Calculate round(avg) from same table and join - mysql

I have two tables,
users
userid fname usertype
1 Steve vendor
2 Peter vendor
3 John normaluser
4 Mark normaluser
5 Kevin vendor
6 Alan vendor
7 Sean vendor
vendor_rating
id userid rating
1 1 4
2 2 3
3 2 2
4 2 4
5 1 3
6 5 2
7 5 2
userid is foreign key.
i want to show all vendors (only usertype vendor) from user table by descending/ascending average rating even if Vendor's rating is not available on table it should show, its information should display at last in descending, at first in ascending.
I want to fetch all users info from first table so i m using left join :
SELECT
users.name,
users.userid,
users.usertype
FROM users
LEFT JOIN (SELECT
ROUND(AVG(rating)) AS rating_avg,
userid
FROM vendor_rating
ORDER BY rating_avg DESC) ven
ON users.usertype = 'vendor'
AND users.userid = ven.userid
ORDER BY ven.rating_avg ASC;
Please help where am i going wrong.
EDIT:
I get this
userid ROUND(AVG(vr.ratings))
28 5
27 4
16 3
26 2
25 0
NULL NULL
NULL NULL
NULL NULL
NULL NULL
if i use
SELECT vr.userid, ROUND(AVG(vr.ratings)) FROM vendor_rating vr
RIGHT JOIN (SELECT users.fname, users.userid, users.usertype FROM users) u
ON u.id = vr.vendor_id WHERE u.usertype = 'vendor' GROUP BY vr.userid,u.fname
ORDER BY round(avg(vr.ratings)) ASC
i get NULL values from users table whose rating is not available in vendor_rating table those should display userids

Try to this
SELECT
vr.userid,
u.fname,
ROUND(AVG(vr.rating))
FROM vendor_rating vr
INNER JOIN users u
ON u.userid = vr.userid
WHERE u.usertype = 'vendor'
GROUP BY vr.userid,
u.fname
ORDER BY round(avg(vr.rating)) ASC

finally i got it
SELECT users.fname, users.userid,users.usertype
FROM users
LEFT JOIN (select ROUND(AVG (ratings)) AS rating_avg,userid FROM
vendor_rating group by userid order by rating_avg desc ) ven
ON users.id=userid
WHERE users.usertype='vendor'
order by rating_avg desc
Thank you all, for sharing views to get idea to solve my problem.

Related

Not able to figure out sql query

i am new to sql and need help with a query. I have two tables user and user_family which contains data like
USER
ID
Date-Of-birth
User_Id
1
2021-05-21
28371
2
2021-04-17
28372
USER_FAMILY
ID
family_detail_id
User_Id
1
1
28371
2
1
28374
3
1
28375
4
2
28372
5
2
28373
6
2
28378
7
2
28379
i want to run a query which checks if current date in equal to someones dob in my user table, if yes i want to return all entries from user_family table which has same family_detail_id to someone whose dob has been matched.
Suppose if the current date is 2021-05-21 then the result should be,
ID
family_detail_id
User_Id
dob
birthday_user_id
2
1
28374
2021-05-21
28371
3
1
28375
2021-05-21
28371
You can use a self-join on the family_detail_id,
select f2.*
from user u join user_family f on f.id=u.id
join user_family f2 on f2.family_detail_id=f.family_detail_id
where u.Date_Of_birth=currdate();
Working Fiddle
I would suggest exists. Assuming that user.id matches to user_family.family_detail_id, then this looks like:
select uf.*
from user_family uf
where exists (select 1
from user u
where u.family_detail_id = u.id and
u.dob = curdate()
);
You may try below query, assuming USER_ID is the linking column between the two tables -
select uf1.*
from user_family uf1
where uf1.family_detail_id in
(select uf.family_detail_id
from user_family uf
inner join user u on u.user_id = uf.user_id
where u.Date-Of-birth = CURDATE()); --DATE(u.Date-Of-birth) = CURDATE()
HTH!

