链表的反转是一个很常见、很基础的数据结构题,输入一个单向链表,输出逆序反转后的链表,如图:上面的链表转换成下面的链表。实现链表反转有两种方式,一种是循环迭代,另外一种方式是递归。
第一种方式:循坏迭代
循坏迭代算法需要三个临时变量:pre、head、next,临界条件是链表为None或者链表就只有一个节点。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
# encoding: utf-8
class Node( object ):
def __init__( self ):
self .value = None
self . next = None
def __str__( self ):
return str ( self .value)
def reverse_loop(head):
if not head or not head. next :
return head
pre = None
while head:
next = head. next # 缓存当前节点的向后指针,待下次迭代用
head. next = pre # 这一步是反转的关键,相当于把当前的向前指针作为当前节点的向后指针
pre = head # 作为下次迭代时的(当前节点的)向前指针
head = next # 作为下次迭代时的(当前)节点
return pre # 返回头指针,头指针就是迭代到最后一次时的head变量(赋值给了pre)
|
测试一下:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
|
if __name__ = = '__main__' :
three = Node()
three.value = 3
two = Node()
two.value = 2
two. next = three
one = Node()
one.value = 1
one. next = two
head = Node()
head.value = 0
head. next = one
newhead = reverse_loop(head)
while newhead:
print (newhead.value, )
newhead = newhead. next
|
输出:
1
2
3
4
5
|
3
2
1
0
2
|
第二种方式:递归
递归的思想就是:
1
2
3
4
5
|
head. next = None
head. next . next = head. next
head. next . next . next = head. next . next
...
...
|
head的向后指针的向后指针转换成head的向后指针,依此类推。
实现的关键步骤就是找到临界点,何时退出递归。当head.next为None时,说明已经是最后一个节点了,此时不再递归调用。
1
2
3
4
5
6
7
|
def reverse_recursion(head):
if not head or not head. next :
return head
new_head = reverse_recursion(head. next )
head. next . next = head
head. next = None
return new_head
|
以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持服务器之家。