CC--Q2.8

2.8 Loop Detection: Given a circular linked list, implement an algorithm that returns the node at the beginning of the loop.

DEFINITION
Circular linked list: A (corrupt) linked list in which a node's next pointer points to an earlier node, so as to make a loop in the linked list
EXAMPLE
Input: A -> B -> C - > D -> E -> C [the same C as earlier)
Output: C

This is a modification of a classic interview problem: detect if a linked list has a loop. Let's apply the Pattern Matching approach.

Part 1: Detect If Linked List Has A Loop

An easy way to detect if a linked list has a loop is through the FastRunner / SlowRunner approach. FastRunner moves two steps at a time, while SlowRunner moves one step. Much like two cars racing around a track at different steps, they must eventually meet.

An astute reader may wonder if FastRunner might "hop over" SlowRunner completely, without ever colliding. That's not possible. Suppose that FastRunner did hop over SlowRunner, such that SlowRunner is at spot i and FastRunner is at spot i + 1. In the previous step, SlowRunner would be at spot i - 1and FastRunner would at spot ((i + 1) - 2),or spot i - 1.That is, they would have collided.

Part 2: When Do They Collide?

Let's assume that the linked list has a "non-looped" part of size k.

If we apply our algorithm from part 1, when will FastRunner and SlowRunner collide?

We know that for every p steps that SlowRunner takes, FastRunner has taken 2p steps. Therefore, when SlowRunner enters the looped portion after k steps, FastRunner has taken 2k steps total and must be 2k - k steps, or k steps, into the looped portion. Since k might be much larger than the loop length, we should actually write this as mod ( k, LOOP_SIZE) steps, which we will denote as K.

At each subsequent step, FastRunner and SlowRunner get either one step farther away or one step closer, depending on your perspective. That is, because we are in a circle, when A moves q steps away from B, it is also moving q steps closer to B.

So now we know the following facts:

  1. SlowRunner is 0 steps into the loop.
  2. FastRunner is K steps into the loop.
  3. SlowRunner is K steps behind FastRunner.
  4. FastRunner is LOOP_SIZE - K steps behind SlowRunner.
  5. FastRunner catches up to SlowRunner at a rate of 1 step per unit of time.

So, when do they meet? Well, if FastRunner is LOOP_SIZE - K steps behind SlowRunner, and FastRunner catches up at a rate of 1 step per unit of time, then they meet after LOOP_SIZE - K steps. At this point, they will be K steps before the head of the loop. Let's call this point CollisionS pot.

Paste_Image.png
Part 3: How Do You Find The Start of the Loop?

We now know that CollisionSpot is K nodes before the start of the loop. Because K = mod (k, LOOP_SIZE) (or, in other words, k = K + M * LOOP_SIZE, for any integer M), it is also correct to say that it is k nodes from the loop start. For example, if node N is 2 nodes into a 5 node loop, it is also correct to say that it is 7, 12, or even 397 nodes into the loop.

Therefore, both CollisionSpot and LinkedListHead are k nodes from the start of the loop.

Now, if we keep one pointer at CollisionSpot and move the other one to LinkedListHead, they will each be k nodes from LoopStart. Moving the two pointers at the same speed will cause them to collide again--this time after k steps, at which point they will both be at LoopStart. All we have to do is return this node.

Part 4: Putting It All Together

To summarize, we move FastPointer twice as fast as SlowPointer. When SlowPointer enters the loop, after k nodes, FastPointer is k nodes into the loop. This means that FastPointer and SlowPointer are LOOP_SIZE - k nodes away from each other.

The head of the linked list is also k nodes from the front of the loop. So, if we keep one pointer where it is, and move the other pointer to the head of the linked list. then they will meet at the front of the loop.
Our algorithm is derived directly from parts 1, 2 and 3.

