程序读取二进制文件并确定文件类型

Question

我写了这段代码，它读取文件二进制文件并确定其文件类型（对于一些测试文件的目的）。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]);显示多个字节（即ffffff而不是仅一个字节）

#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

使用unsigned char为buffer（你也可以为header）以避免问题，如果你的char被签名并给出一个负值（你比较int如0x89，其中第7位被设置）

你也有PNG的问题，因为PNG的值是：

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

没有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来表示您找到了：

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

只需为PNG添加一列也为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;
}

Answer 2

你忘了制作你的缓冲区unsigned char，所以你得到了值的符号扩展。

显然，内置魔法签名表也应该是const unsigned char，并且应该用单个memcmp()调用进行比较。