Linux下C语言用socket获取网页源码…

最新推荐文章于 2021-05-21 06:30:22 发布

原创最新推荐文章于 2021-05-21 06:30:22 发布 · 3.7k 阅读

0 ·

CC 4.0 BY-SA版权

Linux服务器专栏收录该内容

42 篇文章

订阅专栏

本文介绍了使用C++进行网页抓取的三种方法：利用Wget工具、通过socket编程以及使用libcurl库。展示了如何通过这些方法获取网页源码，并提供了详细的代码示例。

第一个为利用linux下的工具来获取网页源码，我用的是Wget，也可以使用Curl，curl的话更加的灵活，可以设置很多参数

C++代码

//通过Wget来获取网页

string GetHtmlByWget(string url)
{
//获取待下载网页文件名
string fileName = url.substr((int)url.find_last_of("/") + 1);
if(fileName != "")
{
string strCom = "wget -q "; //wget命令，-q表示不显示下载信息
strCom.append(url);
system(strCom.c_str()); //执行wget
ifstream fin(fileName.c_str());
if(!fin)
{
return "";
}
string strHtml = "";
char chTemp[1024] = "";
//读取网页文件到内存中
while(fin.getline(chTemp , 1024))
{
strHtml.append(string(chTemp));
strcpy(chTemp , "");
}
fin.close();
strCom = "rm -f "; //删除文件命令,-f表示直接删除不做任何提示
strCom.append(fileName);
system(strCom.c_str()); //删除刚才下载下来的文件
return strHtml; //返回网页源码
}
else
{
return "";
}
}

//通过Wget来获取网页
string GetHtmlByWget(string url)
{
    //获取待下载网页文件名
    string fileName = url.substr((int)url.find_last_of("/") + 1);
    if(fileName != "")
    {
        string strCom = "wget -q "; //wget命令，-q表示不显示下载信息
        strCom.append(url);
        system(strCom.c_str()); //执行wget

        ifstream fin(fileName.c_str());
        if(!fin)
        {
            return "";
        }
        string strHtml = "";
        char chTemp[1024] = "";
        //读取网页文件到内存中
        while(fin.getline(chTemp , 1024))
        {
            strHtml.append(string(chTemp));
            strcpy(chTemp , "");
        }
        fin.close();
        strCom = "rm -f ";  //删除文件命令,-f表示直接删除不做任何提示
        strCom.append(fileName);
        system(strCom.c_str()); //删除刚才下载下来的文件
        return strHtml; //返回网页源码
    }
    else
    {
        return "";
    }
}

第二个是用的socket的来获取源码
C++代码
//通过GET获取网页源码
string GetHtmlByGet(string url)
{
    string strHtmlContent = "";
    int sockfd;
    struct sockaddr_in addr;
    struct hostent *pURL;
    char text[RECVBUF];

    //分析链接
    UrlInfo urlInfo = ParseURL(url);
    string sAccept = "Accept: **\r\nAccept-Language: zh-cn\r\nAccept-Encoding: gzip, deflate";
    //不同的主机UserAgent不同
    string sUserAgent = "Mozilla/5.0 (X11; U; Linux i686; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.224 Safari/534.10";
    //将端口转换为字符串
    char t[6];
    string strPort;
    sprintf(t,"%d", urlInfo.Port);
    strPort = t;
    //构造发送字符串
    string strRequest = "";
    strRequest.append("GET ");
    strRequest.append(urlInfo.File);
    strRequest.append("?");
    strRequest.append(urlInfo.Body);
    strRequest.append(" HTTP/1.1\r\n");
    strRequest.append(sAccept);
    strRequest.append("\r\nUser-Agent:");
    strRequest.append(sUserAgent);
    strRequest.append("\r\nHost:");
    strRequest.append(urlInfo.Host);
    strRequest.append(":");
    strRequest.append(strPort);
    strRequest.append("\r\nConnection: Keep-Alive\r\n\r\n");

    char* host = const_cast<char*>(urlInfo.Host.c_str());
    sockfd = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP); //TCP方式发送
    pURL = gethostbyname(host);
    addr.sin_family = AF_INET;
    addr.sin_addr.s_addr = *((unsigned long*)pURL->h_addr);
    addr.sin_port = htons(80);

    //连接
    connect(sockfd,(struct sockaddr *)&addr,sizeof(addr));
    //发送
    send(sockfd, const_cast<char*>(strRequest.c_str()), strRequest.length(), 0);
    //接受
    while(recv(sockfd, text, RECVBUF, 0) > 0)
    {
        strHtmlContent.append(text);
        bzero(text,RECVBUF);
    }
    //关闭socket
    close(sockfd);
    //返回接受结果
    return strHtmlContent;
}

使用libcurl
1. Java代码
  1. #include <stdio.h>
  2. #include <string.h>
  3. #include <curl/curl.h>
  5. #define MAX_BUF 65536
  7. char wr_buf[MAX_BUF+1];
  8. int wr_index;
  11. size_t write_data( void *buffer, size_t size, size_t nmemb, void *userp )
  12. {
  13. int segsize = size * nmemb;
  16. if ( wr_index + segsize > MAX_BUF ) {
  17. *(int *)userp = 1;
  18. return 0;
  19. }
  22. memcpy( (void *)&wr_buf[wr_index], buffer, (size_t)segsize );
  25. wr_index += segsize;
  28. wr_buf[wr_index] = 0;
  31. return segsize;
  32. }
  36. int main( void )
  37. {
  38. CURL *curl;
  39. CURLcode ret;
  40. int wr_error;
  42. wr_error = 0;
  43. wr_index = 0;
  46. curl = curl_easy_init();
  47. if (!curl) {
  48. printf("couldn't init curl\n");
  49. return 0;
  50. }
  53. curl_easy_setopt( curl, CURLOPT_URL, "www.exampledomain.com" );
  56. curl_easy_setopt( curl, CURLOPT_WRITEDATA, (void *)&wr_error );
  57. curl_easy_setopt( curl, CURLOPT_WRITEFUNCTION, write_data );
  60. ret = curl_easy_perform( curl );
  62. printf( "ret = %d (write_error = %d)\n", ret, wr_error );
  65. if ( ret == 0 ) printf( "%s\n", wr_buf );
  67. curl_easy_cleanup( curl );
  69. return 0;
  70. }
http://yang7229693.iteye.com/blog/855208