BOOST 绝对简明手册(写作中!!!!!!!) -- [email protected] TableOfContents

BOOST THE WORLD .

更新:2006.10 beta

作者:张沈鹏  电子科大大三  生物医学工程
 -- 欢迎指出错误和遗漏,提出建议和意见,请发信到[email protected]

[http://www.cppblog.com/zuroc 我的C++Blog] [http://www.cnweblog.com/zuroc/ 我的文学Blog]

参考:BOOST文档

-- 欢迎转载,但请保留引用网址以获得更新

1. 序言

现在学的东西很容易忘记,写这篇文章的目的是能让我在需要时快速找回当时的感觉.

2. 编译:VC2005注意

在属性->C/C++->预处理器->预处理定义中加入

_CRT_SECURE_NO_DEPRECATE;

来屏蔽不必要的警告

3. Asio 网络库

Boost.Asio是利用当代C++的先进方法,跨平台,异步I/O模型的C++网络库.

3.1. 网络库:VC2005注意

在属性->C/C++->命令行中加入

-DBOOST_REGEX_NO_LIB

来防止自动连接.

3.2. 同步Timer

本章介绍asio如何在定时器上进行阻塞等待(blocking wait).

实现,我们包含必要的头文件.

所有的asio类可以简单的通过include "asio.hpp"来调用.

#include <iostream>
#include <boost/asio.hpp>

此外,这个示例用到了timer,我们还要包含Boost.Date_Time的头文件来控制时间.

#include <boost/date_time/posix_time/posix_time.hpp>

使用asio至少需要一个boost::asio::io_service对象.该类提供了访问I/O的功能.我们首先在main函数中声明它.

int main()
{
boost::asio::io_service io;

下一步我们声明boost::asio::deadline_timer对象.这个asio的核心类提供I/O的功能(这里更确切的说是定时功能),总是把一个io_service对象作为他的第一个构造函数,而第二个构造函数的参数设定timer会在5秒后到时(expired).

boost::asio::deadline_timer t(io, boost::posix_time::seconds(5));

这个简单的示例中我们演示了定时器上的一个阻塞等待.就是说,调用boost::asio::deadline_timer::wait()的在创建后5秒内(注意:不是等待开始后),timer到时之前不会返回任何值.

一个deadline_timer只有两种状态:到时,未到时.

如果boost::asio::deadline_timer::wait()在到时的timer对象上调用,会立即return.

t.wait();

最后,我们输出理所当然的"Hello, world!"来演示timer到时了.

std::cout << "Hello, world!\n";

return 0;
}

完整的代码:

#include <iostream>
#include <boost/asio.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

int main()
{
boost::asio::io_service io;

boost::asio::deadline_timer t(io, boost::posix_time::seconds(5));
t.wait();

std::cout << "Hello, world!\n";

return 0;
}

3.3. 异步Timer

#include <iostream>
#include <asio.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

asio的异步函数会在一个异步操作完成后被回调.这里我们定义了一个将被回调的函数.

void print(const asio::error& /*e*/)
{
  std::cout << "Hello, world!\n";
}

int main()
{
  asio::io_service io;

  asio::deadline_timer t(io, boost::posix_time::seconds(5));

这里我们调用asio::deadline_timer::async_wait()来异步等待

  t.async_wait(print);

最后,我们必须调用asio::io_service::run().

asio库只会调用那个正在运行的asio::io_service::run()的回调函数.

如果asio::io_service::run()不被调用,那么回调永远不会发生.

asio::io_service::run()会持续工作到点,这里就是timer到时,回调完成.

别忘了在调用 asio::io_service::run()之前设置好io_service的任务.比如,这里,如果我们忘记先调用asio::deadline_timer::async_wait()则asio::io_service::run()会在瞬间return.

  io.run();

  return 0;
}

完整的代码:

#include <iostream>
#include <asio.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

void print(const asio::error& /*e*/)
{
  std::cout << "Hello, world!\n";
}

int main()
{
  asio::io_service io;

  asio::deadline_timer t(io, boost::posix_time::seconds(5));
  t.async_wait(print);

  io.run();

  return 0;
}

3.4. 回调函数的参数

这里我们将每秒回调一次,来演示如何回调函数参数的含义

#include <iostream>
#include <asio.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