Complex SQL Union

Any SQL Guru's out there I could use some help! I am creating a stored procedure that I believe needs a Union so that all the results are brought back with 1 SELECT statement.
I have simplified my problem to the tables below:
user
user_id username name DOB
------------------------------------------------------
1 JohnSmith1 John Smith 01/01/1990
2 LisaGreen17 Lisa Green 03/07/1986
3 BarneyB Barney Brown 09/12/1960
user_team
user_team_id user_id team_id total_score
-------------------------------------------------------------
1 1 1 29
2 2 7 37
3 3 2 15
private_league
priv_league_id league_name host_user league_password
-------------------------------------------------------------------------
1 Lisa's League 2 CSUASH429d9
2 Barney's Bonanza 3 Jkap89f5I01
user_team_private_league_M2M
id priv_league_id user_team_id
----------------------------------------
1 1 1
2 1 2
3 1 3
4 2 1
5 2 3
I would like to run a stored procedure with an input of a user_id which will bring back all leagues entered by the user, the host of each of those leagues, how many total players have entered in each league and what position the user is in for each of those leagues(sorted by total score).
At the moment I have:
CREATE DEFINER=`root`#`localhost` PROCEDURE `user_private_leagues`(IN v_user_id INT)
BEGIN
DECLARE userteamid INT;
# Retrieve user team from a user_id
SELECT user_team_id INTO userteamid
FROM user_team
WHERE user_id = v_user_id;
# Retrieve private league name and host user (for a userteam)
SELECT private_league.league_name, private_league.host_user
FROM user_team_private_league_M2M
INNER JOIN privateleague
ON user_team_private_league_M2M.priv_league_id=private_league.priv_league_id
WHERE user_team_id = userteamid;
END
This query does not include the total number of players for each league and the current position of the user
I have created a query to bring back the total users for each private league, with no user filter like so:
SELECT private_league_id, COUNT(*) AS total_users
FROM classicseasonmodel_classicseasonuserteamprivateleague
GROUP BY private_league_id;
A query for the user's current position can be worked out by using the answer to this question and using total_score.
I am extremely stuck with this at the moment - the perfect result from the SP will be as follows:
CALL user_private_leagues(3); (user id of BarneyB)
priv_league_name current_position total_users host_user
-----------------------------------------------------------------------
Lisa's League 3 3 LisaGreen17
Barney's Bonanza 2 2 BarneyB
Thanks!
Sorry but I didn't create the DB to test the SQL below. But you can start from there. No need for UNION. I didn't understand the business rule to compute the user's position in the league, since it may come from the team or the user.
select priv_league_id, league_name, host_user_name, count(*) as total_users
from (
select A.priv_league_id, A.league_name, D.name as host_user_name, B.user_team_id, C.user_id, D.
from private_league A
join user_team_private_league_M2M B
on A.priv_league_id = B. priv_league_id
join user_team C
on B.user_team_id = C. user_team_id
join user D
on A.host_user = D.user_id
) D
group by priv_league_id
Let take it step by step....
First ID, Count for pleague
SELECT private_league_id, COUNT(*) AS total_users
FROM classicseasonmodel_classicseasonuserteamprivateleague
GROUP BY private_league_id;
Now add in Name and host user
SELECT PL.league_name, LC.uCNT, PL.host_user
FROM (SELECT private_league_id AS pID, COUNT(*) AS uCNT
FROM classicseasonmodel_classicseasonuserteamprivateleague
GROUP BY private_league_id ) AS LC
LEFT JOIN private_league PL ON PL.priv_league_id = LC.pID
Now add in host user name
SELECT PL.league_name, LC.uCNT as total_users, hu.name as host_user
FROM (SELECT private_league_id AS pID, COUNT(*) AS uCNT
FROM classicseasonmodel_classicseasonuserteamprivateleague
GROUP BY private_league_id ) AS LC
LEFT JOIN private_league PL ON PL.priv_league_id = LC.pID
LEFT JOIN user hu ON PL.host_user = hu.user_id
Don't know where current position is.
This query will give users and position for each team, join to this and limit by user id to get one users position for each team:
select UT.user_id,
UT.team_id,
ROW_NUMBER() OVER (PARTITION BY team_id ORDER BY total_score DESC) AS team_position
from private_league L
join user_team_private_league_M2M LJ ON L.priv_league_id = LJ.priv_league_id
join user_team UT ON LJ.user_team_id = UT.user_team_id

How to make my mysql queries for showing friends feeds?

