Question

感谢您的精力充沛。

我有两种表用于某种社交网络，表A用于内容，表B用于喜欢：

表A：

CREATE TABLE `A` (
`id` int(10) unsigned NOT NULL,
  `id_user` int(10) unsigned NOT NULL,
  `title` varchar(255) DEFAULT NULL,
  `path` text,
  `ext` varchar(15) DEFAULT NULL,
  `image_type` varchar(100) DEFAULT NULL,
  `link_url` text,
  `creation_date` datetime DEFAULT NULL,
  `size` float DEFAULT NULL,
  `type` int(10) unsigned DEFAULT NULL,
  `number_share` int(10) unsigned NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;


ALTER TABLE `A`
 ADD PRIMARY KEY (`id`), ADD KEY `id_user` (`id_user`);

表B：

CREATE TABLE B (
  `id_b` int(10) unsigned NOT NULL,
  `id_user` int(10) unsigned NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;


ALTER TABLE `B`
 ADD PRIMARY KEY (`id_b`,`id_user`), ADD KEY `id_user` (`id_user`);

我想用这个命令从mysql记录中检索：

    1st most popular 
     1st most recent
     2nd most popular 
     2nd most recent



    .....

任何人都可以告诉我如何通过mysql（如果可能）或易于使用的东西组合这个不是2个不同的查询吗？

编辑1：

如果我有这种情况：

首选（流行）的结果： ID 3 五 12 4 65 2

和

第二次选择（最近）的结果：

我将打印我的元素宽度id：

我希望第二个查询中的id = 3不被考虑，因为已经发布了

谢谢！

Answer 1

<强>计划

使用变量，按顺序排列人气

使用变量，并按排序最近排名

union all

由rnk，cat
重新订购
按ID分组以限制为第一个ID

输入[样本]

+----+---------+--------+--------+--------+------------+----------+--------------------------+--------+--------+--------------+
| id | id_user | title  |  path  |  ext   | image_type | link_url |      creation_date       |  size  |  type  | number_share |
+----+---------+--------+--------+--------+------------+----------+--------------------------+--------+--------+--------------+
|  1 |      22 | (null) | (null) | (null) | (null)     | (null)   | August, 09 2015 10:54:57 | (null) | (null) |            0 |
|  2 |      33 | (null) | (null) | (null) | (null)     | (null)   | August, 09 2015 07:54:57 | (null) | (null) |            0 |
|  3 |      44 | (null) | (null) | (null) | (null)     | (null)   | August, 09 2015 08:54:57 | (null) | (null) |            0 |
+----+---------+--------+--------+--------+------------+----------+--------------------------+--------+--------+--------------+

+------+---------+
| id_b | id_user |
+------+---------+
|    3 |       8 |
|    3 |       9 |
|    3 |      21 |
|    3 |      22 |
|    1 |      55 |
|    3 |      55 |
|    1 |      66 |
|    1 |      77 |
+------+---------+

<强>查询

set @rnk_pop := 0;
set @rnk_rec := 0;

-- popularity ranking
select cat, id, rnk
from
(
select cat, id, rnk
from
(
  select cat, id, occurs, @rnk_pop := @rnk_pop + 1 as rnk
  from
  (
    select 'popularity' as cat, A.id, count(B.id_b) as occurs
    from A
    left join B
    on B.id_b = A.id
    group by A.id
    order by occurs desc
  ) q1
) pop_sort
union all
-- recent ranking
select cat, id, rnk
from
(
  select 'recent' as cat, id, @rnk_rec := @rnk_rec + 1 as rnk
  from A
  order by creation_date desc
) rec_sort
order by rnk, cat
) all_q
group by id
order by rnk, cat
;

<强>输出

+------------+----+-----+
|    cat     | id | rnk |
+------------+----+-----+
| popularity |  3 |   1 |
| recent     |  1 |   1 |
| popularity |  2 |   3 |
+------------+----+-----+

sqlfiddle

备注

上面的排名输出将用于加入以显示对内容有用的任何信息..

mysql从同一个表上的2个查询获得混合结果，其中包含不同的顺序

1 个答案: