主要问题是关系部分的'WHERE IN',它将找到的行数相乘 例如,我添加了GROUP_CONCAT以获得更好的视图:
SELECT COUNT(*),
GROUP_CONCAT(sections.name SEPARATOR '; ')
FROM users
LEFT JOIN users_sections ON (users.id = users_sections.user_id)
LEFT JOIN sections ON (users_sections.section_id = sections.id)
GROUP BY users.id
Finds:
'2', 'Registered; admins'
'3', 'Registered; admins; dudes'
'1', 'Registered'
有3行,这是正确的,但结果没用 现在添加WHERE IN部分:
SELECT COUNT(*),
GROUP_CONCAT(sections.name SEPARATOR '; ')
FROM users
LEFT JOIN users_sections ON (users.id = users_sections.user_id)
LEFT JOIN sections ON (users_sections.section_id = sections.id)
WHERE sections.id IN (2,3)
GROUP BY users.id
Finds:
'2', 'Registered; admins'
'2', 'Registered; admins'
'1', 'Registered'
即使是正确但无用的。我需要一个单独的'3'作为COUNT个结果 好吧,让我们删除GROUP_CONCAT并...祈祷
SELECT COUNT(*)
FROM users
LEFT JOIN users_sections ON (users.id = users_sections.user_id)
LEFT JOIN sections ON (users_sections.section_id = sections.id)
WHERE sections.id IN (2,3)
GROUP BY users.id
Find:
'2'
'2'
'1'
和以前一样。让我们试试这个:
SELECT COUNT(*) > 0
...
Find:
'1'
'1'
'1'
嘿,看起来不错!让我们试试(希望最终):
SELECT SUM(COUNT(*) > 0)
...
Find:
Invalid use of group function
太伤心了...... 我怎么能得到伯爵,或者不可能?
答案 0 :(得分:2)
您想要做什么(如果我正确阅读)是计算特定条件适用的不同用户ID。我无法从中得到真正的问题,但我认为你需要它:
SELECT
COUNT( DISTINCT users.id)
FROM
users
LEFT JOIN users_sections ON (users.id = users_sections.user_id)
LEFT JOIN sections ON (users_sections.section_id = sections.id)
WHERE
sections.id IN (2,3)