CODEWITHSUNDEEP
cfilesplitdata-cleaning
Teja
Teja

Reputation: 13544

How to split a file into two files for 400k records

I am trying to separate my input data file into two output files based on a label.Here is my code below.The below code works only for less number of records but it goes into Segmentation fault for more no of. rows.

#include<stdio.h>
#include<stdlib.h>
#include<string.h>

int main(int argc,char *argv[])
{
        FILE *fp,*fp1,*fp2,*fp3;
        char *filename,line[80],line1[80];
        char *token,*token1,mystr[10];

        filename=(char *)argv[1];
        fp=fopen(filename,"r");

        if(fp ==NULL) //Checking whether the command line argument was correctly or not.
        printf("There is no such file in the directory.\n");

        if(remove("sales_ok_fraud.txt") != 0) //Checking for file existence.
        perror("Error in deleting the file.\n");
        else
        printf("The existing cleansed data file is successfully deleted.\n");

        if(remove("sales_unknwn.txt") != 0) //Checking for file existence.
        perror("Error in deleting the file.\n");
        else
        printf("The existing cleansed data file is successfully deleted.\n");

        while(fgets(line,80,fp)!=NULL) //Reading each line from file to calculate the file size.
        {
                strcpy(line1,line);
                token = strtok(line,",");
                token = strtok(NULL,",");
                token = strtok(NULL,",");
                token = strtok(NULL,",");
                token = strtok(NULL,",");
                token = strtok(NULL,",");
                token1 = strtok(token,"\n");
                memcpy(mystr,&token1[0],strlen(token1)-1);
                mystr[strlen(token1)-1] = '\0';


                if( strcmp(mystr,"ok") == 0 )
                {
                        fp1=fopen("sales_ok_fraud.txt","a");//Opening the file in append mode.
    fprintf(fp1,"%s",line1);//Writing into the file.
                        fclose(fp2);//Closing the file.

                        //printf("Inside ok - %s\n",mystr);
                }
                else if( strcmp(mystr,"fraud") == 0)
                {
                        fp2=fopen("sales_ok_fraud.txt","a");//Opening the file in append mode.
                        fprintf(fp2,"%s",line1);//Writing into the file.
                        fclose(fp2);//Closing the file.
                        //printf("Inside fraud - %s\n",mystr);
                }
                else
                {
                        fp3=fopen("sales_unknwn.txt","a");//Opening the file in append mode.
                        fprintf(fp3,"%s",line1);//Writing into the file.
                        fclose(fp3);//Closing the file.
                        //printf("This is unknown record.\n");
                }
        }

        fclose(fp);

        return 0;
}

Upvotes: 2

Views: 319

Answers (1)

iabdalkader
iabdalkader

Reputation: 17332

I see a few problems in your code, first strlen returns the length of string not including the null byte so you don't need -1 (that's why it probably doesn't match any of the strcmp's)

memcpy(mystr, &token1[0], strlen(token1));
mystr[strlen(token1)] = '\0';

And here I think you should close fp1 

fp1=fopen("sales_ok_fraud.txt","a");  //you open f1
fprintf(fp1,"%s",line1);              //you write
fclose(fp2);//Closing the file.       //you close fp2

Note: you should make sure that token1 doesn't overflow mystr.

Upvotes: 2

Related Questions