver
FROM
tickets t
INNER JOIN
(SELECT users_id, tickets_id
FROM tickets_users
WHERE TYPE = 1) type1 ON t.id = type1.tickets_id
INNER JOIN
(SELECT users_id, tickets_id
FROM tickets_users
WHERE TYPE = 2) type2 ON t.id = type2.tickets_id
Should be something like this.
SELECT
t.id, t.description, u.id, userType
CASE WHEN u.type = 1 THEN 'requester' ELSE 'solver' END
FROM
tickets t
INNER JOIN tickets_users tu ON t.id = tu.ticket_id
INNER JOIN users u ON tu.user_id = u.id
Please provide sqlfiddle in case above example does not work.
With Joins,
SELECT
t.id, t.description, requesters.id, solvers.id
FROM
tickets t
INNER JOIN tickets_users tu ON t.id = tu.ticket_id
LEFT JOIN users requesters ON (tu.user_id = requesters.id AND requesters.type=1)
LEFT JOIN users solvers ON (tu.user_id = solvers.id AND solvers.type=2)
As you can tell, users table is joined twice (as requesters and as solvers with additional conditions). The reason LEFT JOIN is used is because if there's a record with no requesters or no solvers, INNER JOIN would completely ignore the whole record until it has a requester and a solver.

MYSQL compare values from same columns

Okay I tried to look all over stackoverflow, and the closest solution I found is this:
mysql AND clause on same column multiple times
But I can't use statements and "having" syntax won't work because of group by. There MUST be a simple solution to this.
The 2 tables looks like this (simplified):
users:
uid name
1 person 1
2 person 2
3 person 3
categories:
uid value
1 actor
1 musician
2 actor
3 dancer
4 musician
4 dancer
I want to get the uid of those that are 2 values at the same time. For example, I want to get the UID that is an actor AND a musician. Not just one value, but both of them must be required!
First I tried this:
SELECT users.uid, users.name
FROM
users
LEFT OUTER JOIN categories ON users.uid = categories.uid
WHERE (categories.value = 'actor' AND categories.value = 'musician')
GROUP BY u.uid;
This of course does not work since one row can't have 2 values.
Does anyone know a solution?
You can JOIN to the categories table multiple times to get the result:
SELECT users.uid, users.name
FROM users
INNER JOIN categories c1
ON users.uid = c1.uid
INNER JOIN categories c2
ON users.uid = c2.uid
WHERE c1.value = 'actor'
AND c2.value = 'musician';
See SQL Fiddle with Demo
SELECT users.uid, users.name
FROM users
LEFT JOIN categories ON users.uid = categories.uid
WHERE categories.value in ('actor', 'musician')
GROUP BY u.uid, users.name
having count(distinct categories.value) = 2;
Use a having clause
SELECT u.uid, u.name
FROM users u
LEFT OUTER JOIN categories c ON u.uid = c.uid
WHERE c.value = 'actor' OR c.value = 'musician'
GROUP BY u.uid
having count(distinct c.value) > 1
If you really do not want to use having you could try this:
SELECT uid, name
FROM users
WHERE
uid IN (SELECT uid FROM categories WHERE value='actor')
AND uid IN (SELECT uid FROM categories WHERE value='musician')
But there is really nothing wrong with using HAVING ;)