首先,调整一下timer的持续时间,开始一个异步等待.显示,回调函数需要访问timer来实现周期运行,所以我们再介绍两个新参数

指向timer的指针
一个int*来指向计数器

void print(const asio::error& /*e*/,
    asio::deadline_timer* t, int* count)
{

我们打算让这个函数运行6个周期,然而你会发现这里没有显式的方法来终止io_service.不过,回顾上一节,你会发现当asio::io_service::run()会在所有任务完成时终止.这样我们当计算器的值达到5时(0为第一次运行的值),不再开启一个新的异步等待就可以了.

  if (*count < 5)
  {
    std::cout << *count << "\n";
    ++(*count);

然后,我们推迟的timer的终止时间.通过在原先的终止时间上增加延时,我们可以确保timer不会在处理回调函数所需时间内的到期.

(原文:By calculating the new expiry time relative to the old, we can ensure that the timer does not drift away from the whole-second mark due to any delays in processing the handler.)

    t->expires_at(t->expires_at() + boost::posix_time::seconds(1));

然后我们开始一个新的同步等待.如您所见,我们用把print和他的多个参数用boost::bind函数合成一个的形为void(const asio::error&)回调函数(准确的说是function object).

在这个例子中, boost::bind的asio::placeholders::error参数是为了给回调函数传入一个error对象.当进行一个异步操作,开始boost::bind时,你需要使用它来匹配回调函数的参数表.下一节中你会学到回调函数不需要error参数时可以省略它.

    t->async_wait(boost::bind(print,
          asio::placeholders::error, t, count));
  }
}

int main()
{
  asio::io_service io;

  int count = 0;
  asio::deadline_timer t(io, boost::posix_time::seconds(1));

和上面一样,我们再一次使用了绑定asio::deadline_timer::async_wait()

  t.async_wait(boost::bind(print,
        asio::placeholders::error, &t, &count));

  io.run();

在结尾,我们打印出的最后一次没有设置timer的调用的count的值

  std::cout << "Final count is " << count << "\n";

  return 0;
}

完整的代码:

#include <iostream>
#include <asio.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

void print(const asio::error& /*e*/,
    asio::deadline_timer* t, int* count)
{
  if (*count < 5)
  {
    std::cout << *count << "\n";
    ++(*count);

    t->expires_at(t->expires_at() + boost::posix_time::seconds(1));
    t->async_wait(boost::bind(print,
          asio::placeholders::error, t, count));
  }
}

int main()
{
  asio::io_service io;

  int count = 0;
  asio::deadline_timer t(io, boost::posix_time::seconds(1));
  t.async_wait(boost::bind(print,
        asio::placeholders::error, &t, &count));

  io.run();

  std::cout << "Final count is " << count << "\n";

  return 0;
}

3.5. 成员函数作为回调函数

本例的运行结果和上一节类似

#include <iostream>
#include <boost/asio.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

我们先定义一个printer类

class printer
{
public:

构造函数有一个io_service参数,并且在初始化timer_时用到了它.用来计数的count_这里同样作为了成员变量

  printer(boost::asio::io_service& io)
    : timer_(io, boost::posix_time::seconds(1)),
      count_(0)
  {

boost::bind同样可以出色的工作在成员函数上.众所周知,所有的非静态成员函数都有一个隐式的this参数,我们需要把this作为参数bind到成员函数上.和上一节类似,我们再次用bind构造出void(const boost::asio::error&)形式的函数.

注意,这里没有指定boost::asio::placeholders::error占位符,因为这个print成员函数没有接受一个error对象作为参数.

    timer_.async_wait(boost::bind(&printer::print, this));
  }

在类的折构函数中我们输出最后一次回调的conut的值

  ~printer()
  {
    std::cout << "Final count is " << count_ << "\n";
  }

print函数于上一节的十分类似,但是用成员变量取代了参数.

  void print()
  {
    if (count_ < 5)
    {
      std::cout << count_ << "\n";
      ++count_;

      timer_.expires_at(timer_.expires_at() + boost::posix_time::seconds(1));
      timer_.async_wait(boost::bind(&printer::print, this));
    }
  }

private:
  boost::asio::deadline_timer timer_;
  int count_;
};

现在main函数清爽多了,在运行io_service之前只需要简单的定义一个printer对象.

int main()
{
  boost::asio::io_service io;
  printer p(io);
  io.run();

  return 0;
}

完整的代码:

