merging and sorting a text file in C

Question

I am supoosed write a function that reads two text files line by line, compare them, delete the duplicates and but them into a third file in alphabetical order...I have been working on this for over a month and I am still stuck I have tried several ways to do this and come up with nothing...I was in formed that i have to use strcmp to do this and I cant use any other predefined sorting function...I have also looked around on this site and cannot find much that helps with this...any help would be greatly appreciated..Here is what I have so far:

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


main (void)
{
    char str [200];
    char str2 [200];
    char new [100];
    char temp [100];
    int row = 10;
    FILE *fa = fopen ("book1.dat", "r");
    FILE *fb = fopen ("book2.dat", "r");
    FILE *fc = fopen ("fixed.txt", "w");

    int i;
    int j;
    int k;

    while (fgets (str, 200, fa) !=NULL && fgets (str2, 200, fb) !=NULL)
    {
        puts(str);
        puts(str2);

        if (strcmp( str, str2) ==0 )
        {
            strcpy (str , new);
        } else {
            strcpy (new, str);
            strcpy (new, str2);
        }
    }
    for ( i = 0; i < row; i++)
    {
        for (j = i+1; j< row; j++)
        {
            if(strcmp(new[i], new [j]) > 0)
            {
                strcpy (temp, new);
                strcpy(new, new);
                strcpy(new, temp);
            }
        }
    }
    for (i = 0; i < length; i ++)
    {
        fputs(new, fc);
    }
}

Answer 1

Your use of strcpy() is peculiar. Recall its signature:

char *strcpy(char *dest, const char *src)

Here's a usage that doesn't make immediate sense to me:

strcpy (new, str); // new now has str
strcpy (new, str2); // new now has str2

You've effectively overwritten something there. I would start from there, and see what else may not be working as you intend. Furthermore, if you can use gcc, look into using gdb as well to debug your code. (You would need to compile with the -g flag.)

Answer 2

First off, can you assume the duplicates from book1 and book2 line up nicely?

Think about how you would detect if the first entry in book1 is identical to the last entry in book2.

Secondly, you have to sort your output alphabetically. Sorting algorithms is kind of one of those common things that students are forced to do all the time. It builds character. For bonus kudos, implement quick sort.

Answer 3

sample a way.
error handling is omitted.
since we are using the sort function of the library sqort, implement your own.

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

#define LINE_MAX_SIZE 256

typedef struct filePos {
    FILE *fp;
    long pos;
} FilePos;

typedef struct myfile {
    int lines;
    int capacity;
    FILE *fp;
    FilePos *filePoss;
} MyFile;

MyFile *myfopen(const char *filepath){
    char buff[LINE_MAX_SIZE];
    MyFile *mfp;
    mfp = (MyFile*)malloc(sizeof(MyFile));
    mfp->lines = 0;
    mfp->capacity=16;
    mfp->filePoss=NULL;
    mfp->filePoss=(FilePos*)realloc(mfp->filePoss, sizeof(FilePos)*(mfp->capacity *= 2));
    mfp->fp = fopen(filepath, "r");
    do{
        mfp->filePoss[mfp->lines].fp = mfp->fp;
        mfp->filePoss[mfp->lines].pos = ftell(mfp->fp);
        if(++mfp->lines == mfp->capacity){
            mfp->filePoss=(FilePos*)realloc(mfp->filePoss, sizeof(FilePos)*(mfp->capacity *= 2));
        }
    }while(NULL!=fgets(buff, LINE_MAX_SIZE, mfp->fp));
    --mfp->lines;
    return mfp;
}

void myfclose(MyFile *mfp){
    free(mfp->filePoss);
    fclose(mfp->fp);
    free(mfp);
}

char *myfgets(FilePos *p, char *buff){
    fseek(p->fp, p->pos, SEEK_SET);
    return fgets(buff, LINE_MAX_SIZE, p->fp);
}

int myfcomp(const void *a, const void *b){
    char buff_a[LINE_MAX_SIZE];
    char buff_b[LINE_MAX_SIZE];
    FilePos *fpa,*fpb;
    fpa=(FilePos*)a;
    fpb=(FilePos*)b;
    myfgets(fpa, buff_a);
    myfgets(fpb, buff_b);
    return strcmp(buff_a, buff_b);
}

void myfsort(MyFile *mfp){
    qsort(mfp->filePoss, mfp->lines, sizeof(FilePos), myfcomp);
}

void myfprint(MyFile *mfp){
    char buff[LINE_MAX_SIZE];
    int i;

    for(i=0;i<mfp->lines ;++i)
        printf("%s", myfgets(mfp->filePoss + i, buff));
}

void merge(const char *inpfile1, const char *inpfile2, const char *outfile){
    FILE *fo;
    MyFile *fi1, *fi2;
    char buff_f1[LINE_MAX_SIZE];
    char buff_f2[LINE_MAX_SIZE];
    char buff_fo[LINE_MAX_SIZE];
    char *outbuff=NULL;
    int fi1_line, fi2_line;
    int eof1, eof2;

    fo=fopen(outfile, "w");
    fi1=myfopen(inpfile1);
    fi2=myfopen(inpfile2);
    myfsort(fi1);
    myfsort(fi2);
    fi1_line=fi2_line=0;
    eof1=eof2=0;
    *buff_fo='\0';
    while(1){
        if(!eof1 && outbuff != buff_f2){
            myfgets(&(fi1->filePoss[fi1_line]), buff_f1);
        }
        if(!eof2 && outbuff != buff_f1){
            myfgets(&(fi2->filePoss[fi2_line]), buff_f2);
        }
        if(!eof1 && !eof2){
           if(strcmp(buff_f1, buff_f2) <= 0){
                outbuff=buff_f1;
                ++fi1_line;
            } else {
                outbuff=buff_f2;
                ++fi2_line;
            }
        } else if(!eof1 && eof2){
            outbuff=buff_f1;
            ++fi1_line;
        } else if(eof1 && !eof2){
            outbuff=buff_f2;
            ++fi2_line;
        } else {
            break;
        }
        if(strcmp(outbuff, buff_fo) != 0){//duplicate check
            strcpy(buff_fo, outbuff);
            fputs(buff_fo, fo);
        }
        if(fi1->lines == fi1_line)
            eof1 = !0;
        if(fi2->lines == fi2_line)
            eof2 = !0;
    }
    myfclose(fi2);
    myfclose(fi1);
    fclose(fo);
}

int main(){
    merge("book1.txt", "book2.txt", "fixed.txt");
    return 0;
}