B. Memory Manager
time limit per test

1 second

memory limit per test

64 megabytes

input

standard input

output

standard output

There is little time left before the release of the first national operating system BerlOS. Some of its components are not finished yet — the memory manager is among them. According to the developers' plan, in the first release the memory manager will be very simple and rectilinear. It will support three operations:

  • alloc n — to allocate n bytes of the memory and return the allocated block's identifier x;
  • erase x — to erase the block with the identifier x;
  • defragment — to defragment the free memory, bringing all the blocks as close to the beginning of the memory as possible and preserving their respective order;

The memory model in this case is very simple. It is a sequence of m bytes, numbered for convenience from the first to the m-th.

The first operation alloc n takes as the only parameter the size of the memory block that is to be allocated. While processing this operation, a free block of n successive bytes is being allocated in the memory. If the amount of such blocks is more than one, the block closest to the beginning of the memory (i.e. to the first byte) is prefered. All these bytes are marked as not free, and the memory manager returns a 32-bit integer numerical token that is the identifier of this block. If it is impossible to allocate a free block of this size, the function returns NULL.

The second operation erase x takes as its parameter the identifier of some block. This operation frees the system memory, marking the bytes of this block as free for further use. In the case when this identifier does not point to the previously allocated block, which has not been erased yet, the function returns ILLEGAL_ERASE_ARGUMENT.

The last operation defragment does not have any arguments and simply brings the occupied memory sections closer to the beginning of the memory without changing their respective order.

In the current implementation you are to use successive integers, starting with 1, as identifiers. Each successful alloc operation procession should return following number. Unsuccessful alloc operations do not affect numeration.

You are to write the implementation of the memory manager. You should output the returned value for each alloc command. You should also output ILLEGAL_ERASE_ARGUMENT for all the failed erase commands.

Input

The first line of the input data contains two positive integers t and m (1 ≤ t ≤ 100;1 ≤ m ≤ 100), where t — the amount of operations given to the memory manager for processing, and m — the available memory size in bytes. Then there follow t lines where the operations themselves are given. The first operation is alloc n (1 ≤ n ≤ 100), where n is an integer. The second one is erase x, where x is an arbitrary 32-bit integer numerical token. The third operation is defragment.

Output

Output the sequence of lines. Each line should contain either the result of alloc operation procession , or ILLEGAL_ERASE_ARGUMENT as a result of failed erase operation procession. Output lines should go in the same order in which the operations are processed. Successful procession of alloc operation should return integers, starting with 1, as the identifiers of the allocated blocks.

Examples
Input
6 10
alloc 5
alloc 3
erase 1
alloc 6
defragment
alloc 6
Output
1
2
NULL
3

一道纯模拟题,模拟内存分配过程。(os的首次适应算法
a:从前往后寻找一个满足大小的内存空间,放进去并给个编号(编号一直累加
e:清空给定编号的空间
d:把所有内存块移到最前面,空闲块都存在最后面了。

这题有两个大坑:1.内存块移到前面时不改变顺序,这个顺序不是编号的顺序,而是他们内存块的位置顺序。2.不合理编号的处理要细心点。

附ac代码:
  1 #include <cstdio>
2 #include <cstring>
3 #include <string>
4 #include <iostream>
5 #include <algorithm>
6 using namespace std;
7 string op;
8 int mem[111];
9 int idm[111];
10 struct nod
11 {
12 int st;
13 int ed;
14 int id;
15 }nd[111];
16 bool cmp(nod a,nod b)
17 {
18 if(a.st<b.st) return 1;
19 return 0;
20 }
21 int main()
22 {
23 ios::sync_with_stdio(false);
24 cin.tie(0);
25 cout.tie(0);
26 int t,m;
27 cin>>t>>m;
28 for(int i=1;i<=111;++i)
29 nd[i].id=i;
30
31 int u,id=1;
32 for(int i=1;i<=t;++i)
33 {
34
35 cin>>op;
36 if(op[0]=='a')
37 {
38 cin>>u;
39
40
41 int j=1,sum=0;
42 // cout<<"m:"<<mem[1]<<endl;
43 for(j=1;j<=m;++j)
44 {
45 if(mem[j]==0)
46 {
47 sum++;
48 }
49 else
50 {
51 sum=0;
52 }
53 if(sum==u)
54 {
55 int k=j;
56 while(sum!=0)
57 {
58 mem[k-sum+1]=1;
59 sum--;
60 }
61 break;
62 }
63 }
64 if(j==m+1) cout<<"NULL"<<endl;
65 else
66 {
67 cout<<id<<endl;
68 idm[id]=1;
69 nd[id].st=j-u+1;
70
71 nd[id++].ed=j;
72 // cout<<nd[id-1].st<<nd[id-1].ed<<endl;
73 }
74 }
75 else if(op[0]=='e')
76 {
77 cin>>u;
78 if(u<0 || u>id || idm[u]==0)
79 {
80 cout<<"ILLEGAL_ERASE_ARGUMENT"<<endl;
81 }
82 else
83 {
84 // cout<<nd[id-1].st<<nd[id-1].ed<<endl;
85 idm[u]=0;
86 for(int k=1;k<id;++k)
87 {
88
89 if(nd[k].id==u)
90 {
91
92 for(int j=nd[k].st;j<=nd[k].ed;++j)
93 {
94
95 mem[j]=0;
96 }
97 }
98 }
99
100 }
101 }
102 else
103 {
104 int sum=0,x=1;
105 for(int k=1;k<=m;++k)
106 {
107 if(mem[k]==1)
108 sum++;
109 }
110 memset(mem,0,sizeof(mem));
111 for(int k=1;k<=sum;++k)
112 mem[k]=1;
113 sort(nd+1,nd+id,cmp);
114 for(int k=1;k<id;++k)
115 {
116
117 if(idm[nd[k].id]==1)
118 {
119 // cout<<nd[k].id<<endl;
120 int len=nd[k].ed-nd[k].st+1;
121 nd[k].st=x;
122 nd[k].ed=x+len-1;
123 x=nd[k].ed+1;
124 // cout<<nd[k].st<<nd[k].ed<<endl;
125 }
126 }
127 }
128 }
129 return 0;
130 }

