如何从我的查询结果中删除反向重复？

Question

我解决了 Lagunita Stanford 的 SQL 练习，我在第 3 题（SQL 社交网络查询练习）处停下来。我不知道如何从我的查询结果中删除反向重复：

    Gabriel    9  |  Cassandra  9
    ------------------------------
    Cassandra  9  |  Gabriel    9
    ------------------------------
    Jessica    11 |  Kyle       12
    ------------------------------
    Kyle       12 |  Jessica    11

答案应该是：

        Cassandra  9  |  Gabriel    9
        ------------------------------
        Jessica    11 |  Kyle       12

这是我的代码：

SELECT h1.name,
   h1.grade,
   h2.name,
   h2.grade          
FROM likes AS likes1
INNER JOIN likes AS likes2
   ON likes1.id1 = likes2.id
   AND likes1.id2 = likes2.id1
INNER JOIN highschooler  AS h1 ON likes1.id1 = h1.id
INNER JOIN highschooler  AS h2 ON likes2.id1 = h2.id

表： https://lagunita.stanford.edu/c4x/DB/SQL/asset/socialdata.html

而且我不能使用窗口函数和 cte 的。

我很好奇这是否可以删除同一查询中的重复项。

Answer 1

您当前形成的结果集包含每对两次，删除两行之一所需要做的就是应用WHERE 子句：

SELECT h1.name name1,
       h1.grade grade1,
       h2.name name2,
       h2.grade grade2
FROM likes AS likes1
INNER JOIN likes AS likes2 ON likes1.id1 = likes2.id2 AND likes1.id2 = likes2.id1
INNER JOIN highschooler  AS h1 ON likes1.id1 = h1.id
INNER JOIN highschooler  AS h2 ON likes2.id1 = h2.id
WHERE h1.name < h2.name 

也可以是：

WHERE likes1.id1 < likes1.id2  

如果有两个高中生同名的情况。
请参阅demo。
结果：

| name1     | grade1 | name2   | grade2 |
| --------- | ------ | ------- | ------ |
| Cassandra | 9      | Gabriel | 9      |
| Jessica   | 11     | Kyle    | 12     |