The following is what I got right now, which does not work properly because it checks one row for two different values.
SELECT users.*
FROM users INNER JOIN roles_users ru ON users.id = ru.user_id
WHERE ru.role_id = 1 AND ru.role_id = 2
I would like to select all users that have two rows in roles_users. The one rows role_id should have one and the second should have role_id two.
So select all users that have two rows in the roles_users where one of them has role_id = 1 and the other has role_id = 2.
The above query selects all users that have one row in roles_users that has first one and then two, that's why I get no results and it does not work. So how can I do this right?
SELECT users.id
FROM users INNER JOIN roles_users ON users.id = roles_users.user_id
WHERE roles_users.role_id IN (1, 2)
GROUP BY users.id
HAVING COUNT(*) = 2
Why not just join in on roles_users twice? Ala:
SELECT users.* FROM users
INNER JOIN roles_users ru1 ON users.id = ru1.role_id AND ru1.role_id = 1
INNER JOIN roles_users ru2 ON users.id = ru2.role_id AND ru2.role_id = 2
You need to get a (distinct) list of the users having the required roles. Try this instead:
SELECT users.{column_list}
FROM users as a
JOIN (SELECT user_id
FROM roles_users
WHERE role_id IN (1, 2)
GROUP BY user_id
HAVING COUNT(DISTINCT role_id) = 2) required_role
ON required_role.user_id = users.id
Related
I want to show all those users from Users table, whose id is present in table favorite_group_users as user_id for a user whose favorite_groups.id = favorite_group_users.id.
I used below query but it is returning null.
select users.id as user_id, users.first_name as name,
favorite_groups.id as group_id,
favorite_groups_users.user_id as carrier_id
from users
inner join favorite_groups
on users.id = favorite_groups.user_id
inner join favorite_groups_users
on favorite_groups.id = favorite_groups_users.favorite_group_id
where users.id = 38;
You may try nested select
select
favorite_groups.group_id, users_group.user_id
from
favorite_groups_users ,
(select
favorite_groups_users.favorite_group_id,
users.user_id
from
users, favorite_groups_users
where
users.id = 38 and
users.id = favorite_groups.user_id
) users_group
where
users_group.favorite_group_id=favorite_groups.group_id
i want to return follower and following numbers for just one user_id from existing two tables which one is users and the other one is follow:
my code is here
SELECT users.id,
users.name,
COUNT(folow.following),
COUNT(folow.follower)
FROM users
JOIN folow ON folow.following = users.id
WHERE users.id = '$other' LIMIT 10
It's probably simplest to use correlated subqueries here:
SELECT id,
name,
(SELECT COUNT(*) FROM folow WHERE following = users.id) followers,
(SELECT COUNT(*) FROM folow WHERE follower = users.id) following
FROM users
WHERE id = ?
I have two table users and album. In users table there is user_id primary key .In other table albums there are multiple rows with that user_id because every time when a user upload a new album it uploads with user_id as foreign key. I want to select only once the user_id with other table(album) ignore other result set.
How can I achieve this?
SELECT a.*, b.*
FROM users a
INNER JOIN album b
ON a.user_ID = b.user_ID
INNER JOIN
(
SELECT user_ID, MAX(photo_id) max_rec
FROM album
GROUP BY user_ID
) c ON b.user_ID = c.user_ID AND
b.photo_id = c.max_rec
SELECT album.* FROM album LEFT JOIN users ON user.id = album.id WHERE user.id = SOMEIDHERE
I believe this will work, your not giving me a whole lot of info to work with.
SELECT *
FROM ( SELECT u.*, a.*
FROM users AS u
INNER JOIN album AS a
ON u.user_ID = a.user_ID
ORDER BY a.created DESC) AS h
GROUP BY user_ID
ORDER BY b.created DESC -> ORDER BY whatever row you wish for. In this case the newest one is chosen.
I've been fiddleing with this about 20 minutes now and though I might get some help here. I have two tables, in a many-to-one relationship. users (one) and events (many). The tables, simplified, look like this:
users
-----------
user_id
group_id
events
-----------
user_id
I'm trying to produce a report that will show the number of users with at least one event, per group. so a result set would hopefully look like:
group_id users_with_at_least_one_event
1 689
2 312
SELECT group_id, count(*) users_with_at_least_one_event
FROM
(
SELECT u.user_id, u.group_id, count(*) as event_count
FROM users u
JOIN events e
ON u.user_id = e.user_id
GROUP BY u.user_id, u.group_id
HAVING count(*) > 1
) sub
GROUP BY group_id
SELECT group_id, COUNT(users.user_id) AS users_with_at_least_one_event,
FROM users, events
WHERE users.user_id = events.user_id
GROUP BY group_id
maybe?
try
select users.group_id, count(distinct users.user_id) as users_with_at_least_one_event
from
users, events
where users.user_id = events.user_id
group by users.group_id
SELECT users.group_id, COUNT(*) FROM users
WHERE users.user_id IN (SELECT events.user_id FROM events)
I have the following table structure:
- table users -
user_id
user_name
- table groups -
group_id
group_name
- table group_members -
group_id
user_id
Let's also say I have this in the DB:
users:
1:administrator
groups:
1:administrators
2:superusers
3:normal users
group_members:
1:1 (user_id 1 is member of group_id 1)
1:2
1:3
Now, how do I go efficient with selecting all users, and with that, the groups they're member of? Do I have to execute 3 queries selecting all rows from all 3 tables, and fetch it with arrays in PHP, or is there a more effecient way to achieve this?
select u.*, g.group_name
from users u
inner join group_members gm on gm.user_id = u.user_id
inner join groups g on g.group_id = gm.group_id
where u.user_id = 123
This query should work if there are no null fields in group and group_members table
if there are null fields replace inner join with left join
select users.username, group.group_name
from users
inner join group_members on group_members.user_id = users.user_id
inner join groups on group.group_id = group_members.group_id