  1. Create two pointers, FastPointer and SlowPointer.
  2. Move FastPointer at a rate of 2 steps and SlowPointer at a rate of 1 step.
  3. When they collide, move SlowPointer to LinkedListHead. Keep FastPointer where it is.
  4. Move SlowPointer and FastPointer at a·rate of one step. Return the new collision point.
最后编辑于
©著作权归作者所有,转载或内容合作请联系作者
  • 序言:七十年代末,一起剥皮案震惊了整个滨河市,随后出现的几起案子,更是在滨河造成了极大的恐慌,老刑警刘岩,带你破解...
    沈念sama阅读 212,383评论 6 493
  • 序言:滨河连续发生了三起死亡事件,死亡现场离奇诡异,居然都是意外死亡,警方通过查阅死者的电脑和手机,发现死者居然都...
    沈念sama阅读 90,522评论 3 385
  • 文/潘晓璐 我一进店门,熙熙楼的掌柜王于贵愁眉苦脸地迎上来,“玉大人,你说我怎么就摊上这事。” “怎么了?”我有些...
    开封第一讲书人阅读 157,852评论 0 348
  • 文/不坏的土叔 我叫张陵,是天一观的道长。 经常有香客问我,道长,这世上最难降的妖魔是什么? 我笑而不...
    开封第一讲书人阅读 56,621评论 1 284
  • 正文 为了忘掉前任,我火速办了婚礼,结果婚礼上,老公的妹妹穿的比我还像新娘。我一直安慰自己,他们只是感情好,可当我...
    茶点故事阅读 65,741评论 6 386
  • 文/花漫 我一把揭开白布。 她就那样静静地躺着,像睡着了一般。 火红的嫁衣衬着肌肤如雪。 梳的纹丝不乱的头发上,一...
    开封第一讲书人阅读 49,929评论 1 290
  • 那天,我揣着相机与录音,去河边找鬼。 笑死,一个胖子当着我的面吹牛,可吹牛的内容都是我干的。 我是一名探鬼主播,决...
    沈念sama阅读 39,076评论 3 410
  • 文/苍兰香墨 我猛地睁开眼,长吁一口气:“原来是场噩梦啊……” “哼!你这毒妇竟也来了?” 一声冷哼从身侧响起,我...
    开封第一讲书人阅读 37,803评论 0 268
  • 序言:老挝万荣一对情侣失踪,失踪者是张志新(化名)和其女友刘颖,没想到半个月后,有当地人在树林里发现了一具尸体,经...
    沈念sama阅读 44,265评论 1 303
  • 正文 独居荒郊野岭守林人离奇死亡,尸身上长有42处带血的脓包…… 初始之章·张勋 以下内容为张勋视角 年9月15日...
    茶点故事阅读 36,582评论 2 327
  • 正文 我和宋清朗相恋三年,在试婚纱的时候发现自己被绿了。 大学时的朋友给我发了我未婚夫和他白月光在一起吃饭的照片。...
    茶点故事阅读 38,716评论 1 341
  • 序言:一个原本活蹦乱跳的男人离奇死亡,死状恐怖,灵堂内的尸体忽然破棺而出,到底是诈尸还是另有隐情,我是刑警宁泽,带...
    沈念sama阅读 34,395评论 4 333
  • 正文 年R本政府宣布,位于F岛的核电站,受9级特大地震影响,放射性物质发生泄漏。R本人自食恶果不足惜,却给世界环境...
    茶点故事阅读 40,039评论 3 316
  • 文/蒙蒙 一、第九天 我趴在偏房一处隐蔽的房顶上张望。 院中可真热闹,春花似锦、人声如沸。这庄子的主人今日做“春日...
    开封第一讲书人阅读 30,798评论 0 21
  • 文/苍兰香墨 我抬头看了看天上的太阳。三九已至,却和暖如春,着一层夹袄步出监牢的瞬间,已是汗流浃背。 一阵脚步声响...
    开封第一讲书人阅读 32,027评论 1 266
  • 我被黑心中介骗来泰国打工, 没想到刚下飞机就差点儿被人妖公主榨干…… 1. 我叫王不留,地道东北人。 一个月前我还...
    沈念sama阅读 46,488评论 2 361
  • 正文 我出身青楼,却偏偏与公主长得像,于是被迫代替她去往敌国和亲。 传闻我的和亲对象是个残疾皇子,可洞房花烛夜当晚...
    茶点故事阅读 43,612评论 2 350

推荐阅读更多精彩内容

  • 其实你始终很孤独 但还是对生活充满热情 其实你始终坚持一些梦想 也明明知道有些梦不可能实现 其实你始终相信爱情 也...
    时光慢递阅读 361评论 0 1
  • 六 折腾的周六 2011年7月23日,李杨和段工要先撤退回国。两个不会日语的人,要安全到达成田机场,确实有点难度。...
    明哥明说阅读 349评论 0 2
  • 1. 人际沟通,自我概念,自我应验预言,身份认同管理 澄清一些大众对沟通的误解:意思不在字眼里;沟通不见得越多越好...
    bfe736b4c320阅读 321评论 0 0
  • 请家人们在下面的评论中回复学习心得。
    林玉珍阅读 147评论 5 0
  • 这是一个关注留守儿童的文字。同时也是我一路看着过来的书,在书里面,我经历了我那留守的童年,空巢的长辈对家的坚守,对...
    海熠阅读 183评论 2 1