codeforces 7B的更多相关文章

  1. python爬虫学习(5) —— 扒一下codeforces题面

    上一次我们拿学校的URP做了个小小的demo.... 其实我们还可以把每个学生的证件照爬下来做成一个证件照校花校草评比 另外也可以写一个物理实验自动选课... 但是出于多种原因,,还是绕开这些敏感话题 ...

  2. 【Codeforces 738D】Sea Battle(贪心)

    http://codeforces.com/contest/738/problem/D Galya is playing one-dimensional Sea Battle on a 1 × n g ...

  3. 【Codeforces 738C】Road to Cinema

    http://codeforces.com/contest/738/problem/C Vasya is currently at a car rental service, and he wants ...

  4. 【Codeforces 738A】Interview with Oleg

    http://codeforces.com/contest/738/problem/A Polycarp has interviewed Oleg and has written the interv ...

  5. CodeForces - 662A Gambling Nim

    http://codeforces.com/problemset/problem/662/A 题目大意: 给定n(n <= 500000)张卡片,每张卡片的两个面都写有数字,每个面都有0.5的概 ...

  6. CodeForces - 274B Zero Tree

    http://codeforces.com/problemset/problem/274/B 题目大意: 给定你一颗树,每个点上有权值. 现在你每次取出这颗树的一颗子树(即点集和边集均是原图的子集的连 ...

  7. CodeForces - 261B Maxim and Restaurant

    http://codeforces.com/problemset/problem/261/B 题目大意:给定n个数a1-an(n<=50,ai<=50),随机打乱后,记Si=a1+a2+a ...

  8. CodeForces - 696B Puzzles

    http://codeforces.com/problemset/problem/696/B 题目大意: 这是一颗有n个点的树,你从根开始游走,每当你第一次到达一个点时,把这个点的权记为(你已经到过不 ...

  9. CodeForces - 148D Bag of mice

    http://codeforces.com/problemset/problem/148/D 题目大意: 原来袋子里有w只白鼠和b只黑鼠 龙和王妃轮流从袋子里抓老鼠.谁先抓到白色老鼠谁就赢. 王妃每次 ...

随机推荐

  1. 多视图子空间聚类/表示学习(Multi-view Subspace Clustering/Representation Learning)

    多视图子空间聚类/表示学习(Multi-view Subspace Clustering/Representation Learning) 作者:凯鲁嘎吉 - 博客园 http://www.cnblo ...

  2. Java基础复习3

    循环的嵌套 public class demo8 {     public static void main(String[] args) {         /*  输出########       ...

  3. 配置HDFS的HA

    配置前准备: -- 配置hadoop -- 配置ZooKeeper,传送门:https://www.cnblogs.com/zhqin/p/11906106.html 安装配置好hadoop和ZooK ...

  4. 解决windows git乱码问题

    在windows中打开git bash git config --global i18n.commitencoding utf-8       设置提交日志使用utf-8 git config --g ...

  5. 中央事件总线 事件驱动架构(EDA) 解析事件总线的4种实现方式

    事件驱动架构(EDA)https://mp.weixin.qq.com/s/nA8XFD2Rx_7qA_LxltGGHw https://mp.weixin.qq.com/s/cD3auglgKzOb ...

  6. Java泛型机制

    泛型程序设计 1.泛型程序设计的起源? 泛型是JDK1.5增加的新特性. 2.使用泛型的好处? 使用泛型机制编写的代码比那些杂乱使用Object变量,然后再进行强制类型转换的代码具有更好的安全性和可读 ...

  7. 从零搭建一个IdentityServer——初识OpenIDConnect

    上一篇文章实现了IdentityServer4与Asp.net core Identity的集成,可以使用通过identity注册功能添加的用户,以Password的方式获取Access token, ...

  8. (二)基于shard-jdbc中间件,实现数据分库分表

    基于shard-jdbc中间件,实现数据分库分表 Sharding-JDBC简介 Sharding配置示意图 1.水平分割 1.1 水平分库 1.2 水平分表 2.Shard-jdbc中间件 2.1 ...

  9. Spring5源码,@ModelAttribute

    一.什么是@ModelAttribute注解 二.@ModelAttribute注解相关代码详解 一.什么是@ModelAttribute注解 @ModelAttribute注解主要用来将请求转换为使 ...

  10. java架构《并发线程高级篇二》

    本章主要记录讲解并发线程的线程池.使用Executor框架自定义线程池. 自定义线程池使用Queue队列所表示出来的形式: 1 ArrayBlockingQueue<Runnable>(3 ...