云计算实验

int main()

{

clock_t start, stop;

start = clock();//记录程序开始时间

FILE *fp;

int isfirst = 1;//文件第一个单词

int isfound = 0;//是否发现新单词，表示未发现

myword *head;

myword *p1, *p2, *p3,*p4;

p1 = (myword*)malloc(sizeof(myword));

p1->occnum=0;

p1->next=NULL;p2 = p1;p3 = p1;head = NULL;

fp = fopen("//home//huchaowei//Desktop//data(1).dat", "r");//打开数据文件

while (!feof(fp))//当文件未结束

{if (isfirst == 1)

{fscanf(fp, "%s", p1->name);p1->occnum = 1;head = p1;isfirst = 0;}

else{char temp[MAX]={0};

fscanf(fp, "%s", temp);//fscanf函数遇空格或回车当成\0处理。

change(temp);

p3 = head;isfound = 0;

while (p3 != NULL)

{if (strcmp(p3->name, temp) != 0){p3 = p3->next;}

else{p3->occnum += 1;isfound = 1;break;}

}

if (isfound == 0)//若发现新单词，则新建结点插入链表

{

p4 = (myword*)malloc(sizeof(myword);

p4->occnum=0;

p4->next=NULL;

p2=head;

while(p2->next!=NULL)

p2=p2->next;

strcpy(p4->name,temp);

p4->occnum = 1;

p2->next = p4;

}

}

}

select(head);//选出前三

fclose(fp);

stop = clock(); //记录程序结束时间

printf("Time:");

printf("%f\n", (double)(stop - start) / CLOCKS_PER_SEC);//打印程序运行时间

system("pause");

return 0;

}

typedef struct myword

{

char name[MAX];

int occnum;

struct myword *next;

}myword;

void select(myword *head)

{

FILE *fp2;

myword *p3 = head;

int na,nb,nc;

char* a=(char*)malloc(20*sizeof(char));char* b=(char*)malloc(20*sizeof(char));

char* c=(char*)malloc(20*sizeof(char));na=nb=nc=0;

while (p3 != NULL)

{printf("name=%s,count=%d\n", p3->name, p3->occnum);

if(naoccnum)

{nc=nb;c=b;nb=na；b=a;na=p3->occnum;a=p3->name;}

else if(nboccnum)

{nc=nb; c=b；nb=p3->occnum;b=p3->name;}

else if(ncoccnum)

{nc=p3->occnum;c=p3->name;}

p3 = p3->next;}

printf("\nTop three：%s:%d %s:%d %s:%d\n",a,na,b,nb,c,nc);

fp2 = fopen("//home//huchaowei//Desktop//data.dat", "w");

fprintf(fp2,"\nTop three：%s:%d %s:%d %s:%d\n",a,na,b,nb,c,nc);//统计结果写入文件中

fclose(fp2);}

void change(char *str)//将从文件中读出的字符串中含有的标点符号去掉，并将大写转为小写

{char ch;char temp[MAX];

memset(temp,'\0',sizeof(temp)/sizeof(char));int i=0,j=0;

while((ch=str[i])!='\0')

{if(ch>96&&ch<123){temp[j] = ch;j++;}

else if(ch>64&&ch<91){temp[j] = ch + 32;j++;}i++;}

memset(str,'\0',sizeof(str)/sizeof(char));strcpy(str,temp);}

public class Topcount {

public static class TokenizerMapper extends

Mapper

{

private final static IntWritable one = new IntWritable(1);

private Text word = new Text();

public void map(Object key, Text value, Context context)

throws IOException, InterruptedException

{

String line = value.toString().toLowerCase(); // 全部转换为小写字母

StringTokenizer itr = new StringTokenizer(line, " \t\n\f\" . , : ; ? ! [ ] ' - ");

//常见的标点符号识别

while (itr.hasMoreTokens())

{

word.set(itr.nextToken());

context.write(word, one);

}

}

}

public static class TokenizerMapper2

extends Mapper

{

int c=0;

public void map(Object key, Text value, Context context)

throws IOException, InterruptedException

{

StringTokenizer itr = new StringTokenizer(value.toString());

IntWritable a=new IntWritable(Integer.parseInt(itr.nextToken()));

Text b=new Text(itr.nextToken());

if(c<3)//取前三

{

context.write(a, b);

c++;

}

}

}

public static class IntSumReducer extends

Reducer

{

private IntWritable result = new IntWritable();

public void reduce(Text key, Iterable values,

Context context) throws IOException, InterruptedException {

int sum = 0;