#include <iostream>
#include <boost/asio.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

class printer
{
public:
  printer(boost::asio::io_service& io)
    : timer_(io, boost::posix_time::seconds(1)),
      count_(0)
  {
    timer_.async_wait(boost::bind(&printer::print, this));
  }

  ~printer()
  {
    std::cout << "Final count is " << count_ << "\n";
  }

  void print()
  {
    if (count_ < 5)
    {
      std::cout << count_ << "\n";
      ++count_;

      timer_.expires_at(timer_.expires_at() + boost::posix_time::seconds(1));
      timer_.async_wait(boost::bind(&printer::print, this));
    }
  }

private:
  boost::asio::deadline_timer timer_;
  int count_;
};

int main()
{
  boost::asio::io_service io;
  printer p(io);
  io.run();

  return 0;
}

3.6. 多线程回调同步

本节演示了使用boost::asio::strand在多线程程序中进行回调同步(synchronise).

先前的几节阐明了如何在单线程程序中用boost::asio::io_service::run()进行同步.如您所见,asio库确保仅当当前线程调用boost::asio::io_service::run()时产生回调.显然,仅在一个线程中调用boost::asio::io_service::run() 来确保回调是适用于并发编程的.

一个基于asio的程序最好是从单线程入手,但是单线程有如下的限制,这一点在服务器上尤其明显:

当回调耗时较长时,反应迟钝.
在多核的系统上无能为力

如果你发觉你陷入了这种困扰,可以替代的方法是建立一个boost::asio::io_service::run()的线程池.然而这样就允许回调函数并发执行.所以,当回调函数需要访问一个共享,线程不安全的资源时,我们需要一种方式来同步操作.

#include <iostream>
#include <boost/asio.hpp>
#include <boost/thread.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

在上一节的基础上我们定义一个printer类,此次,它将并行运行两个timer

class printer
{
public:

除了声明了一对boost::asio::deadline_timer,构造函数也初始化了类型为boost::asio::strand的strand_成员.

boost::asio::strand可以分配的回调函数.它保证无论有多少线程调用了boost::asio::io_service::run(),下一个回调函数仅在前一个回调函数完成后开始,当然回调函数仍然可以和那些不使用boost::asio::strand分配,或是使用另一个boost::asio::strand分配的回调函数一起并发执行.

  printer(boost::asio::io_service& io)
    : strand_(io),
      timer1_(io, boost::posix_time::seconds(1)),
      timer2_(io, boost::posix_time::seconds(1)),
      count_(0)
  {

当一个异步操作开始时,用boost::asio::strand来 "wrapped(包装)"回调函数.boost::asio::strand::wrap()会返回一个由boost::asio::strand分配的新的handler(句柄),这样,我们可以确保它们不会同时运行.

    timer1_.async_wait(strand_.wrap(boost::bind(&printer::print1, this)));
    timer2_.async_wait(strand_.wrap(boost::bind(&printer::print2, this)));
  }

  ~printer()
  {
    std::cout << "Final count is " << count_ << "\n";
  }

多线程程序中,回调函数在访问共享资源前需要同步.这里共享资源是std::cout 和count_变量.

  void print1()
  {
    if (count_ < 10)
    {
      std::cout << "Timer 1: " << count_ << "\n";
      ++count_;

      timer1_.expires_at(timer1_.expires_at() + boost::posix_time::seconds(1));
      timer1_.async_wait(strand_.wrap(boost::bind(&printer::print1, this)));
    }
  }

  void print2()
  {
    if (count_ < 10)
    {
      std::cout << "Timer 2: " << count_ << "\n";
      ++count_;

      timer2_.expires_at(timer2_.expires_at() + boost::posix_time::seconds(1));
      timer2_.async_wait(strand_.wrap(boost::bind(&printer::print2, this)));
    }
  }

private:
  boost::asio::strand strand_;
  boost::asio::deadline_timer timer1_;
  boost::asio::deadline_timer timer2_;
  int count_;
};

main函数中boost::asio::io_service::run()在两个线程中被调用:主线程,一个boost::thread线程.

