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.
 
 
 
 
 
 

2392 lines
73 KiB

//+-------------------------------------------------------------------------
//
// Microsoft Windows
// Copyright (C) Microsoft Corporation, 1996 - 2000.
//
// File: linkhits.cxx
//
// Contents: classes to insert links among query hits in HTML format
//
//--------------------------------------------------------------------------
#include<pch.cxx>
#pragma hdrstop
#include <locale.h>
#include <mbutil.hxx>
#include <htmlchar.hxx>
#include <codepage.hxx>
#include <cgiesc.hxx>
#include <weblcid.hxx>
#include "webdbg.hxx"
#include "whmsg.h"
#include "linkhits.hxx"
#include "whtmplat.hxx"
extern int _cdecl ComparePositions(const void* pPos1, const void* pPos2);
const WCHAR VPathWebHitsFile[]=L"CiWebHitsFile";
const WCHAR CommandLineVarName[]=L"QUERY_STRING";
const WCHAR RestrictionVarName[]=L"CiRestriction";
const WCHAR IDQFilenameVarName[]=L"CiQueryFile";
const WCHAR WTXFilenameVarName[]=L"CiTemplateFile";
const WCHAR HiliteTypeVarName[]=L"CiHiliteType";
const WCHAR ColorVarName[]=L"CiHiliteColor";
const WCHAR BoldVarName[]=L"CiBold";
const WCHAR ItalicVarName[]=L"CiItalic";
const WCHAR MaxLineLength[]=L"CiMaxLineLength";
const WCHAR LocaleVar[]=L"CiLocale";
const WCHAR BeginHiliteVar[]=L"CiBeginHilite";
const WCHAR EndHiliteVar[]=L"CiEndHilite";
const WCHAR NullHTWFile[]=L"null.htw";
const WCHAR CodepageVar[]=L"CiCodepage";
const WCHAR DialectVar[]=L"CiDialect";
const WCHAR ParaTag[]=L"<P>\n";
const WCHAR HRule[]=L"<HR>\n";
const WCHAR Red24BitMask[]=L"#FF0000";
const WCHAR Blue24BitMask[]=L"#0000FF";
const WCHAR Green24BitMask[]=L"#00FF00";
const WCHAR Black24BitMask[]=L"#000000";
const WCHAR Yellow24BitMask[]=L"#FFFF00";
//
// List of replacable parameters in the htx file for output generation.
//
const WCHAR wcsParamCiUrl[] = L"CIURL";
const WCHAR wcsParamRestriction[] = L"CIRESTRICTION";
const WCHAR wcsCharSet[] = L"CICHARSET";
const WCHAR wcsLocale[] = L"CILOCALE";
const WCHAR wcsCodepage[] = L"CICODEPAGE";
const WCHAR wcsUserParamPrefix[] = L"CIUSERPARAM";
const ULONG cwcUserParamPrefix = (sizeof(wcsUserParamPrefix)/sizeof(WCHAR))-1;
const WCHAR * awcsUserParamNames[] = {
L"CIUSERPARAM1",
L"CIUSERPARAM2",
L"CIUSERPARAM3",
L"CIUSERPARAM4",
L"CIUSERPARAM5",
L"CIUSERPARAM6",
L"CIUSERPARAM7",
L"CIUSERPARAM8",
L"CIUSERPARAM9",
L"CIUSERPARAM10",
};
//+-------------------------------------------------------------------------
//
// Member: CInternalQuery::CinternalQuery, public constructor
//
// Arguments: [rGetEnvVars] - object that contains CGI env. variables
// [rPList] - property list of properties to query
// [lcid] - LCID (locale identifier)
//
// Synopsis: rGetEnvVars yields the textual restriction.
// GetStringDbRestriction then converts it into a DbRestriction.
//
//--------------------------------------------------------------------------
CInternalQuery::CInternalQuery(
CGetEnvVars& rGetEnvVars,
CEmptyPropertyList& rPList,
LCID lcid ):
_pDbRestriction(0),
_pISearch(0)
{
TRY
{
_pDbRestriction = GetStringDbRestriction( rGetEnvVars.GetRestriction(),
rGetEnvVars.GetDialect(),
&rPList,
lcid);
}
CATCH( CException, e)
{
webDebugOut(( DEB_ERROR, "WEBHITS: failed to get DbRestriction\n" ));
delete _pDbRestriction;
RETHROW();
}
END_CATCH
}
void CInternalQuery::CreateISearch( WCHAR const * pwszPath )
{
Win4Assert( 0 == _pISearch );
SCODE sc = MakeISearch( &_pISearch, _pDbRestriction, pwszPath );
if ( FAILED( sc ) )
THROW( CException( sc ) );
} //CreateISearch
//+-------------------------------------------------------------------------
//
// Member: CURLUnescaper::CURLUnescaper public constructor
//
//--------------------------------------------------------------------------
CURLUnescaper::CURLUnescaper( UINT codePage ) :
_codePage(codePage)
{
}
//+---------------------------------------------------------------------------
//
// Member: CURLUnescaper::UnescapeAndConvertToUnicode
//
// Synopsis: Converts the given multi-byte string into a unicode string
// based on the code page. Decodes URL escape sequences along
// the way.
//
// Arguments: [pszMbStr] - smart array pointer to input string
// [cch] - length of input string
// [xwcsBuffer] -
//
// Returns: Number of characters in target buffer, excluding the
// terminating NULL.
//
// History: 24 Nov 1997 AlanW Created
//
//----------------------------------------------------------------------------