Here is my table structure.(fun_friends)
id user_id,friend_id,status,createdat,updatedat
1 1 2 1 123456 125461
2 1 3 1 454545 448788
3 2 4 1 565659 898889
4 1 5 1 877878 878788
Here is the table structure of user_uploads
id user_id parent_id category_id title slug tags description video_type source video_link video_thumb
1 2 1 2 fun fun ['4','5'] coolvid 1 ytu link thumb
I need to show the latest upload of my friends
Can you tell me how can i join this tables together? i tried with
SELECT * FROM fun_friends WHERE (user_id= '".$_SESSION['user_row_id']."' AND `status` =1) OR (friend_id= '".$_SESSION['user_row_id']."' AND `status` =1)
and it is showing all friends of logged-in user
You can just join both table using user_id field. sample query bellow will return one record with latest user_uploads.id.
select *
from fun_friends a
inner join user_uploads b on a.user_id = b.user id
order by b.id desc limit 0,1
how about using UNION to get the friends user and wrapping it inside a subquery which later join on the other tabel,
SELECT usr.*
FROM user_uploads usr
INNER JOIN
(
SELECT user_ID AS ID
FROM Fun_Friends
WHERE friend_ID = 'ID_HERE'
UNION
SELECT friend_ID As ID
FROM Fun_Friends
WHERE ser_ID = 'ID_HERE'
) idList ON usr.user_ID = idList.ID

Counting votes and if user voted in MySQL table

I have 3 MySQL tables:
Features table:
id name
----------
1 feature 1
2 feature 2
3 feature 3
4 feature 4
5 feature 5
Votes table:
userid featureid
----------------
1 1
1 2
1 3
1 4
2 1
2 2
2 3
2 4
3 1
3 2
3 3
4 1
Users table:
id name
--------
1 John
2 Alice
3 Bob
4 Mark
5 Jane
6 Mary
7 Ann
I need to fetch in single query:
always all features, regardless if they have votes or not
each feature must be listed only once regardless of number of votes even if it has no votes
votes count for each feature listed
a special mark if currently logged user voted for current feature in the list - for example if user 3 is logged in, then list all features with votes count for all users and if user 3 voted for some features have a special field indicating his vote or NULL if he didn't (other data from votes table must be included too so it needs to be LEFT JOINED)
So far I did this:
SELECT *,(SELECT COUNT(*) FROM votes WHERE votes.featureid=features.id) AS "votecnt"
FROM features
LEFT JOIN votes ON votes.featureid=features.id
LEFT JOIN users ON users.id=votes.userid
GROUP BY features.id
It lists all features but has no special field if user 3 voted. I tried with IF, various WHERE conditions and after a lot of tries... ran out of ideas.
Desired output might look like this:
features.id features.name votes.userid votes.otherfields users.id users.name
1 feature 1 4 - 4 Mark
2 feature 2 NULL - NULL NULL
3 feature 3 NULL - NULL NULL
4 feature 4 NULL - NULL NULL
5 feature 5 NULL - NULL NULL
All the features are listed and only those where user 4 voted have other joined tables filled, others are simply NULL. If someone else voted for feature 2 it is still NULL as it is of no relevance for user 4 because in this example user 4 is logged in.
Here is the problem:
http://sqlfiddle.com/#!2/c3d10/3
http://sqlfiddle.com/#!2/c3d10/4
http://sqlfiddle.com/#!2/c3d10/5
http://sqlfiddle.com/#!2/c3d10/6
http://sqlfiddle.com/#!2/c3d10/7
All of above queries in SQLFiddle it should output all 5 features regardless of the userid. So the query must be modified somehow to show all features - even if other people voted for a feature or if there are no votes or if current user voted for feature.
this should do it:
SELECT tmp1.id, name, votecnt, user_id, user_name FROM
(SELECT *,(SELECT COUNT(*) FROM votes WHERE votes.featureid=features.id) AS "votecnt" FROM features) as tmp1
LEFT JOIN (SELECT features.id as feature_id, users.id as user_id, users.name as user_name FROM features
JOIN votes ON votes.featureid=features.id
JOIN users ON users.id=votes.userid
WHERE users.id=3) as tmp2 on tmp1.id = tmp2.feature_id
probably it's not the prettiest sql, and most likely there's also room for optimization
I think you're asking to have all features returned, all counts returned, but only user information if they voted for that feature.
I get the results you're looking for if you specify the userid
select f.id
, f.name
, u.id
, u.name
, v.votecnt
from features f
left join (select featureid, COUNT(userid) votecnt from votes group by featureid) v on v.featureid = f.id
left join votes v1 on v1.featureid = f.id and v1.userid = 4
left join users u on u.id = v1.userid
I chose to specify the userid inside the left join to the votes. Anywhere else and it limits the total number of rows returned.
Results:
1 feature 1 4 Mark 4
2 feature 2 NULL NULL 3
3 feature 3 NULL NULL 3
4 feature 4 NULL NULL 2
5 feature 5 NULL NULL NULL
Example with "Bob"
select f.id
, f.name
, u.id
, u.name
, v.votecnt
from features f
left join (select featureid, COUNT(userid) votecnt from votes group by featureid)v on v.featureid = f.id
left join votes v1 on v1.featureid = f.id and v1.userid = 3
left join users u on u.id = v1.userid
Results:
1 feature 1 3 Bob 4
2 feature 2 3 Bob 3
3 feature 3 3 Bob 3
4 feature 4 NULL NULL 2
5 feature 5 NULL NULL NULL

