如何从一个fstream正确地128字节读入一个字符串对象？ [英] how can I read exactly 128 bytes from an fstream into a string object?

问题描述

我如何从一个fstream读取128字节到一个字符串对象？

How do I read exactly 128 bytes from an fstream into a string object?

我写了一些代码来读取文件的前128个字节并打印然后最后128个字节的文件并打印。最后一部分工作，因为你可以轻松地迭代到EOF，但是如何从正面得到正好128字节？下面的代码不工作，因为你不能添加128到ifstream迭代器，它不可索引，只有incrementable（看起来）。

I wrote some code to read the first 128 bytes of a file and print it and then the last 128 bytes of the file and print that. The last part works, since you can easily iterate to EOF, but how do I get exactly 128 bytes from the front? The code below doesn't work since you can't add 128 to an ifstream iterator, it's not indexable, only incrementable (it seems).

迭代器和* ++它128次，但是必须有一行简单的方法来做到这一点，对吧？

Sure I could make an iterator and *++ it 128 times, but there must be a single line straightforward way to do it, right?

#include <iostream>
#include <fstream>
#include <string>

int main(int argc, char **argv)
{
    std::ifstream ifs ("input.txt",std::ifstream::in | std::ifstream::binary);

    if (ifs.good())
    {
    // read first 128 bytes into a string
        ifs.seekg(0,std::ifstream::beg);
        std::string first128((std::istreambuf_iterator<char>(ifs)),
                             (std::istreambuf_iterator<char>(ifs))+128);

        std::cout << first128 << std::endl;

    // read last 128 bytes into a string
        ifs.seekg(-128,std::ifstream::end);
        std::string last128((std::istreambuf_iterator<char>(ifs)),
                            std::istreambuf_iterator<char>());

        std::cout << last128 << std::endl;

        return 0;
    }

    return 1;
}

推荐答案

，但也许你会很高兴它使用迭代器从缓冲区初始化字符串。

My answer uses an intermediate buffer, but perhaps you will be happy that it uses iterators to initialize the string from the buffer.

std::vector<char> buffer(128); // create a buffer
ifs.read( &buffer[0], buffer.size() ); // read to buffer
std::string first128( buffer.begin(), buffer.end() ); // copy from vector

对我来说，似乎他们有一个太可爱的iostreams 。尝试对流I / O使用迭代器是非常复杂的。

To me it seems like they got a little too cute with the implementation of iostreams. Trying to use iterators for stream I/O is overly complicated.

顺便说一下，我怀疑你正在尝试的实现将覆盖下，做各种中间缓冲（也许在内核中，一些在库中），以及重新分配和复制字符串几次，因为它增长。

By the way, I suspect the implementation that you were attempting will, under the covers, do a variety of intermediate buffering (perhaps some in the kernel, some in the library) as well as re-allocating and copying the string several times as it grows.

另一个想法：你真的需要一个标准字符串的结果吗？你可能只是从向量工作 - 避免了复制到字符串的最后一步。或者，如果你感觉冒险，你可以创建自己的字符串类， 允许你以与向量相同的方式暴露内部缓冲区。

One other idea: Do you really need the result in a standard string? You might just work from the vector -- avoiding the final step of copying to a string. Or, if you are feeling adventurous, you could create your own string class that does allow you to expose the internal buffer in the same way that vector does.

这篇关于如何从一个fstream正确地128字节读入一个字符串对象？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！