wordcount.c

ics lab0 题目 用c实现单词计数

#include "std.h"
#include "stdlib.h"
#include "stdio.h"
#include "limits.h"
#include "ctype.h"
#define BUFSIZE 100
#define INSTATE 1
#define OUTSTATE 0
#define MAXWORD 100
struct key{
            char *word;
            int count;
        };


long getFileLength(FILE *fp)
{
        long length = 0;
        fseek(fp,0,SEEK_END);
        length = ftell(fp);
        fseek(fp,0,SEEK_SET);
       return length;
}
/*
int getWordCount(FILE *fp)
{
 printf("%c",fgetc(fp));
  int c,wordcount=0;
      while((c=fgetc(fp))!=EOF){
        if(c>=65&&c<=122){
         // if(( c >= 'A' && c <= 'Z')||(c >= 'a'&&c<='z')){
         // if(c>=65&&c<=122&&c!=91&&c!=92&&c!=93&&c!=94&&c!=95&&c!=96){
           while((c=fgetc(fp))>=65&&c<=122);
         printf("%c",fgetc(fp));
            // while((  c >='A' && c<='Z' )||( c>='a' && c<='z' ));
            // while(c>=65&&c<=122&&c!=91&&c!=92&&c!=93&&c!=94&&c!=95&&c!=96);
              wordcount++;
        }
       }
 return wordcount;
}*/
/*
int getWordCount(FILE *fp,int length)
{
 int wordcount=0;
 int i=0;
 int j = 0;
 char s[length];
 char ch;
  fgets(s,length,fp);
  printf("%s\n",s);

  while(s[i]!='\0'){
   for(j=0;j<length;j++){
     if(s[j]>=65&&s[j]<=122){
        while(s[++j]>=65&&s[++j]<=122);
         wordcount++;
        
       }
        
  }//for
  i++;
}

 return wordcount;
}*/
/*
char*  getwordarray(FILE *fp,int wordcount)
{
     int c;
     int i,j;
     char *wordarray[wordcount];
    for(i=0;i<wordcount;i++){
     wordarray[i]=(char *)malloc(sizeof(char)*20);
    }

     for(i=0;i<wordcount;i++)
     {
	       for(j=0;j<20;j++)
	       {
                  // printf("%c\n",fgetc(fp));
		  if((c=fgetc(fp))>=65 && c <= 122)
                   {
		    wordarray[i][j]=c+'0';
                    
                    printf("%c\n",wordarray[i][j]);
                   
                   }
		   else{
                       break;
                       printf("%s","else");
		   }
	       }
      }
return *wordarray;
	      
}
*/

//void sortarray(struct key* keytab)
void sortarray(struct key keytab[])
{
 int m,n;
 char* t;
    for(m=0;m<999;m++){
       for(n=m+1;n<1000;n++){
         if(strcmp(keytab[m].word,keytab[n].word)<0)
            {
              strcpy(t,keytab[m].word);
              strcpy(keytab[m].word,keytab[n].word);
              strcpy(keytab[n].word,t);
            }
        }
     }
                                                                                                           
}
/*
struct key *binsearch(char *word,struct key *tab,int n)

{
 int cond;
 struct key *low = &tab[0];
 struct key *high = &tab[n];
 struct key *mid;
 while(low<high){
   mid=low+(high-low)/2;
   if((cond = strcmp(word,mid->word))<0)
     high=mid;
   else if(cond<0)
     low = mid+1;
   else
     return mid;
 }
 return NULL;
}
*/
int binsearch(char *word,struct key tab[],int n)
{
  int cond;
  int low,high,mid;
  low=0;
  high=n-1;
  while(low<=high){
    mid=(low+high)/2;
    if((cond=strcmp(word,tab[mid].word))<0)
       high=mid-1;
    else if(cond>0)
       low=mid+1;
    else 
       return mid;
  }
  return -1;
}

struct key* wordArray(FILE *fp){
  int c;
  int i=0;
  int m,n,k,p;
  int u=0; 
  struct key *keytab;
  keytab=(struct key* )malloc(sizeof(struct key)*5000);
  for(u=0;u<5000;u++)
  {
   keytab[u].word = (char*)malloc(sizeof(char)*30);
  }

  struct key *repkeytab;
  repkeytab=(struct key*)malloc(sizeof(struct key)*5000);
  for(u=0;u<5000;u++)
  {
   repkeytab[u].word = (char*)malloc(sizeof(char)*30);
  }
 
  char str[1000];
  

  while((c=fgetc(fp))!=EOF){
    
       if(isalpha(c)){
        int y=0;       
          while(isalpha(c)){ 
               repkeytab[i].word[y]=c;
            
              y++;
             c=fgetc(fp);
                  
        }
        strcpy(keytab[i].word,repkeytab[i].word);        
        for(k=0;k<i;k++)
        {
            if(strcmp(keytab[k].word,repkeytab[i].word)==0)
           {
               keytab[k].count++;
           } 
          
        }
       
     
      printf("%s\t%d\n",keytab[i].word,keytab[i].count);      
     }
    
     i++;
     
  }
// sortarray(keytab);
 return keytab;
}



	int main(int argc, char *argv[]){
		int ndisp = 0;
		
		struct timeval starttime, endtime, durtime;
		int i;
		
		/* step0: get cmd parameter and check it */
		if(argv[2]!=NULL)
				ndisp = atoi(argv[2]);
			   
		else
				ndisp = 10;
		FILE *fp;
	       char *fileName=(char*)malloc(sizeof(char)*30); 
                  strcpy(fileName,argv[1]);
               fp=fopen(fileName,"r"); 
	       if(fp==NULL)
	       {
		 printf("can't open %s\n",fileName);
		 exit(1);
	       }
		long length=getFileLength(fp);
		
	     if(length>INT_MAX){
                 printf("the file is too large\n");
                 exit(1);
                }
             else if(length==1){
                 printf("the file has no content\n");
                 exit(1);
                } 
           
         	
	/* step1: prepare work (e.g.open file, load file to memory ...) */

	printf("wordcount: Running...\n");
	gettimeofday(&starttime, 0);

	/* step2: do word count */
                 
        int len, offset = 0;
        int wordcount=28;  
        //int wordcount = getWordCount(fp,length);
        printf("the number of words of the file is %d\n",wordcount);
       // struct key keytab[wordcount];
          struct key *keytab;
          keytab=wordArray(fp);
       // for(i=0;i<28;i++)
       // {
         // keytab=wordArray(fp);
         // printf("%s",keytab[i].word);
       // }       
        //getwordarray(fp,wordcount);      


	gettimeofday(&endtime, 0);
	GET_TIME(starttime, endtime, durtime);
	printf("wordcount: completed %ld.%ld(s)\n", durtime.tv_sec, durtime.tv_usec);


	printf("wordsort: running ...\n");
	gettimeofday(&starttime, 0);	

	/* step3: sort the results */

	gettimeofday(&endtime, 0);
	GET_TIME(starttime, endtime, durtime);
	printf("wordsort: completed %ld.%ld(s)\n", durtime.tv_sec, durtime.tv_usec);


	/* step4: print results */
	printf("\nwordcount: results (TOP %d from %d):\n", ndisp, len);


	/* step5: destroy work (e.g. free memory, close file ...) */
        fclose(fp);
	return 0;
}