2016-11-29 42 views
3

我有一个类别表,我已将其解析为categories_to_categories关于many-to-many关系的另一个表categories_to_categories。 现在我想只选择那些具有类别Mysql-查询层次结构中仅达到第二级别的项目

  1. 要么PARENT_ID = 0
  2. 无论他们的父母的PARENT_ID = 0

这里是我的Fiddle,在结果集,我有2个问题,

  1. 结果复制
  2. 定制衬衫类不应该“T在那里,因为它已经有一个家长和其父没有parent_id =0

DDL:不工作

CREATE TABLE `categories` (
    `id` int(11) NOT NULL AUTO_INCREMENT, 
    `uuid` varchar(25) DEFAULT NULL, 
    `name` varchar(255) DEFAULT NULL, 
    `description` text, 
    `status` enum('ACTIVE','INACTIVE') DEFAULT 'ACTIVE', 
    `created_at` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, 
    `updated_at` timestamp NOT NULL DEFAULT '0000-00-00 00:00:00', 
    `banner_path` varchar(255) DEFAULT NULL, 
    `franchise_id` int(11) DEFAULT NULL, 
    PRIMARY KEY (`id`), 
    UNIQUE KEY `categories_uuid_uindex` (`uuid`) 
); 



INSERT INTO `categories` VALUES (1,'xyz','Printing','test','ACTIVE','2016-11-29 13:54:15','2016-11-29 13:54:18',NULL,3),(2,'abc','Digital','test','ACTIVE','2016-11-29 14:33:48','2016-11-29 14:33:53',NULL,3),(3,'def','Video','test','ACTIVE','2016-11-29 14:34:25','2016-11-29 14:34:29',NULL,3),(4,'s','Merchandise printing','test','ACTIVE','2016-11-29 14:35:02','2016-11-29 14:35:04',NULL,3),(5,'4','D/C','test','ACTIVE','2016-11-29 14:35:24','2016-11-29 14:35:27',NULL,3),(6,'2','Goods','test','ACTIVE','2016-11-29 14:35:49','2016-11-29 14:35:51',NULL,3),(8,'5','B/A Templates','test','ACTIVE','2016-11-29 14:36:26','2016-11-29 14:36:28',NULL,3),(9,'gggg','Customized Shirts','test','ACTIVE','2016-11-29 14:37:00','2016-11-29 14:37:01',NULL,3); 



CREATE TABLE `categories_to_categories` (
    `id` int(11) NOT NULL AUTO_INCREMENT, 
    `category_id` int(11) DEFAULT NULL, 
    `parent_id` int(11) DEFAULT NULL, 
    `created_at` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP, 
    `updated_at` timestamp NOT NULL DEFAULT '0000-00-00 00:00:00', 
    PRIMARY KEY (`id`) 
); 



INSERT INTO `categories_to_categories` VALUES (1,1,0,'2016-11-29 13:54:56','2016-11-29 13:54:59'),(2,2,0,'2016-11-29 14:38:10','2016-11-29 14:38:17'),(3,3,0,'2016-11-29 14:38:28','2016-11-29 14:38:29'),(4,4,1,'2016-11-29 14:38:48','2016-11-29 14:38:51'),(5,5,2,'2016-11-29 14:39:28','2016-11-29 14:39:30'),(6,6,0,'2016-11-29 14:39:41','2016-11-29 14:39:43'),(7,4,6,'2016-11-29 14:39:52','2016-11-29 14:39:55'),(8,7,1,'2016-11-29 14:40:11','2016-11-29 14:40:17'),(9,8,4,'2016-11-29 14:40:29','2016-11-29 14:40:32'),(10,9,2,'2016-11-29 14:40:40','2016-11-29 14:40:42'); 

查询:

SELECT * 
    FROM CATEGORIES      A 
INNER JOIN CATEGORIES_TO_CATEGORIES B 
    ON A.ID = B.ID 
AND B.PARENT_ID = 0 
    OR (SELECT PARENT_ID 
     FROM CATEGORIES_TO_CATEGORIES 
     WHERE ID = B.PARENT_ID 
    ) = 0 ; 
+2

请直接在问题中添加所有必要的信息。 –

+0

其实,'Customized Shirts'应该在结果集中 - 它只有一个父节点'Digital',它是根类别(parent_id = 0) – krlv

+0

从sql小提琴中添加的代码 – DForck42

回答

1

一个可能的解决方案是使用根类别和他们的孩子的联合:

SELECT c.id, c.name FROM categories c 
WHERE c.id IN (
    SELECT cc.category_id FROM categories_to_categories cc 
    WHERE cc.parent_id = 0 

    UNION 

    SELECT pc.category_id FROM categories_to_categories cc 
    LEFT JOIN categories_to_categories pc ON cc.category_id = pc.parent_id 
    WHERE cc.parent_id = 0 
) 
ORDER BY c.id; 

结果此查询将是根类别列表/没有重复子类别(“商品印刷”是两个“打印”和“商品”子节点,但在结果集中只能出现一次):

+----+----------------------+ 
| id | name     | 
+----+----------------------+ 
| 1 | Printing    | 
| 2 | Digital    | 
| 3 | Video    | 
| 4 | Merchandise printing | 
| 5 | D/C     | 
| 6 | Goods    | 
| 9 | Customized Shirts | 
+----+----------------------+