Question

Consider the following tables:

users                    messages
-------------------      ----------------------
user_id avg_quality      msg_id user_id quality
-------------------      ----------------------
1                        1      1       1
2                        2      1       0
3                        3      1       0
                         4      1       1
                         5      1       1
                         6      2       0
                         7      2       0
                         8      3       1

messages.quality is either 0 or 1. I need to calculate the average message quality per user and update users.avg_quality accordingly. So the desired output would be modified users table like so:

users
-------------------
user_id avg_quality  <-- DECIMAL (8,2)
-------------------
1       0.60         <-- (3x1 + 2x0) / 5
2       0.00         <-- (2x0) / 2
3       1.00         <-- (1x1) / 1

I’ve begun my query like this, I know the syntax is incorrect but have no better idea. Do you?

UPDATE messages m, users u
SET avg_quality = (SELECT COUNT(m.msg_id) / SUM(m.quality))
WHERE m.user_id = u.user_id

Answer 1

This should work:

UPDATE users u
       INNER JOIN (SELECT a.user_id, AVG(quality) avg_quality 
                   FROM messages a
                        INNER JOIN users b
                            ON a.user_id = b.user_id
                   GROUP BY a.user_id
                  ) tmp
                  ON u.user_id = tmp.user_id
SET u.avg_quality = tmp.avg_quality;