C ++的二进制数据转换为十六进制字符串和背部 [英] C++ Converting binary data to a hex string and back

问题描述

我有一对匹配的一个工具类的静态函数，我使用二进制数据（无符号字符）之间的转换，它的字符串重新presentation（A-F和0-9）。他们似乎正常工作，但最近我试图编译我的code在Visual C ++（2010年防爆preSS）和我失望的是，它们会导致什么，但堆损坏错误。我在做什么错了？

I have a matching pair of static functions in a utility class that I use to convert between binary data (unsigned characters) and it's string representation (a-f and 0-9). They seemed to work correctly but recently I tried to compile my code under Visual C++ (2010 Express) and to my dismay, they cause nothing but heap corruption errors. What am I doing wrong?

void Utility::string_to_binary(const std::string source, unsigned char* destination, unsigned int length)
{
    unsigned int effective_length = min(length, (unsigned int) source.length() / 2);
    for(unsigned int b = 0; b < effective_length; b++)
    {
        sscanf(source.data() + (b * 2), "%02x", (unsigned int*) &destination[b]);
    }
}

void Utility::binary_to_string(const unsigned char* source, unsigned int length, std::string& destination)
{
    destination.clear();
    for(unsigned int i = 0; i < length; i++)
    {
        char digit[3];
        sprintf(digit, "%02x", source[i]);
        destination.append(digit);
    }
}

编辑：下面是说明该问题的完整方案

#include <iostream>
#include <hdcs/Utility.h>

using namespace std;

int main(int argc, char* argv[])
{
    //Generate some data
    unsigned int size = 1024;
    unsigned char* data = new unsigned char[size];

    //Convert it to it's string representation
    string hex;
    Utility::binary_to_string(data, size, hex);

    //Output it to the screen
    cout << hex << endl;

    //Clear the data buffer
    memset(data, 0, sizeof(unsigned char) * size);

    //Convert the hex string back to binary
    Utility::string_to_binary(hex, data, size);

    //Cleanup
    delete[] data;
}

在错误发生，删除[]数据。

推荐答案

在此code，

for(unsigned int b = 0; b < effective_length; b++)
{
    sscanf(source.data() + (b * 2), "%02x", (unsigned int*) &destination[b]);
}

您似乎写一个 unsigned int类型的位置目标，目的地+ 1 ，目的地+ 2 ，和C。当你接近你的目标缓冲区的字节最后，你会写超出其极限。

you seem to be writing an unsigned int at locations destination, destination+1, destination+2, &c. As you approach the final bytes of your destination buffer, you will write beyond its limit.

例如起见，我们假定目标是一个四字节的缓冲区，而的sizeof（unsigned int类型）是在您的环境4。然后，每个的sscanf 正在写四个字节。

For the sake of example, let us assume that destination is a four-byte buffer, and that sizeof (unsigned int) is 4 in your environment. Then each sscanf is writing four bytes.

第一次迭代写入字节0，1，2，3

The first iteration writes bytes 0, 1, 2, 3

第二iteratino写入字节1，2，3，4

The second iteratino writes bytes 1, 2, 3, 4

第三迭代写字节2，3，4，5

The third iteration writes bytes 2, 3, 4, 5

的最终迭代写字节3，4，5，6

The final iteration writes bytes 3, 4, 5, 6

由于缓冲区是开始，你写超出你缓冲区末尾只有四个字节。热潮。

Since the buffer was only four bytes to start with, you have written beyond the end of your buffer. Boom.

---

修改

要避免这种特定的错误所需的最小变动如下：

The minimum change required to avoid this particular bug follows:

for(unsigned int b = 0; b < effective_length; b++)
{
    unsigned int ui;
    sscanf(source.data() + (b * 2), "%02x", &ui);
    destination[b] = ui;
}

这篇关于C ++的二进制数据转换为十六进制字符串和背部的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！