Select the opposite row of a WHERE clause in a joined jquery result

I have a database with users and games. 1 game can have multiple users so I made a linking table called users_games. The crux is that a game can always only have 2 players since it is a board game. I know which player I am, i have my user_id and my email, but I would like to gain a result that gives me a list of all games I am in WITH the user_id and email of the other fellow. So a query that looks to all games I am in and give the other row, with the name of the player.
My tables:
games
id (int)
board (varchar) representation of the board
users
id (int)
email (varchar)
password (varchar MD5)
users_games
id (int)
user_id (int)
game_id (int)
For clarification this query
SELECT *
FROM `tic_users_games` AS ug
LEFT JOIN tic_users AS u
ON ug.user_id = u.id
RIGHT JOIN tic_games AS g
ON ug.game_id = g.id
And result
id user_id game_id id email password id board created updated
1 1 1 1 ME#gmail.com d56b699830e77ba53855679cb1d252da 1 0|0|0|0|0|0|0|0|0 2012-04-02 16:56:06 2012-04-02 16:56:06
2 2 1 2 FOE1#gmail.com d56b699830e77ba53855679cb1d252da 1 0|0|0|0|0|0|0|0|0 2012-04-02 16:56:06 2012-04-02 16:56:06
3 3 2 2 FOE2#gmail.com d56b699830e77ba53855679cb1d252da 2 0|0|0|0|0|0|0|0|0 2012-04-02 16:56:06 2012-04-02 16:56:06
4 1 2 1 ME#gmail.com d56b699830e77ba53855679cb1d252da 2 0|0|0|0|0|0|0|0|0 2012-04-02 16:56:06 2012-04-02 16:56:06
See: In the above case I just want 2 rows: game_id 1 and 2, with FOE1#... and FOE2#...
Thanks
This should do it for you. The syntax might not be exact for mysql but you should get there. Basically get all users that have games in the user_games table with the same game_ID as the games I am in:
SELECT
User_Games.Game_ID,
Users.ID,
Users.Email
FROM
Users
LEFT JOIN User_Games ON Users.ID = User_Games.User_Id
WHERE
Users.User_ID <> #yourUserID
AND EXISTS
(SELECT
NULL
FROM
User_Games AS MyUserGames
WHERE
User_Games.Game_ID = MyUserGames.Game_ID
AND MyUserGames.User_ID = #yourUserID)
Try this (where #userid represents the user you are searching games for, in this case, 1):
select *
from tic_users_games ug1
left join tic_users_games ug2 on ug1.game_id = ug2.game_id
left join tic_users u on ug2.user_id = u.id
right join tic_games g on ug2.game_id = g.id
where ug1.user_id = #userid
and ug2.user_id <> #userid
Demo: http://www.sqlfiddle.com/#!2/7b0f6/2