C＋＋抓取网页内容
Windows VC下的做法：

#include <stdio.h>
#include <afxinet.h>
 
int main(int argc, char* argv[])
{
    CInternetSession session("HttpClient");
    char * url = " http://www.imobile.com.cn/simcard.php?simcard=1392658";
    CHttpFile* pfile = (CHttpFile *)session.OpenURL(url);
    
    DWORD dwStatusCode;
    pfile -> QueryInfoStatusCode(dwStatusCode);
    if(dwStatusCode == HTTP_STATUS_OK)
    {
        CString content;
        CString data;
        while (pfile -> ReadString(data))
        {
            content  += data + "\r\n";
        }
        content.TrimRight();
        printf(" %s\n " ,(LPCTSTR)content);
    } 
    pfile -> Close();
    delete pfile;
    session.Close();
    
    return   0 ;
}

Windows下用socket：

#include <string>
#include <iostream>
#include <fstream>
#include "winsock2.h"
#include <time.h>

#pragma comment(lib, "ws2_32.lib")

using namespace std;

#define DEFAULT_PAGE_BUF_SIZE 1048576

void main()
{
    WSADATA wsaData;
    int err;
    err = WSAStartup(MAKEWORD(2,2), &wsaData);
    if( err != 0 )
    {
        return;
    }

    // timer is start

    clock_t start, finish;
    double duration;
    start = clock();

    char host[] = "www.sina.com.cn";
    char *request = "GET / HTTP/1.0\r\nHost: www.sina.com.cn\r\nConnection: Close\r\n\r\n";

    struct hostent *hp;
    hp = gethostbyname(host);
    if(hp == NULL)
    {
        cout << "gethostbyname() error in GetIpByHost: " << host << endl;
        return;
    }
  
    // 获取域名对应的IP

    struct in_addr inAddr;
    LPSTR lpAddr;
    lpAddr = hp->h_addr;
    memmove(&inAddr,lpAddr,4);

    int sock, ret = 0, optval = 1;
    struct sockaddr_in sa;
    sa.sin_family = AF_INET;       
    sa.sin_port = htons(80);
    sa.sin_addr.s_addr = inet_addr(inet_ntoa(inAddr));
  
    sock = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP);
    connect(sock, (SOCKADDR*)&sa, sizeof(sa));
    if(sock == -1)
    {
        return;
    }
    if(sock == -2)
    {
        return;
    }

    // send the "GET" data

    ret = send(sock, request, strlen(request), 0);

    // 网页内容长度。可以从http头部数据中获取 "Content-Length:"

    int m_nContentLength = DEFAULT_PAGE_BUF_SIZE;
  
    char *pageBuf;
    pageBuf = (char *)malloc(m_nContentLength);
    memset(pageBuf, 0, m_nContentLength);

    int bytesRead = 0;
    while(ret > 0)
    {
        ret = recv(sock, pageBuf + bytesRead, m_nContentLength - bytesRead, 0);
       
        if(ret > 0)
        {
            bytesRead += ret;
        }
    }
    pageBuf[bytesRead] = '\0';

    cout << bytesRead << endl;

    // write the html content to the file

    ofstream ofs;
    ofs.open("ofs.txt");
    ofs << pageBuf << endl;

    ofs.close();
    free(pageBuf);
    closesocket(sock);
    WSACleanup();

    // timer is finish

    finish = clock();
    duration = (double)(finish - start) / CLOCKS_PER_SEC;
    cout << "have cost " << duration << " seconds\n";
  
    return;
}

其他如不从缓存中读取内容及如何使用代理连接现在就不说了，可以参考下面的链接，或者下次补上。另外不妨看看 Java 是如何读取 URL 内容的，更简单

GetMethod httpMethod = new GetMethod("http://unmi.blogcn.com"); 
int statusCode = new HttpClient().executeMethod(httpMethod); 
if(statusCode == HttpStatus.SC_OK) 
{ 
    System.out.println(httpMethod.getResponseBodyAsString()); 
} 
httpMethod.releaseConnection(); 
GetMethod httpMethod = new GetMethod("http://unmi.blogcn.com");
int statusCode = new HttpClient().executeMethod(httpMethod);
if(statusCode == HttpStatus.SC_OK)
{
    System.out.println(httpMethod.getResponseBodyAsString());
}
httpMethod.releaseConnection();
 
内容取过来之后，总是希望从中拣出需要的数据，可惜 VC6 中没有自己的正则表达式库，所以下一步要学用 boost 的正则表达式库。

Linux 下最简单，三种办法：

如果在C程序想抓取网页内容，比如百度
方法1. 执行System("wget http：//www.baidu.com -q -O baidu.html")，然后再分析baidu.html。
方法2. system(curl http://www.baidu.com)
方法3. 直接写socket程序获取http://www.baidu.com的内容存入buffer中。

参考链接： 1. VC++6.0 通过HTTP方式获取网页
         2. CInternetSession获取网页内容的问题
         3. 利用CInternetSession从网站获取信息，并利用CString拆分查找特定信息