在实际工作中,一个递归方法被重复调用的话,如果使用的static方式,由于上一次的数据并不会消失,第二次调用会造成数据的混乱。所以出现这种情况,需要换一种方式来解决。这里用的&引用传递来代替上一种方法。
下面是网上的一个例子,贴出来大家看下:
$items
=
array
(
1 =>
array
(
'id'
=> 1,
'pid'
=> 0,
'name'
=>
'江西省'
),
2 =>
array
(
'id'
=> 2,
'pid'
=> 0,
'name'
=>
'黑龙江省'
),
3 =>
array
(
'id'
=> 3,
'pid'
=> 1,
'name'
=>
'南昌市'
),
4 =>
array
(
'id'
=> 4,
'pid'
=> 2,
'name'
=>
'哈尔滨市'
),
5 =>
array
(
'id'
=> 5,
'pid'
=> 2,
'name'
=>
'鸡西市'
),
6 =>
array
(
'id'
=> 6,
'pid'
=> 4,
'name'
=>
'香坊区'
),
7 =>
array
(
'id'
=> 7,
'pid'
=> 4,
'name'
=>
'南岗区'
),
8 =>
array
(
'id'
=> 8,
'pid'
=> 6,
'name'
=>
'和兴路'
),
9 =>
array
(
'id'
=> 9,
'pid'
=> 7,
'name'
=>
'西大直街'
),
10 =>
array
(
'id'
=> 10,
'pid'
=> 8,
'name'
=>
'东北林业大学'
),
11 =>
array
(
'id'
=> 11,
'pid'
=> 9,
'name'
=>
'哈尔滨工业大学'
),
12 =>
array
(
'id'
=> 12,
'pid'
=> 8,
'name'
=>
'哈尔滨师范大学'
),
13 =>
array
(
'id'
=> 13,
'pid'
=> 1,
'name'
=>
'赣州市'
),
14 =>
array
(
'id'
=> 14,
'pid'
=> 13,
'name'
=>
'赣县'
),
15 =>
array
(
'id'
=> 15,
'pid'
=> 13,
'name'
=>
'于都县'
),
16 =>
array
(
'id'
=> 16,
'pid'
=> 14,
'name'
=>
'茅店镇'
),
17 =>
array
(
'id'
=> 17,
'pid'
=> 14,
'name'
=>
'大田乡'
),
18 =>
array
(
'id'
=> 18,
'pid'
=> 16,
'name'
=>
'义源村'
),
19 =>
array
(
'id'
=> 19,
'pid'
=> 16,
'name'
=>
'上坝村'
),
);
function
genTree(
$items
) {
$tree
=
array
();
//格式化好的树
foreach
(
$items
as
$item
)
if
(isset(
$items
[
$item
[
'pid'
]]))
$items
[
$item
[
'pid'
]][
'son'
][] = &
$items
[
$item
[
'id'
]];
else
$tree
[] = &
$items
[
$item
[
'id'
]];
return
$tree
;
}
打印结果,太长了,只贴一部分。
Array
(
[0] => Array
(
[id] => 1
[pid] => 0
[name] => 江西省
[son] => Array
(
[0] => Array
(
[id] => 3
[pid] => 1
[name] => 南昌市
)
[1] => Array
(
[id] => 13
[pid] => 1
[name] => 赣州市
[son] => Array
(
[0] => Array
(
[id] => 14
[pid] => 13
[name] => 赣县
[son] => Array
(
[0] => Array
(
[id] => 16
[pid] => 14
[name] => 茅店镇
[son] => Array
(
[0] => Array
(
[id] => 18
[pid] => 16
[name] => 义源村
)
[1] => Array
(
[id] => 19
[pid] => 16
[name] => 上坝村
)
)
)
[1] => Array
(
[id] => 17
[pid] => 14
[name] => 大田乡
)
)
)
[1] => Array
(
[id] => 15
[pid] => 13
[name] => 于都县
)
)
)
)
)
[1] => Array
(
)
)*/
第一次看到这个方法时我没看懂,今天回答贴子的问题时,也尝试了用&,才明白。
&符号:$a=&$b , $b变量的内存地址与$a变量的内存地址,指向了同一个地方.
这样,已经生成的数组,会随着后面值的变化不断变化,比如$arr['son'] = &$array, 虽然此次循环已经完毕,但是$arr['son'] 指向了$array同一个地方,再后面的循环中,会随着$array值的变化,随时随地的变化。由此顶级分类不断的壮大,最终组装成想要的结构。形成了数据的同步。
为了更好的说明&的作用。下面是我刚作的一个例子。
$data[] = array('id'=>"1601");
$data[] = array('id'=>"160101");
$data[] = array('id'=>"160102");
$data[] = array('id'=>"16010201");
$data[] = array('id'=>"1601020101");
$data[] = array('id'=>"1601020102");
$data[] = array('id'=>"160103");
$data[] = array('id'=>"16010301");
foreach ($data as $k => $v) {
$arr[ $v['id'] ] = $v;
}
foreach ($arr as $k => &$v) {
for($i=1;$i<100;$i++)
{
$oo = $i<10?'0'.$i:$i;
$sub_k = $k.$oo;
if( isset( $arr[$sub_k] ) )
{
$v['child'][$sub_k] = &$arr[$sub_k];//这里使用了&,数据实时同步。
$temp[] = $sub_k;
}
}
}
//如测试&注释掉下面的
/*for($i=0;$i<count($temp);$i++){unset($arr[$temp[$i]]);
}*/
如果不用&,会啥样呢?
Array ( [1601] => Array ( [id] => 1601 [child] => Array ( [160101] => Array ( [id] => 160101 ) [160102] => Array ( [id] => 160102 ) [160103] => Array ( [id] => 160103 ) ) ) [160101] => Array ( [id] => 160101 ) [160102] => Array ( [id] => 160102 [child] => Array ( [16010201] => Array ( [id] => 16010201 ) ) ) [16010201] => Array ( [id] => 16010201 [child] => Array ( [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) ) ) [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) [160103] => Array ( [id] => 160103 [child] => Array ( [16010301] => Array ( [id] => 16010301 ) ) ) [16010301] => Array ( [id] => 16010301 ) )
加上&:
Array ( [1601] => Array ( [id] => 1601 [child] => Array ( [160101] => Array ( [id] => 160101 ) [160102] => Array ( [id] => 160102 [child] => Array ( [16010201] => Array ( [id] => 16010201 [child] => Array ( [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) ) ) ) ) [160103] => Array ( [id] => 160103 [child] => Array ( [16010301] => Array ( [id] => 16010301 ) ) ) ) ) [160101] => Array ( [id] => 160101 ) [160102] => Array ( [id] => 160102 [child] => Array ( [16010201] => Array ( [id] => 16010201 [child] => Array ( [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) ) ) ) ) [16010201] => Array ( [id] => 16010201 [child] => Array ( [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) ) ) [1601020101] => Array ( [id] => 1601020101 ) [1601020102] => Array ( [id] => 1601020102 ) [160103] => Array ( [id] => 160103 [child] => Array ( [16010301] => Array ( [id] => 16010301 ) ) ) [16010301] => Array ( [id] => 16010301 ) )
发现,加上&后,
结果:
这样就通过引用数据同步的原理,同样实现了递归的效果,而且引用的效率要比递归高。
有兴趣的朋友可以,拷贝下来,去掉&,看看是什么结果。加上&又是什么结果。
end;