sql - PostgreSQL - 选择条件满足的不同(列1，列2)

Question

我有下表和其中的一些示例记录:

  id  | attr1_id | attr2_id |      user_id      | rating_id |
------+----------+----------+-------------------+-----------+
 1    |      188 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="70050315022f4130141f1d11191e5e131f1d" rel="noreferrer noopener nofollow">[email protected]</a> |         3 |
 2    |      193 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="bcc9cfd9cee38efcd8d3d1ddd5d292dfd3d1" rel="noreferrer noopener nofollow">[email protected]</a> |         2 |
 3    |      193 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="84f1f7e1f6dbb6c4e0ebe9e5edeaaae7ebe9" rel="noreferrer noopener nofollow">[email protected]</a> |         1 |
 4    |      194 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="0d787e687f523f4d6962606c6463236e6260" rel="noreferrer noopener nofollow">[email protected]</a> |         1 |
 5    |      194 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="ee9b9d8b9cb1dfae8a81838f8780c08d8183" rel="noreferrer noopener nofollow">[email protected]</a> |         1 |
 6    |      192 |      201 | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="fe8b8d9b8ca1ccbe9a91939f9790d09d9193" rel="noreferrer noopener nofollow">[email protected]</a> |         1 |

(attr1_id、attr2_id、user_id) 的组合是UNIQUE，这意味着每个用户只能创建具有一对特定属性 ID 的一条记录。

我的目标是选择 (attr1_id, attr2_id) 的所有不同组合，其中 rating_id = 1，但仅选择 attr1_id 和 attr2_id 的每个组合code> 仅一次，且仅在不存在任何其他行(由其他用户)具有 rating_id > 1 且引用相同 attr1_id 和 的情况下attr2_id。 请注意，attr1_id 和 attr2_id 的组合可以互换，因此给出这两条记录:

  id  | attr1_id | attr2_id |      user_id       | rating_id | override_comment
------+----------+----------+--------------------+-----------+------------------
  20  |       5  |       2  | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="44313721361b7504202b29252d2a6a272b29" rel="noreferrer noopener nofollow">[email protected]</a>  |         3 |
------+----------+----------+--------------------+-----------+------------------
  21  |       2  |       5  | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="14616771664b2654707b79757d7a3a777b79" rel="noreferrer noopener nofollow">[email protected]</a>  |         1 |

不应计算任何行，因为这些行引用 attr_ids 的相同组合，并且其中之一的 rating_id > 1。

但是，如果这两行存在:

  id  | attr1_id | attr2_id |      user_id       | rating_id | override_comment
------+----------+----------+--------------------+-----------+------------------
  20  |       5  |       2  | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="52272137200d6312363d3f333b3c7c313d3f" rel="noreferrer noopener nofollow">[email protected]</a>  |         1 |
------+----------+----------+--------------------+-----------+------------------
  21  |       2  |       5  | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="d0a5a3b5a28fe290b4bfbdb1b9befeb3bfbd" rel="noreferrer noopener nofollow">[email protected]</a>  |         1 |
------+----------+----------+--------------------+-----------+------------------
  22  |       2  |       5  | <a href="https://stackoverflow.com/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="295c5a4c5b761a694d4644484047074a4644" rel="noreferrer noopener nofollow">[email protected]</a>  |         1 |

所有行只能算作一行，因为它们都共享相同的 attr1_id 和 attr2_id 组合，并且都具有 rating_id = 1 .

此外，还有一些通过连接表列进行的连接和过滤，我将省略这些操作，但我想我还是要提及它。

SQL Fiddle 现在对我不起作用，但我已经 uploaded some sample data from the compatibility table.

到目前为止我的查询是这样的:

SELECT distinct(a1, a2),
       a1,
       a2
FROM
  ( SELECT c.*,
           least(attr1_id, attr2_id) AS a1,
           greatest(attr1_id, attr2_id) AS a2
   FROM compatibility c
   JOIN attribute a ON c.attr1_id = a.id
   JOIN PARAMETER pa ON a.parameter_id = pa.id
   JOIN problem p ON pa.problem_id = p.id
   WHERE p.id = 1
   GROUP BY 1,
            2 HAVING NOT bool_or(rating_id > 1)) s;

样本中共有 144 个评分。每个用户都创建了 7 个评分，其 rating_id > 1 且在这 14 个评分中，有 2 个引用同一组 (attr1_id,attr2_id )。 因此，我要查找的数字为 (77-12) = 65。然而，这里的结果似乎是77-2 = 75。因此，只有存在两个具有相同属性 ID 的评级的行才会被丢弃。

我还要指出my previous question for this matter我被要求开一个新的。

Answer 1

我认为这符合您的描述:

select least(attr1_id, attr2_id) as attr1, greatest(attr1_id, attr2_id) as attr2
from table t
group by least(attr1_id, attr2_id), greatest(attr1_id, attr2_id) 
having bool_and(rating_d = 1) ;

我不理解您查询中的其他表，因为您从一个包含您需要的所有内容的表开始。