url.cpp

小型搜索引擎,用C/C++编写,属于全文搜索引擎

/* URL handling
 */

#include <iostream>
#include <string>
#include "Url.h"

/* Is X "."?  */
#define DOTP(x) ((*(x) == '.') && (!*(x + 1)))
/* Is X ".."?  */
#define DDOTP(x) ((*(x) == '.') && (*(x + 1) == '.') && (!*(x + 2)))

struct scheme_data
{
	char *leading_string;
	int default_port;
	int enabled;
};

/* Supported schemes: */
static struct scheme_data supported_schemes[] =
{
	{ "http://",  DEFAULT_HTTP_PORT,  1 },
	{ "ftp://",   DEFAULT_FTP_PORT,   1 },

	/* SCHEME_INVALID */
	{ NULL,       -1,                 0 }
};

/* Returns the scheme type if the scheme is supported, or
   SCHEME_INVALID if not.  */
void CUrl::ParseScheme (const char *url)
{
	int i;

	for (i = 0; supported_schemes[i].leading_string; i++)

		if (0 == strncasecmp (url, supported_schemes[i].leading_string,
                          strlen (supported_schemes[i].leading_string))) {

			if (supported_schemes[i].enabled){
				this->m_eScheme = (enum url_scheme) i;
				return;
			}else{
				this->m_eScheme = SCHEME_INVALID;
				return;
			}
		}

	this->m_eScheme = SCHEME_INVALID;
	return;
}


/* scheme://user:pass@host[:port]... 
 *                    ^              
 * We attempt to break down the URL into the components path,
 * params, query, and fragment.  They are ordered like this:
 * scheme://host[:port][/path][;params][?query][#fragment] 
 */

bool CUrl::ParseUrl(string strUrl)
{
	string::size_type idx;

	this->ParseScheme(strUrl.c_str());	
	if( this->m_eScheme != SCHEME_HTTP )
		return false;

	// get host name
	this->m_sHost = strUrl.substr(7);
	idx = m_sHost.find('/');
	if(idx != string::npos){
		m_sHost = m_sHost.substr(0,idx);
	}

	this->m_sUrl = strUrl;

	return true;
}

CUrl::CUrl()
{
	this->m_sUrl = ""; 
	this->m_eScheme= SCHEME_INVALID;
        
	this->m_sHost = "";  
	this->m_nPort = DEFAULT_HTTP_PORT; 
        
	this->m_sPath = "";
	this->m_sParams = "";
	this->m_sQuery = "";
	this->m_sFragment = "";

	this->m_sDir = "";
	this->m_sFile = "";
        
        this->m_sUser = "";
	this->m_sPasswd = "";

}

CUrl::~CUrl()
{

}