读取二进制文件并确定文件类型的程序

Question

我写了这段代码，它读取二进制文件并确定其文件类型（用于一些测试文件结尾）。它适用于 PDF、MP3，但不适用于 jpg。

问题是什么？对于 jpg，行 printf("%s[%d]: %x\n", "Buffer", j, buffer[j]); 显示了多个字节（即 ffffff 而不是一个字节）

#include <stdio.h>


const int header[6][8] =    { 
                            {0x89,0x50,0x4E,0x47,0x0D,0x0A,0x1A,0x0A},
                            {0xFF,0xD8,0x00,0x00,0x00,0x00,0x00,0x00},
                            {0xFF,0xFB,0x00,0x00,0x00,0x00,0x00,0x00},
                            {0x49,0x44,0x33,0x00,0x00,0x00,0x00,0x00},
                            {0x25,0x50,0x44,0x46,0x2D,0x00,0x00,0x00},
                            {0x42,0x4C,0x45,0x4E,0x44,0x45,0x52,0x00} 
                            };

const char* filetype[6] = {"PNG","JPG","MP3","MP3v2","PDF","Blender"};

int main()
{
    FILE *fd;
    char buffer[8];


    if ((fd = fopen("C:\Users\***\Desktop\Unnamed.jpg", "rb")) == NULL) {
        return -1;
    }

    //fread(buffer, sizeof(char), 8, fd);
    fread(buffer, sizeof(buffer), 1, fd);

    for (int i = 0; i < 6; i++) {
        for(int j = 0; j < 8; j++){
            printf("%s[%d][%d]: %x\n","Header",i,j,header[i][j]);
            printf("%s[%d]: %x\n", "Buffer", j, buffer[j]);

            if (header[i][j] == 0x00) {
                printf("%s: %s","Found file type",filetype[i]);
                return 1;
            }
            if (header[i][j] != buffer[j]) {
                break;
            }
        }
    }
    printf("%s", "Couldn't determine filetype - Not in library");
    return 0;
}

Answer 1

您忘记创建缓冲区 unsigned char，因此您对值进行了符号扩展。

显然，内置魔术签名的 table 也应该是 const unsigned char，并且应该通过单个 memcmp() 调用进行比较。

Answer 2

使用 unsigned char 表示 buffer（您也可以表示 header）以避免出现问题，如果您的 char 已签名并给出负值（您与 int 就像 0x89，其中设置了第 7 位）

PNG 也有问题，因为 PNG 的值为：

{0x89,0x50,0x4E,0x47,0x0D,0x0A,0x1A,0x0A},

不像所有其他情况那样以 0 结尾，这是必需的，因为您的算法需要找到 0 以表明您找到了：

        if (header[i][j] == 0x00) {
            printf("%s: %s","Found file type",filetype[i]);
            return 1;
        }

只需为 PNG 添加一列也有 0。

最后：

const unsigned char header[6][9] =    { 
                            {0x89,0x50,0x4E,0x47,0x0D,0x0A,0x1A,0x0A, 0x00},
                            {0xFF,0xD8,0x00,0x00,0x00,0x00,0x00,0x00, 0x00},
                            {0xFF,0xFB,0x00,0x00,0x00,0x00,0x00,0x00, 0x00},
                            {0x49,0x44,0x33,0x00,0x00,0x00,0x00,0x00, 0x00},
                            {0x25,0x50,0x44,0x46,0x2D,0x00,0x00,0x00, 0x00},
                            {0x42,0x4C,0x45,0x4E,0x44,0x45,0x52,0x00, 0x00} 
                            };

const char* filetype[6] = {"PNG","JPG","MP3","MP3v2","PDF","Blender"};

int main()
{
    FILE *fd;
    unsigned char buffer[sizeof(header[0])];


    if ((fd = fopen("C:\Users\***\Desktop\Unnamed.jpg", "rb")) == NULL) {
        return -1;
    }

    fread(buffer, sizeof(buffer), 1, fd);

    for (int i = 0; i < ; i++) {
        for(int j = 0; j < sizeof(header[0]); j++){
            printf("%s[%d][%d]: %x\n","Header",i,j,header[i][j]);
            printf("%s[%d]: %x\n", "Buffer", j, buffer[j]);

            if (header[i][j] == 0x00) {
                printf("%s: %s","Found file type",filetype[i]);
                return 1;
            }
            if (header[i][j] != buffer[j]) {
                break;
            }
        }
    }
    printf("%s", "Couldn't determine filetype - Not in library");
    return 0;
}

读取二进制文件并确定文件类型的程序

Program to read binary and determine filetype

c

binary

fread