1

到目前为止，我有这个：

SELECT connections.id,
       connections.word_id,
       connections.order_id,
       connections.top,
       connections.deleted,
       (UNIX_TIMESTAMP() - UNIX_TIMESTAMP(connections.modified)) AS modified_ago,
       words_en.word,
       (CASE WHEN words_en.user_id = 1 THEN "You" ELSE words_en.user_id END) AS created_by
FROM connections, words_en
WHERE connections.word_id = words_en.id AND connections.user_id = 1
ORDER BY connections.order_id

我想添加一些东西，但我无法使语法正确。

  1. (CASE WHEN words_en.user_id = 1 THEN "You" ELSE words_en.user_id END)在这里，当 ELSE 时，我想name_surname从用户表中获取该用户的列，而不是 ID。所以我需要做另一个加入吗？但既然是在案例之内，我该怎么做呢？

  2. connections.word_id从选择。除了 ID，我还想知道这个 word_id 在连接表中出现了多少次。但从列表中排除当前用户。

我希望解释足够清楚。

4

2 回答 2

0 
SELECT connections.id,
       connections.word_id,
       connections.order_id,
       connections.top,
       connections.deleted,
       (UNIX_TIMESTAMP() - UNIX_TIMESTAMP(connections.modified)) AS modified_ago,
       words_en.word,
       (CASE WHEN words_en.user_id = 1 THEN "You" ELSE users.name_surname END) AS created_by
FROM connections JOIN words
WHERE connections.word_id = words_en.id AND connections.user_id = 1 JOIN users ON users.id=words_en.user_id
ORDER BY connections.order_id

我无法从您的回答中看出如何链接用户表，发布表模式。只是一个疯狂的猜测。

于 2013-11-02T11:10:46.033 回答
0

首先，使用 ANSI 连接重写您的查询，并添加表别名，如下所示：

SELECT c.id,
       c.word_id,
       c.order_id,
       c.top,
       c.deleted,
       (UNIX_TIMESTAMP() - UNIX_TIMESTAMP(c.modified)) AS modified_ago,
       words_en.word,
       (CASE WHEN words_en.user_id = 1 THEN "You" ELSE w.user_id END) AS created_by
FROM connections c
JOIN words_en w ON c.word_id = w.id
WHERE c.user_id = 1
ORDER BY c.order_id

现在扩展这个查询变得更容易了：要通过 引入用户w.user_id，添加另一个连接：

SELECT c.id,
       c.word_id,
       c.order_id,
       c.top,
       c.deleted,
       (UNIX_TIMESTAMP() - UNIX_TIMESTAMP(c.modified)) AS modified_ago,
       words_en.word,
       (CASE WHEN words_en.user_id = 1 THEN "You" ELSE u. name_surname END) AS created_by
FROM connections c
JOIN words_en w ON c.word_id = w.id
JOIN users u ON w.user_id = u.id
WHERE c.user_id = 1
ORDER BY c.order_id

要添加计数，请使用子查询，如下所示：

SELECT c.id,
       c.word_id,
       c.order_id,
       c.top,
       c.deleted,
       (UNIX_TIMESTAMP() - UNIX_TIMESTAMP(c.modified)) AS modified_ago,
       words_en.word,
       (CASE WHEN words_en.user_id = 1 THEN "You" ELSE u. name_surname END) AS created_by,
       (
           SELECT COUNT(*)
           FROM connections cc
           WHERE cc.word_id=c.word_id     -- It's the same word
             AND cc.user_id <> c.user_id  -- user by a different user
       ) as uses_by_others
FROM connections c
JOIN words_en w ON c.word_id = w.id
JOIN users u ON w.user_id = u.id
WHERE c.user_id = 1
ORDER BY c.order_id
于 2013-11-02T11:10:56.563 回答