好用的http client库CPP REST SDK

前言

C++中http client库自己就少，好用的就更少了，在了解微软开源的CPP REST SDK库以前，我知道的C++ http client库有libcurl（这个是C语言的），Qt的QNetworkAccessManager，还有VC++ http client，Qt的QNetworkAccessManager库我在开发CZPlayer的时候用来下载过音乐、专辑图片和歌词，不得不说Qt提供的API仍是比较好用的，若是不涉及界面开发，难道咱们在linux上就只能用libcurl，在windows上就用VC++的http client？答案是否认的，在绝望之际CPP REST SDK出如今个人眼前，CPP REST SDK是微软开源的基于PPL的异步http client，网络层使用的是Boost.Asio，跨平台，而且支持json解析，在使用CPP REST SDK以前要确保你已经安装了boost和openssl，下面是微软官方提供的例子。

微软官方例子

#include <cpprest/http_client.h>
#include <cpprest/filestream.h>

using namespace utility;                    // Common utilities like string conversions
using namespace web;                        // Common features like URIs.
using namespace web::http;                  // Common HTTP functionality
using namespace web::http::client;          // HTTP client features
using namespace concurrency::streams;       // Asynchronous streams

int main(int argc, char* argv[])
{
    auto fileStream = std::make_shared<ostream>();

    // Open stream to output file.
    pplx::task<void> requestTask = fstream::open_ostream(U("results.html")).then([=](ostream outFile)
    {
        *fileStream = outFile;

        // Create http_client to send the request.
        http_client client(U("http://www.bing.com/"));

        // Build request URI and start the request.
        uri_builder builder(U("/search"));
        builder.append_query(U("q"), U("cpprestsdk github"));
        return client.request(methods::GET, builder.to_string());
    })

    // Handle response headers arriving.
    .then([=](http_response response)
    {
        printf("Received response status code:%u\n", response.status_code());

        // Write response body into the file.
        return response.body().read_to_end(fileStream->streambuf());
    })

    // Close the file stream.
    .then([=](size_t)
    {
        return fileStream->close();
    });

    // Wait for all the outstanding I/O to complete and handle any exceptions
    try
    {
        requestTask.wait();
    }
    catch (const std::exception &e)
    {
        printf("Error exception:%s\n", e.what());
    }

    return 0;
}

上面的例子主要内容是访问一个网站并将该内容保存在results.html里面，这里用到了微软自家的PPL并行计算库（还有一个是英特尔的TBB），该例子使用lambda表达式做为异步回调的handler，打开文件流、请求、回应和关闭文件流都是异步的，为了方便起见，咱们将该例子改成同步方式。

使用同步方式

auto fileStream = std::make_shared<ostream>();
ostream outFile = fstream::open_ostream(U("results.html")).get();
*fileStream = outFile;

// Create http_client to send the request.
http_client client(U("http://www.bing.com/"));

// Build request URI and start the request.
uri_builder builder(U("/search"));
builder.append_query(U("q"), U("cpprestsdk github"));
http_response response = client.request(methods::GET, builder.to_string()).get();
// Write response body into the file.
response.body().read_to_end(fileStream->streambuf()).get();
fileStream->close().get();

使用同步方式的代码比较清晰，能够看到每个函数调用后面都会调用get函数，由于CPP REST SDK是基于PPL的，因此在request、read_to_end、close等函数调用后都会返回一个Task对象，而Task里面的get和wait函数是等待任务执行完成。

实战

我以前在看别人博客的时候看到每一篇博客前都有一副图，看起来比较好看，听博主讲他是使用bing的每日一图，后来我也效仿，我在主题之家找到了本身比较喜欢的图片，最开始本身写博客的时候都是直接去主题之家一页一页的找看，而后再下载图片，写了几篇博客以后感受每次都要去下载图片感受有点low，因此我决定直接将主题之家的某类型的图片都下载下来，这样选图片就方便多了，下面是我在主题之家抓图的代码。

