Count unique rows mysql（使用返回非不同值的Count）

我被卡住了，我试图从数据库中获取统计数据

我的查询需要返回连接表中至少有1个条目的用户计数

所以我试着：

SELECT DISTINCT (u.id) AS total_has_atleast_one_word FROM users AS u
LEFT JOIN connections AS c
ON u.id = c.user_id
WHERE c.word_id IS NOT null;

这将返回正确的用户id，我有3行正确的id，这一切都很好

但当我计算u.id时，它返回35，应该是3。我的理解是，这是在计算不明显的行数。那我该怎么办

作为我问题的最后一部分，我如何将其与我的其他统计查询结合起来

/*SELECT COUNT(u.id) AS total_users,
       sum(u.created < (NOW() - INTERVAL 7 DAY)) as total_seven_day_period,
       sum(u.verified = 1) as total_verified,
       sum(u.level = 3) as total_passed_tut,
       sum(u.avatar IS NOT null) as total_with_avatar,
       sum(u.privacy = 0) as total_private,
       sum(u.privacy = 2) as total_friends_only,
       sum(u.privacy = 3) as total_public,
       sum(u.sex = "F") as total_female,
       sum(u.sex = "M") as total_male
FROM users AS u;*/

测试场：

第一部分：

/* user ids of those who have ever connected */
Select user_id from connections group by user_id

/* to get those who have connected after a particular date time ... */
Select user_id from connections group by user_id where connection_time > '2013-11-23' 

/* join with user table to get user details e.g. .. */   
Select u.name , u.address from user u
join on (Select user_id from connections group by user_id) c on c.user_id =u.user_id

---

这个查询也有同样的问题。是的，这更简单，但我仍然需要将它与我的第二个查询结合起来。给出示例数据、预期结果和实际结果。给你：应该返回4。因为有4个用户在连接表中至少有一个单词。我写这个答案的时候一定是睡着了。修正了，所以我把它当作子查询？选择COUNTDISTINCT user\u id FROM connections作为总计\u至少\u一个单词

/* user ids of those who have ever connected */
Select user_id from connections group by user_id

/* to get those who have connected after a particular date time ... */
Select user_id from connections group by user_id where connection_time > '2013-11-23' 

/* join with user table to get user details e.g. .. */   
Select u.name , u.address from user u
join on (Select user_id from connections group by user_id) c on c.user_id =u.user_id