cgetwebpage.h

可以进行膜板定制的动态网页下载分析的源程序

//#include <TIME.H>
//#include "IGet_Public.h"
#include <time.h>
#ifdef WIN32
#include <winsock2.h> 
#else
#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <string.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <fcntl.h>
#include <sys/socket.h>
#include <arpa/inet.h>
#include <netdb.h>
#include <pthread.h>
#include <unistd.h>
#endif
#define MAX_INET_BUFFER 1024000
#define REDIRECTED_BUT_NOT_FOUND      -8801
#define TIMEOUT_ON_REQUEST            -8802
#define PAGE_OR_FILE_NOT_FOUNT        -8803
#define ACCESS_DENIED                 -8804
#define PROXY_AUTHENTICATION_REQUIRED -8805
#define WININET_ERROR                 -8806
#define OPEN_SESSION_ERROR            -8807
#define DYNBUFFER 151200
typedef struct _BUFFERADDR
{
	char curlhost[128];
	struct sockaddr_in serverlast;
	
} bufferaddr;//缓冲IP地址

class cgetwebpage  
{
	char chttphead[10240];
	char chttpget[10240];
	char Bufferhead[10240];
	char Buffer[DYNBUFFER];
	bufferaddr gbufferaddr;
	int m_nport;
	char m_chost[128];
	char m_curlbak[51200];
	char m_cookie[10240];
	int standardurl(char *url,char *newurl);
	char *getmyhostname(char *url);
	int detecttime(char *pmodifyplace);
	int analypagehead(char *ppagehead,time_t lasttime);
	int getwebstatus(char *ppagehead);
	int getredirectfun(char *ppagehead);
	int IsUrl(char *url);
	char *outgetmodheadinfo(char *buffer);
	int getapptype(char *pheader);
public:
	time_t FileTime;
	int m_dwRet;
	int FileSize;
	char m_credirecturl[10000];
	char *HTMLFileBuf;//[MAX_INET_BUFFER+2];
	cgetwebpage();
	virtual ~cgetwebpage();
	int DoGet(char *strU, const char *httpHeaders, const char *httpBodys,int PageType,int AcceptType,time_t fTime,int flag=0);
};