我想按类别名称统计给定月份的项目状态更新/更改
我有一个类别(类别列表)、投资(投资者投资列表，包括类别中的category_id)、状态表，其中包含预实现、实现和运营状态的名称。 项目包含项目 ID、名称、开始日期字段。 状态和项目都有多对多的关系，称为project_status表包含project_id、status_id、date_of_progress。
我想统计本月内按类别名称更新/更改其 status_id 的所有项目
从实现前到实现 从实现到运营 从前期实现到运营
CREATE TABLE `categories` (
`cat_id` int(11) NOT NULL,
`name` varchar(50) NOT NULL,
`description` varchar(255) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
INSERT INTO `categories` (`cat_id`, `name`, `description`) VALUES
(1, 'agriculture', ''),
(2, 'manufactures ', ''),
(3, 'Technology', ''),
(4, 'services', '');
CREATE TABLE `investments` (
`investment_id` int(11) NOT NULL,
`investment_name` varchar(100) NOT NULL,
`cat_id` int(11) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
INSERT INTO `investments` (`investment_id`, `investment_name`, `cat_id`) VALUES
(1, 'Company A', 3),
(2, 'Company B', 4),
(3, 'Company C', 1),
(4, 'Company D', 2),
(5, 'Company E', 3),
(6, 'Company F', 4),
(7, 'Company G', 1),
(8, 'Company H', 2);
CREATE TABLE `status` (
`status_id` int(11) NOT NULL,
`status_name` varchar(30) NOT NULL,
`description` varchar(255) DEFAULT NULL,
`status_name_tg` varchar(50) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
INSERT INTO `status` (`status_id`, `status_name`, `description`, `status_name_tg`) VALUES
(1, 'Pre Implementation', 'Operational', 'Pre Implementation'),
(2, 'Implementation', NULL, 'Implementation'),
(3, 'Operational', NULL, 'Operational'),
(4, 'Inactive', NULL, 'Inactive'),
(5, 'Cancellation', NULL, 'Cancellation');
CREATE TABLE `project_status` (
`project_status_id` int(11) NOT NULL,
`status_id` int(11) NOT NULL,
`time` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP,
`project_id` int(11) NOT NULL,
`reason_for_cancellation` varchar(255) DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
INSERT INTO `project_status` (`project_status_id`, `status_id`, `time`, `project_id`, `reason_for_cancellation`) VALUES
(1, 1, '2019-02-24 21:51:50', 1, NULL),
(2, 2, '2019-03-26 21:52:57', 1, NULL),
(3, 1, '2019-04-30 21:57:57', 2, NULL),
(4, 1, '2019-05-26 22:04:08', 3, NULL),
(5, 3, '2019-08-24 22:06:36', 1, NULL),
(6, 2, '2019-08-11 22:07:05', 3, NULL),
(8, 1, '2019-08-01 00:14:41', 6, NULL),
(9, 1, '2019-08-09 12:11:22', 7, NULL),
(10, 1, '2019-08-09 12:15:22', 8, NULL),
(11, 3, '2019-08-14 10:07:49', 7, NULL),
(12, 2, '2019-08-14 10:10:45', 8, NULL),
(13, 2, '2019-08-26 17:16:02', 6, NULL),
(14, 3, '2019-08-26 17:16:02', 6, NULL);
CREATE TABLE `projects` (
`project_id` int(11) NOT NULL,
`name` varchar(150) NOT NULL,
`start_date` datetime NOT NULL,
`investment_id` int(11) NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
INSERT INTO `projects` (`project_id`, `name`, `start_date`, `investment_id`) VALUES
(1, 'site A', '2019-02-01 00:00:00', 1),
(2, 'site B', '2019-03-12 00:00:00', 2),
(3, 'Site C', '2019-04-04 00:00:00', 3),
(4, 'Site D', '2019-05-03 00:00:00', 4),
(5, 'site E', '2019-06-01 00:00:00', 5),
(6, 'site F', '2019-08-02 00:00:00', 6),
(7, 'Site J', '2019-08-01 00:00:00', 7),
(8, 'Site H', '2019-08-05 00:00:00', 8);
以下是2019年8月的结果
+--------------+-------------------+---------------+---------------+
| category_name| pre_implemntation | imp_operation | pre_operation |
+------+-------+------------+-------------++-------------++--------+
| agriculture | 1 | - | 1 |
| manufactures | 1 | - | - |
| Technology | - | 1 | - |
| services | - | 1 | - |
+--------------+-------------------+---------------+---------------+
最佳答案
<强> SQL DEMO
SELECT q.year
, q.month
, q.name
, COUNT(CASE WHEN q.last_status = 2 THEN 1 END) as pre_implementation
, COUNT(CASE WHEN q.last_status = 3
AND q.previous_status = 2 THEN 1 END) as implementation_operation
, COUNT(CASE WHEN q.last_status = 3
AND (q.previous_status = 1 or q.previous_status IS NULL)
THEN 1 END) as pre_operation
FROM (
SELECT *, (SELECT MAX(p.status_id)
FROM project_status p
WHERE p.time < CONCAT(t.year,'/', t.month,'/1')
AND p.project_id = t.project_id
) as previous_status
FROM (
SELECT YEAR(time) as year
, MONTH(time) as month
, c.name
, ps.project_id
, MAX(status_id) as last_status
FROM project_status ps
JOIN projects p
ON ps.project_id = p. project_id
JOIN investments i
ON p.investment_id = i.investment_id
JOIN categories c
ON i.cat_id = c.cat_id
GROUP BY YEAR(time), MONTH(time), c.name, ps.project_id
) t
) q
GROUP BY q.year, q.month, q.name
输出
+--------+-------+---------------+--------------------+--------------------------+---------------+
| year | month | name | pre_implementation | implementation_operation | pre_operation |
+--------+-------+---------------+--------------------+--------------------------+---------------+
| 2019 | 2 | Technology | 0 | 0 | 0 |
| 2019 | 3 | Technology | 1 | 0 | 0 |
| 2019 | 4 | services | 0 | 0 | 0 |
| 2019 | 5 | agriculture | 0 | 0 | 0 |
+--------+-------+---------------+--------------------+--------------------------+---------------+
| 2019 | 8 | Technology | 0 | 1 | 0 |
| 2019 | 8 | agriculture | 1 | 0 | 1 |
| 2019 | 8 | services | 0 | 0 | 1 |
| 2019 | 8 | manufactures | 1 | 0 | 0 |
+--------+-------+---------------+--------------------+--------------------------+---------------+
关于mysql - 我如何计算特定月份按类别更改/更新的状态，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/57800331/