c++ - 将二进制文件(exe，zip等)读取为char *，c++

Question

我正在尝试读取文件并将字节放入字节缓冲区中，但是当我尝试读取exe或zip文件时，并非所有字节都被加载到缓冲区中。我的功能:

char* read_file_bytes(const string &name) {
   FILE *img = fopen(name.c_str(), "rb");
   fseek(img, 0, SEEK_END);
   unsigned long filesize = ftell(img);
   char *buffer = (char*)malloc(sizeof(char)*filesize);
   rewind(img);
   fread(buffer, sizeof(char), filesize, img);
   return buffer;
}

检查缓冲区的代码段:

char* bytes = read_file_bytes(path);
for(int i = 0; i < strlen(bytes); i++)
    cout << hex << (unsigned int)(bytes[i]);

Answer 1

strlen()设计用于文本字符，而不用于二进制字节。当遇到nul char(0x00)(可能包含二进制数据)时，它将停止计数。

您的read_file_bytes()函数知道它读取了多少字节。您需要将该数字返回给调用方，例如:

typedef unsigned char byte;

byte* read_file_bytes(const std::string &name, unsigned long &filesize)
{
    filesize = 0;

    FILE *img = fopen(name.c_str(), "rb");
    if (!img)
        return NULL;

    if (fseek(img, 0, SEEK_END) != 0)
    {
        fclose(img);
        return NULL;
    }

    long size = ftell(img);
    if (size == -1L)
    {
        fclose(img);
        return NULL;
    }

    byte *buffer = static_cast<byte*>(std::malloc(size));
    if (!buffer)
    {
        fclose(img);
        return NULL;
    }

    rewind(img);
    if (fread(buffer, 1, size, img) < size)
    {
        free(buffer);
        close(img);
        return NULL;
    }

    fclose(img);

    filesize = size;
    return buffer;
}

unsigned long filesize;
byte* bytes = read_file_bytes(path, filesize);

for(unsigned long i = 0; i < filesize; ++i)
    std::cout << std::hex << static_cast<unsigned int>(bytes[i]);

free(bytes);

请注意，这种方法非常C式且容易出错。更多的C++方法看起来像这样:

using byte = unsigned char;
// or, use std::byte in C++17 and later...

std::vector<byte> read_file_bytes(const std::string &name)
{
    std::ifstream img;
    img.exceptions(std::ifstream::failbit | std::ifstream::badbit);
    img.open(name.c_str(), std::ifstream::binary | std::ifstream::ate);

    std::ifstream::pos_type size = img.tellg();
    ifs.seekg(0, std::ios::beg);
    // or, use std::filesystem::file_size() instead...

    std::vector<byte> buffer(size);
    img.read(buffer.data(), size);

    return buffer;
}

std::vector<byte> bytes = read_file_bytes(path);
for(byte b : bytes)
    std::cout << std::hex << static_cast<unsigned int>(b);