#include <iostream>
#include <string>
#include <vector>
#include <fstream>
#include <memory>
#include <regex>
#include <cpprest/http_client.h>
#include <cpprest/filestream.h>
#include <cpprest/containerstream.h>

// 请求并解析url
bool get_result(const std::string& url, const std::string& pattern, std::vector<std::string>& vec)
{
    try
    {
        web::http::client::http_client client(web::uri(utility::conversions::to_string_t(url)));
        web::http::http_response response = client.request(web::http::methods::GET).get();

        concurrency::streams::stringstreambuf buffer;
        response.body().read_to_end(buffer).get();
        std::string& str = buffer.collection();
        
            // 使用C++11提供的正则表达式库
        std::regex r(pattern);
        for (std::sregex_iterator iter(str.begin(), str.end(), r), end; iter != end; ++iter)
        {
            std::cout << iter->str() << std::endl;
            vec.emplace_back(iter->str());
        }
    }
    catch (std::exception& e)
    {
        std::cout << "Exception: " << e.what() << std::endl;
        return false;
    }

    return true;
}

// 获取图片
bool get_result(const std::string& url, std::string& picture)
{
    try
    {
        web::http::client::http_client client(web::uri(utility::conversions::to_string_t(url)));
        web::http::http_response response = client.request(web::http::methods::GET).get();

        concurrency::streams::stringstreambuf buffer;
        response.body().read_to_end(buffer).get();
        picture = buffer.collection();
    }
    catch (std::exception& e)
    {
        std::cout << "Exception: " << e.what() << std::endl;
        return false;
    }

    return true;
}

// 保存图片
bool write_to_file(const std::string& file_path, const std::string& data)
{
    try
    {
        std::ofstream file;
        file.open(file_path, std::ios::out | std::ios::binary);
        if (!file.good())
        {
            return false;
        }
        file.write(data.c_str(), data.size());
        file.close();
    }
    catch (std::exception& e)
    {
        std::cout << "Exception: " << e.what() << std::endl;
        return false;
    }

    return true;
}

int main()
{
    // [1] 请求每一页，将子页面的url保存在sub_url_vec里面
    std::vector<std::string> sub_url_vec;
    std::string pattern = "/desk/[0-9]+.htm";
    for (int i = 1; i <= 32; ++i)
    {
        // 创意主题
        std::string url = "http://www.51ztzj.com/dbizhi/category_27_" + std::to_string(i) + ".htm#content_anchor";
        std::cout << "Start get " << i << " page, url: " << url << std::endl;
        // 请求并解析url
        if (!get_result(url, pattern, sub_url_vec))
        {
            std::cout << "Get " << i << " page failed" << std::endl;
        }
    }

    // 最终的图片url：http://img.51ztzj.com//upload/image/20130220/2013022014_670x419.jpg
    // [2] 将子页面的图片url解析出来放入picture_url_vec
    std::vector<std::string> picture_url_vec;
    pattern = "http://img.51ztzj.com//upload/image/.+/.+_670x419.jpg";
    for (std::size_t i = 0; i < sub_url_vec.size(); ++i)
    {
        std::string url = "http://www.51ztzj.com" + sub_url_vec[i];
        std::cout << "Start get " << i + 1 << " sub page, url: " << url << std::endl;
        // 请求并解析url
        if (!get_result(url, pattern, picture_url_vec))
        {
            std::cout << "Get " << i + 1 << " sub page failed" << std::endl;
        }
    }

    // [3] 最后遍历picture_url_vec，而后一个一个的下载图片
    for (std::size_t i = 0; i < picture_url_vec.size(); ++i)
    {
        std::cout << "Start download " << i + 1 << " picture, url: " << picture_url_vec[i] << std::endl;
        std::string picture;
        // 获取图片
        if (!get_result(picture_url_vec[i], picture))
        {
            std::cout << "Download " << i + 1 << " picture failed" << std::endl;
        }

        std::string file_path = "./download/" + std::to_string(i) + ".jpg";
        // 保存图片
        if (!write_to_file(file_path, picture))
        {
            std::cout << "Write to file failed: " << i + 1 << std::endl;
        }
    }

    return 0;
}