From: Robert Taylor [taylor_robert@jpmorgan.com]
Sent: Friday, October 20, 2000 5:36 AM
To: Info-VAX@Mvb.Saic.Com
Subject: Re: VMS C Compiler

For whats its worth heres the code. I needed it quickly so no sub-routines
or anything. If anyone feels like taking up the task I thought future
enhancements could be :

A) Allowing a specified delimiter
B) Allowing ignored columns to be of variable length and thus different in
each file being scanned.
C) Allowing more than one column to be ignored.

I know this isn't a C group but it might be of use to somebody. I'm done
with it for now.

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

/* Compare 2 CSV files, ignoring specified columns
*/

main(   int argc,   char *argv[] )
{
    char file1_rec  [1024];
    char file2_rec  [1024];
    FILE *Fp1 = 0;
    FILE *Fp2 = 0;

    int current_col = 1;
    int idx = 0;
    int line_num = 0;
    long ignore_col;

    if ( argc != 4 )
    {
        printf("Usage : CSV_COMPARE FILENAME1 FILENAME2 COLUMN_TO_IGNORE");

    }
    else
    {
        ignore_col = atoi(argv[3]);

        printf("File1 : %s\nFile2 %s\nIgnore Column %s\n", argv[1],
argv[2], argv[3] );
        printf("-=-=-=-=-=-=-=-=-=-=-=-\n");

        /* Open file1 and file 2 */
        Fp1 = fopen(argv[1],"r");
        Fp2 = fopen(argv[2],"r");

        do{
                /*
                Test if EOF FILE1 has been reached, further if it has test
that FILE2 has
                been completely read also
                */
                if (!fgets(file1_rec,sizeof(file1_rec),Fp1))
                {
                        printf("End of file1 detected - quitting\n");
                        if (fgets(file2_rec,sizeof(file2_rec),Fp2))
                        {
                                printf("Warning FILE2 still has more
data\n");
                        }
                        break;
                }

                /*
                Test if EOF FILE2 has been reached
                */
                if (!fgets(file2_rec,sizeof(file2_rec),Fp2))
                {
                        printf("End of file1 detected - quitting\n");
                        printf("This means FILE1 wasn't completely read -
ERROR\n");
                        break;
                }

                current_col = 1;
                ++line_num;
                /* So we have a record from each file, now compare content
*/
                for ( idx = 0; idx <= (strlen(file1_rec)-1); ++idx ){
                        if ( file1_rec[idx] == ',' ){
                                ++current_col;
                        }
                        else{
                                if ( current_col != ignore_col
                                   && file1_rec[idx] != file2_rec[idx] )
                                {
                                        printf ("Mismatch Line %d , char %d
\n", line_num, idx);
                                        printf ("*FILE1* %s", file1_rec );
                                        printf ("*FILE2* %s", file2_rec );
                                        break;
                                }
                        }
                }


        } while (1);


        /* Close file 1 and file 1 */
        fclose(Fp1);
        fclose(Fp2);
    }

    return 0;
}