正如单线程中那样,并发的boost::asio::io_service::run()会一直运行直到完成任务.后台的线程将在所有异步线程完成后终结.

int main()
{
  boost::asio::io_service io;
  printer p(io);
  boost::thread t(boost::bind(&boost::asio::io_service::run, &io));
  io.run();
  t.join();

  return 0;
}

完整的代码:

#include <iostream>
#include <boost/asio.hpp>
#include <boost/thread.hpp>
#include <boost/bind.hpp>
#include <boost/date_time/posix_time/posix_time.hpp>

class printer
{
public:
  printer(boost::asio::io_service& io)
    : strand_(io),
      timer1_(io, boost::posix_time::seconds(1)),
      timer2_(io, boost::posix_time::seconds(1)),
      count_(0)
  {
    timer1_.async_wait(strand_.wrap(boost::bind(&printer::print1, this)));
    timer2_.async_wait(strand_.wrap(boost::bind(&printer::print2, this)));
  }

  ~printer()
  {
    std::cout << "Final count is " << count_ << "\n";
  }

  void print1()
  {
    if (count_ < 10)
    {
      std::cout << "Timer 1: " << count_ << "\n";
      ++count_;

      timer1_.expires_at(timer1_.expires_at() + boost::posix_time::seconds(1));
      timer1_.async_wait(strand_.wrap(boost::bind(&printer::print1, this)));
    }
  }

  void print2()
  {
    if (count_ < 10)
    {
      std::cout << "Timer 2: " << count_ << "\n";
      ++count_;

      timer2_.expires_at(timer2_.expires_at() + boost::posix_time::seconds(1));
      timer2_.async_wait(strand_.wrap(boost::bind(&printer::print2, this)));
    }
  }

private:
  boost::asio::strand strand_;
  boost::asio::deadline_timer timer1_;
  boost::asio::deadline_timer timer2_;
  int count_;
};

int main()
{
  boost::asio::io_service io;
  printer p(io);
  boost::thread t(boost::bind(&boost::asio::io_service::run, &io));
  io.run();
  t.join();

  return 0;
}

3.7.

Daytime.1 - A synchronous TCP daytime client This tutorial program shows how to use asio to implement a client application with TCP.

We start by including the necessary header files.

#include <iostream> #include <boost/array.hpp> #include <boost/asio.hpp>

The purpose of this application is to access a daytime service, so we need the user to specify the server.

using boost::asio::ip::tcp;

int main(int argc, char* argv[]) {

try {
- if (argc != 2) {
  - std::cerr << "Usage: client <host>" << std::endl; return 1;
  }

All programs that use asio need to have at least one boost::asio::io_service object.

boost::asio::io_service io_service;

We need to turn the server name that was specified as a parameter to the application, into a TCP endpoint. To do this we use an boost::asio::ip::tcp::resolver object.

tcp::resolver resolver(io_service);

A resolver takes a query object and turns it into a list of endpoints. We construct a query using the name of the server, specified in argv[1], and the name of the service, in this case "daytime".

tcp::resolver::query query(argv[1], "daytime");

The list of endpoints is returned using an iterator of type boost::asio::ip::tcp::resolver::iterator. A default constructed boost::asio::ip::tcp::resolver::iterator object is used as the end iterator.

tcp::resolver::iterator endpoint_iterator = resolver.resolve(query); tcp::resolver::iterator end;

Now we create and connect the socket. The list of endpoints obtained above may contain both IPv4 and IPv6 endpoints, so we need to try each of them until we find one that works. This keeps the client program independent of a specific IP version.

tcp::socket socket(io_service); boost::asio::error error = boost::asio::error::host_not_found;
while (error && endpoint_iterator != end) {
- socket.close(); socket.connect(*endpoint_iterator++, boost::asio::assign_error(error));
} if (error)
- throw error;

The connection is open. All we need to do now is read the response from the daytime service.

We use a boost::array to hold the received data. The boost::asio::buffer() function automatically determines the size of the array to help prevent buffer overruns. Instead of a boost::array, we could have used a char [] or std::vector.

for (;;) {
- boost::array<char, 128> buf; boost::asio::error error; size_t len = socket.read_some(
  - boost::asio::buffer(buf), boost::asio::assign_error(error));

When the server closes the connection, the boost::asio::ip::tcp::socket::read_some() function will exit with the boost::asio::error::eof error, which is how we know to exit the loop.

if (error == boost::asio::error::eof)
- break; // Connection closed cleanly by peer.
else if (error)
- throw error; // Some other error.
std::cout.write(buf.data(), len);

}

Finally, handle any exceptions that may have been thrown.

}
catch (std::exception& e) {
- std::cerr << e.what() << std::endl;
}

See the full source listing Return to the tutorial index Next: Daytime.2 - A synchronous TCP daytime server

4. Filesystem Library 文件系统