Show offsets of holes and data in a possibly sparse file on Linux

sparsedump.c

/*
 * sparsedump: Show offsets of holes and data in a possibly sparse file on Linux
 *
 * Compile using:
 *     gcc -o sparsedump sparsedump.c -ggdb -Wall
 *
 * Ben Cohen, February 2025.
 */

/*
 * Test cases:
 *   echo XXX > W
 *
 *   truncate -s 10000000 X
 *
 *   truncate -s 10000000 Y
 *   echo XXX >> Y
 *   truncate -s 20000000 Y
 *   echo XXX >> Y
 *
 *   echo XXX >> Z
 *   truncate -s 10000000 Z
 *   echo XXX >> Z
 *   truncate -s 20000000 Z
 *
 * Output:
 *   $ ./sparsedump W
 *   OFFSET              WHAT
 *   ------------------  ----
 *   0x0000000000000000  DATA
 *   0x0000000000000004  EOF
 *   $ ./sparsedump X
 *   OFFSET              WHAT
 *   ------------------  ----
 *   0x0000000000000000  HOLE
 *   0x0000000000989680  EOF
 *   $ ./sparsedump Y
 *   OFFSET              WHAT
 *   ------------------  ----
 *   0x0000000000000000  HOLE
 *   0x0000000000989000  DATA
 *   0x000000000098a000  HOLE
 *   0x0000000001312000  DATA
 *   0x0000000001312d04  EOF
 *   $ ./sparsedump Z
 *   OFFSET              WHAT
 *   ------------------  ----
 *   0x0000000000000000  DATA
 *   0x0000000000001000  HOLE
 *   0x0000000000989000  DATA
 *   0x000000000098a000  HOLE
 *   0x0000000001312d00  EOF
 */

#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>
#include <stdbool.h>
#include <errno.h>

off_t lseek_checked(int fd, off_t offset, int whence)
{
    off_t result = lseek(fd, offset, whence);
    if (errno == 0
        || (errno == ENXIO && (whence == SEEK_HOLE || whence == SEEK_DATA)))
    {
        return result;
    }

    perror("lseek");
    exit(1);
}

int main(int argc, char **argv)
{
    if (argc != 2)
    {
       printf("Usage: %s <file>\n", argv[0]);
       exit(1);
    }

    int fd = open(argv[1], O_RDONLY);
    if (fd < 0)
    {
        perror("couldn't open file");
        exit(1);
    }

    off_t end_of_file = lseek_checked(fd, 0, SEEK_END);

    printf("OFFSET              WHAT\n");
    printf("------------------  ----\n");

    if (lseek_checked(fd, 0, SEEK_HOLE) == 0)
    {
        printf("0x0000000000000000  HOLE\n");
    }

    off_t current_offset = 0;
    bool looking_for_data = true;
    while (true)
    {
        int whence = looking_for_data ? SEEK_DATA : SEEK_HOLE;
        char *what = looking_for_data ? "DATA" : "HOLE";

        current_offset = lseek_checked(fd, current_offset, whence);
        if (current_offset == end_of_file || errno == ENXIO)
        {
            printf("0x%016lx  EOF\n", end_of_file);
            exit(0);
        }

        printf("0x%016lx  %s\n", current_offset, what);

        looking_for_data = !looking_for_data;
    }
}