Leaked source code of windows server 2003
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

438 lines
8.6 KiB

/*++
Copyright (c) 1996 Microsoft Corporation
Module Name :
linkpars.cpp
Abstract:
Link parser class implementation. This class responsible for
parsing the html file for hyperlink.
Author:
Michael Cheuk (mcheuk)
Project:
Link Checker
Revision History:
--*/
#include "stdafx.h"
#include "LinkPars.h"
#include "link.h"
#include "lcmgr.h"
#ifdef _DEBUG
#define new DEBUG_NEW
#undef THIS_FILE
static char THIS_FILE[] = __FILE__;
#endif
// Constants
const CString strLocalHost_c(_T("localhost"));
void
CLinkParser::Parse(
const CString& strData,
const CString& strBaseURL,
CLinkPtrList& rLinkPtrList
)
/*++
Routine Description:
Parse a page of html data
Arguments:
strData - page of html
strBaseURL - base URL
rLinkPtrList - reference to links list. The new links will
will be added to this list.
Return Value:
N/A
--*/
{
// Look for the first '<'
LPCTSTR lpszOpen = _tcschr(strData, _TUCHAR('<'));
while(lpszOpen != NULL)
{
// Look for the '>'
LPCTSTR lpszClose = _tcschr(lpszOpen, _TUCHAR('>'));
if(lpszClose)
{
// The possible tag must be longer than 7 bytes (a href=)
int iCount = (int)(lpszClose - lpszOpen) - 1; // skip the '<'
if( iCount > 7 )
{
int iIndex = lpszOpen - ((LPCTSTR)strData) + 1; // skip the '<'
CString strPossibleURL(strData.Mid(iIndex, iCount));
// Parse the possible tag
if(ParsePossibleTag(strPossibleURL))
{
CString strURL;
BOOL fLocalLink;
// We found a valid tag. Time to create new link.
if( CreateURL(strPossibleURL, strBaseURL, strURL, fLocalLink) )
{
rLinkPtrList.AddLink(strURL, strBaseURL, strPossibleURL, fLocalLink);
}
}
}
}
// Look for the next '<'
lpszOpen = _tcschr(++lpszOpen, _TUCHAR('<'));
}
} // CLinkParser::Parse
BOOL
CLinkParser::ParsePossibleTag(
CString& strTag
)
/*++
Routine Description:
Parse a single "<.....>" for possible hyperlink
Arguments:
strTag - value inside a "<.....>" excluding '<' & '>'
If this is a hyperlink tag, the hyperlink URL
will be put in strTag.
Return Value:
BOOL - TRUE if hyperlink tag. FALSE otherwise.
--*/
{
// Make a working copy
CString strWorkCopy(strTag);
// Let's work with lower case
strWorkCopy.MakeLower();
//
// Check for,
//
// HyperLink:
// <a href="url" ...>
// <a href="url#anchor" ...>
// <a href="#anchor" ...>
//
// CGI
// <a href="url?parameters" ...>
//
// Style Sheet
// <link rel="stylesheet" href="url" ...>
//
if( strWorkCopy[0] == _T('a') ||
strWorkCopy.Find(_T("link")) == 0 )
{
return GetTagValue(strTag, CString(_T("href")));
}
//
// Check for,
//
// <body background="url" ...>
//
// Table:
// <table background="url" ...>
// <th background="url" ...>
// <td background="url" ...>
//
else if( strWorkCopy.Find(_T("body")) == 0 ||
strWorkCopy.Find(_T("table")) == 0 ||
strWorkCopy.Find(_T("th")) == 0 ||
strWorkCopy.Find(_T("td")) == 0 )
{
return GetTagValue(strTag, CString(_T("background")));
}
//
// Check for,
//
// Sound:
// <bgsound src="url" ...>
// <sound src="url" ...>
//
// Frame:
// <frame src="url" ...>
//
// Netscape embeded:
// <embed src="url" ...>
//
// JavaScript & VB Script
// <script src="url" language="java or vbs" ...>
//
else if( strWorkCopy.Find(_T("bgsound")) == 0 ||
strWorkCopy.Find(_T("sound")) == 0 ||
strWorkCopy.Find(_T("frame")) == 0 ||
strWorkCopy.Find(_T("embed")) == 0 ||
strWorkCopy.Find(_T("script")) == 0 )
{
return GetTagValue(strTag, CString(_T("src")));
}
// Check for,
//
// Image:
// <img src="url" ...>
//
// Video:
// <img dynsrc="url">
//
// VRML:
// <img vrml="url">
//
else if( strWorkCopy.Find(_T("img")) == 0 )
{
if(GetTagValue(strTag, CString(_T("src"))))
{
return TRUE;
}
if(GetTagValue(strTag, CString(_T("dynsrc"))))
{
return TRUE;
}
return GetTagValue(strTag, CString(_T("vrml")));
}
// Java
// <applet code="name.class" codebase="url" ...>
else if( strWorkCopy.Find(_T("applet")) == 0 )
{
return GetTagValue(strTag, CString(_T("codebase")));
}
// Form
// <form action="url" ...>
else if( strWorkCopy.Find(_T("form")) == 0 )
{
return GetTagValue(strTag, CString(_T("action")));
}
return FALSE;
} // CLinkParser::ParsePossibleTag
BOOL
CLinkParser::GetTagValue(
CString& strTag,
const CString& strParam
)
/*++
Routine Description:
Get the hyperlink value from "<.....>"
Arguments:
strTag - value inside a "<.....>" excluding '<' & '>'
If this is a hyperlink tag, the hyperlink URL
will be put in strTag.
strParam - parameter to look for. For example, src or href
Return Value:
BOOL - TRUE if hyperlink tag. FALSE otherwise.
