DDR爱好者之家 Design By 杰米

无限分类是个老话题了,来看看PHP结合Mysql如何实现。

第一种方法

这种方法是很常见、很传统的一种,先看表结构

表:category
id int 主键,自增
name varchar 分类名称
pid int 父类id,默认0
顶级分类的 pid 默认就是0了。当我们想取出某个分类的子分类树的时候,基本思路就是递归,当然,出于效率问题不建议每次递归都查询数据库,通常的做法是先讲所有分类取出来,保存到PHP数组里,再进行处理,最后还可以将结果缓存起来以提高下次请求的效率。

先来构建一个原始数组,这个直接从数据库中拉出来就行:

复制代码 代码如下:
$categories = array(
    array('id'=>1,'name'=>'电脑','pid'=>0),
    array('id'=>2,'name'=>'手机','pid'=>0),
    array('id'=>3,'name'=>'笔记本','pid'=>1),
    array('id'=>4,'name'=>'台式机','pid'=>1),
    array('id'=>5,'name'=>'智能机','pid'=>2),
    array('id'=>6,'name'=>'功能机','pid'=>2),
    array('id'=>7,'name'=>'超级本','pid'=>3),
    array('id'=>8,'name'=>'游戏本','pid'=>3),
);

目标是将它转化为下面这种结构

电脑
    笔记本
        超级本
        游戏本
    台式机
手机
    智能机
    功能机
用数组来表示的话,可以增加一个 children 键来存储它的子分类:

复制代码 代码如下:
array(
    //1对应id,方便直接读取
    1 => array(
        'id'=>1,
        'name'=>'电脑',
        'pid'=>0,
        children=>array(
            &array(
                'id'=>3,
                'name'=>'笔记本',
                'pid'=>1,
                'children'=>array(
                    //此处省略
                )
            ),
            &array(
                'id'=>4,
                'name'=>'台式机',
                'pid'=>1,
                'children'=>array(
                    //此处省略
                )
            ),
        )
    ),
    //其他分类省略
)

处理过程:

复制代码 代码如下:
$tree = array();
//第一步,将分类id作为数组key,并创建children单元
foreach($categories as $category){
    $tree[$category['id']] = $category;
    $tree[$category['id']]['children'] = array();
}
//第二部,利用引用,将每个分类添加到父类children数组中,这样一次遍历即可形成树形结构。
foreach ($tree as $k=>$item) {
    if ($item['pid'] != 0) {
        $tree[$item['pid']]['children'][] = &$tree[$k];
    }
}
print_r($tree);

打印结果如下:

复制代码 代码如下:
Array
(
    [1] => Array
        (
            [id] => 1
            [name] => 电脑
            [pid] => 0
            [children] => Array
                (
                    [0] => Array
                        (
                            [id] => 3
                            [name] => 笔记本
                            [pid] => 1
                            [children] => Array
                                (
                                    [0] => Array
                                        (
                                            [id] => 7
                                            [name] => 超级本
                                            [pid] => 3
                                            [children] => Array
                                                (
                                                )
                                        )
                                    [1] => Array
                                        (
                                            [id] => 8
                                            [name] => 游戏本
                                            [pid] => 3
                                            [children] => Array
                                                (
                                                )
                                        )
                                )
                        )
                    [1] => Array
                        (
                            [id] => 4
                            [name] => 台式机
                            [pid] => 1
                            [children] => Array
                                (
                                )
                        )
                )
        )
    [2] => Array
        (
            [id] => 2
            [name] => 手机
            [pid] => 0
            [children] => Array
                (
                    [0] => Array
                        (
                            [id] => 5
                            [name] => 智能机
                            [pid] => 2
                            [children] => Array
                                (
                                )
                        )
                    [1] => Array
                        (
                            [id] => 6
                            [name] => 功能机
                            [pid] => 2
                            [children] => Array
                                (
                                )
                        )
                )
        )
    [3] => Array
        (
            [id] => 3
            [name] => 笔记本
            [pid] => 1
            [children] => Array
                (
                    [0] => Array
                        (
                            [id] => 7
                            [name] => 超级本
                            [pid] => 3
                            [children] => Array
                                (
                                )
                        )
                    [1] => Array
                        (
                            [id] => 8
                            [name] => 游戏本
                            [pid] => 3
                            [children] => Array
                                (
                                )
                        )
                )
        )
    [4] => Array
        (
            [id] => 4
            [name] => 台式机
            [pid] => 1
            [children] => Array
                (
                )
        )
    [5] => Array
        (
            [id] => 5
            [name] => 智能机
            [pid] => 2
            [children] => Array
                (
                )
        )
    [6] => Array
        (
            [id] => 6
            [name] => 功能机
            [pid] => 2
            [children] => Array
                (
                )
        )
    [7] => Array
        (
            [id] => 7
            [name] => 超级本
            [pid] => 3
            [children] => Array
                (
                )
        )
    [8] => Array
        (
            [id] => 8
            [name] => 游戏本
            [pid] => 3
            [children] => Array
                (
                )
        )
)

优点:关系清楚,修改上下级关系简单。

缺点:使用PHP处理,如果分类数量庞大,效率也会降低。

第二种方法

这种方法是在表字段中增加一个path字段:

表:category
id int 主键,自增
name varchar 分类名称
pid int 父类id,默认0
path varchar 路径
示例数据:

id        name        pid        path
1         电脑        0          0
2         手机        0          0
3         笔记本      1          0-1
4         超级本      3          0-1-3
5         游戏本      3          0-1-3
path字段记录了从根分类到上一级父类的路径,用id+'-'表示。

这种方式,假设我们要查询电脑下的所有后代分类,只需要一条sql语句:

select id,name,path from category where path like (select concat(path,'-',id,'%') as path from category where id=1);
结果:

+----+-----------+-------+
| id | name      | path  |
+----+-----------+-------+
| 3  | 笔记本 | 0-1   |
| 4  | 超级本 | 0-1-3 |
| 5  | 游戏本 | 0-1-3 |
+----+-----------+-------+
这种方式也被很多人所采纳,我总结了下:

优点:查询容易,效率高,path字段可以加索引。

缺点:更新节点关系麻烦,需要更新所有后辈的path字段。

以上就是本文的全部内容了,两种方式,你喜欢哪种?希望大家能够喜欢。

DDR爱好者之家 Design By 杰米
广告合作:本站广告合作请联系QQ:858582 申请时备注:广告合作(否则不回)
免责声明:本站资源来自互联网收集,仅供用于学习和交流,请遵循相关法律法规,本站一切资源不代表本站立场,如有侵权、后门、不妥请联系本站删除!
DDR爱好者之家 Design By 杰米