--*/
{
// Make a copy of original tag
CString strWorkCopy(strTag);
strWorkCopy.MakeLower();
int iLength = strParam.GetLength();
// Look for the parameter
int iIndex = strWorkCopy.Find(strParam);
if(iIndex == -1)
{
return FALSE;
}
// Remove the parameter from the tag
CString strResult( strTag.Mid(iIndex + iLength) );
// Look for '='
iIndex = strResult.Find(_T("="));
if(iIndex == -1)
{
return FALSE;
}
// Remove the '=' from the tag
strResult = strResult.Mid(iIndex+1);
// Look for the value
int iStart = -1;
int iEnd = -1;
int fPara = FALSE; // is the tag start with "
// Search for the value
for(int i=0; i<strResult.GetLength(); i++)
{
// If we found the starting index of value, look
// for the end of the value
if(iStart!=-1 &&
( !fPara && strResult[i] == _TCHAR(' ') ||
( fPara && strResult[i] == _TCHAR('\"') )
)
)
{
iEnd = i;
break;
}
// Look for the starting index of value
if(iStart==-1 && strResult[i] != _TCHAR(' ') && strResult[i] != _TCHAR('\"') )
{
iStart = i;
if(i - 1 >= 0)
{
fPara = (strResult[i-1] == _TCHAR('\"')); // found a "
}
}
}
// Found the starting index
if(iStart != -1)
{
// If we didn't find the end of value, use the
// last character as end
if(iEnd == -1)
{
iEnd = strResult.GetLength();
}
// Copy the value to the input
strTag = strResult.Mid(iStart, (iEnd - iStart));
// Change '\' to '/'
CLinkCheckerMgr::ChangeBackSlash(strTag);
return TRUE;
}
return FALSE;
} // CLinkParser::GetTagValue
BOOL
CLinkParser::CreateURL(
const CString& strRelativeURL,
const CString& strBaseURL,
CString& strURL,
BOOL& fLocalLink
)
/*++
Routine Description:
Create a URL from base URL & relative URL. It also check
the result for local or remote link
Arguments:
strRelativeURL - relative URL
strBaseURL - base URL
strURL - result URL
fLocalLink - will be set to TRUE if this is a local link
Return Value:
BOOL - TRUE if sucess. FALSE otherwise.
--*/
{
ASSERT(CWininet::IsLoaded());
// Remove the anchor from the relative URL
CString strNewRelativeURL(strRelativeURL);
int i = strNewRelativeURL.ReverseFind(_TCHAR('#'));
if(i != -1)
{
strNewRelativeURL = strNewRelativeURL.Left(i);
}
// Combine the URLs
DWORD dwLength = INTERNET_MAX_URL_LENGTH;
LPTSTR lpBuffer = strURL.GetBuffer(dwLength);
CWininet::InternetCombineUrlA(
strBaseURL,
strNewRelativeURL,
lpBuffer,
&dwLength,
ICU_ENCODE_SPACES_ONLY);
strURL.ReleaseBuffer();
// Check for local or remote link
URL_COMPONENTS urlcomp;
memset(&urlcomp, 0, sizeof(urlcomp));
urlcomp.dwStructSize = sizeof(urlcomp);
urlcomp.dwHostNameLength = 1;
VERIFY(CWininet::InternetCrackUrlA(strURL, strURL.GetLength(), NULL, &urlcomp));
// Check for possible local link
if((int)urlcomp.dwHostNameLength == m_strLocalHostName.GetLength() ||
(int)urlcomp.dwHostNameLength == strLocalHost_c.GetLength()) // localhost
{
if( _tcsnccmp( urlcomp.lpszHostName, m_strLocalHostName, m_strLocalHostName.GetLength() ) == 0 ||
_tcsnccmp( urlcomp.lpszHostName, strLocalHost_c, strLocalHost_c.GetLength() ) == 0)
{
fLocalLink = TRUE;
// Local link
if(GetLinkCheckerMgr().GetUserOptions().IsCheckLocalLinks())
{
return TRUE;
}
else
{
return FALSE;
}
}
}
// Remote link
fLocalLink = FALSE;
if(GetLinkCheckerMgr().GetUserOptions().IsCheckRemoteLinks())
{
return TRUE;
}
else
{
return FALSE;
}
} // CLinkParser::CreateURL