ULONG CURLUnescaper::UnescapeAndConvertToUnicode( char const * pszMbStr,
ULONG cch,
XArray<WCHAR> & xwcsBuffer )
{
if ( xwcsBuffer.Get() == 0 || cch+1 > xwcsBuffer.Count() )
{
delete [] xwcsBuffer.Acquire();
xwcsBuffer.Init( cch+1 );
}
DecodeURLEscapes( (BYTE *)pszMbStr,
cch,
xwcsBuffer.Get(),
_codePage );
return cch;
}
//+-------------------------------------------------------------------------
//
// Member: CCollectVar::CCollectVar, public constructor
//
// Arguments: [rUnescaper] - reference to an unescaper object - it will
// be used to unescape variables stored in
// the local buffer
// [webServer] - web server object from which variables
// are read.
//
//--------------------------------------------------------------------------
CCollectVar::CCollectVar(
CURLUnescaper& rUnescaper,
CWebServer & webServer ):
_rUnescaper(rUnescaper),
_webServer( webServer ),
_xwcsBuffer(DEFAULT_BUFF_SIZE),
_cwcVarSize(0),
_xszBuffer(DEFAULT_BUFF_SIZE),
_cbVarSize(0)
{
}
//+-------------------------------------------------------------------------
//
// Member: CCollectVar::GetEnvVar
//
// Arguments: [pwcsVariableName] - the name of the variable to be retrieved
//
// Synopsis: Returns TRUE if the specified variable was retrieved and
// FALSE otherwise.
//
//--------------------------------------------------------------------------
BOOL CCollectVar::GetEnvVar( CHAR const * pszVariableName)
{
//
// Prime to indicate that the WIDE-CHAR form of the string is
// not valid anymore.
//
_cwcVarSize = 0;
_cbVarSize = 0;
//
// Retrieve the variable as an ascii string
//
webDebugOut(( DEB_ITRACE, "WEBHITS: Getting environment variable\n" ));
ULONG cb = _xszBuffer.Count();
if ( ! _webServer.GetCGIVariable( pszVariableName,
_xszBuffer.GetPointer(),
& cb ) )
{
if ( cb > _xszBuffer.Count() )
{
//
// have to re-size and incur the cost of the memory allocation
//
delete[] _xszBuffer.Acquire();
_xszBuffer.Init( cb );
if ( ! _webServer.GetCGIVariable( pszVariableName,
_xszBuffer.GetPointer(),
& cb ) )
return FALSE;
}
else
{
return FALSE;
}
}
_cbVarSize = cb;
return TRUE;
}
//+-------------------------------------------------------------------------
//
// Member: CCollectVar::UnescapeAndConvertToUnicode
//
// Synopsis: Unescapes the variable currently stored in _xszBuffer.
//
//--------------------------------------------------------------------------
inline void CCollectVar::UnescapeAndConvertToUnicode()
{
//
// Convert the multi-byte string into a UNICODE string.
//
_cwcVarSize = _rUnescaper.UnescapeAndConvertToUnicode( _xszBuffer.Get(),
_cbVarSize,
_xwcsBuffer );
}
//+-------------------------------------------------------------------------
//
// Member: CSmartByteArray::CSmartByteArray - public constructor
//
// Synopsis: allocates memory for the buffer
//--------------------------------------------------------------------------
CSmartByteArray::CSmartByteArray():
_xszBuffer(DEFAULT_BUFFER_SIZE)
{
}
//+-------------------------------------------------------------------------
//
// Member: CSmartByteArray::CopyTo
//
// Arguments: [pwcText] - pointer to text to be copied
// [cwcToCopy] - the number of characters to be copied
//
// Synopsis: copies the text pointed to by [pwcText] to the buffer,
// enlarging it if necessary. A L'\0' is appended to the end
// automatically, since this method is used to copy non-null
// terminated strings.
//
//--------------------------------------------------------------------------
void CSmartByteArray::CopyTo(char * pszText, ULONG cbToCopy)
{
//
// if a buffer larger than the default one is being allocated, then
// re-size it
//
if ( cbToCopy >= _xszBuffer.Count() )
{
delete[] _xszBuffer.Acquire();
ULONG cbNew = max(2*_xszBuffer.Count(), cbToCopy+1);
_xszBuffer.Init(cbNew);
}
RtlCopyMemory(_xszBuffer.GetPointer(),pszText,cbToCopy);
// insert null terminator
_xszBuffer[cbToCopy] = 0;
}
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::CQueryStringParser, public constructor
//
// Arguments: [pwszQUERY_STRING] - buffer in which the *ESCAPED*
// QUERY_STRING is stored
// [rUnescaper] - ref. to unescaper object
//
// Synopsis: The QUERY_STRING must be passed in *ESCAPED* form, otherwise
// it's difficult to say for example whether an ampersand delimits
// consecutive variables or is part of the restriction
//
//--------------------------------------------------------------------------
CQueryStringParser::CQueryStringParser( char * pszQUERY_STRING,
CURLUnescaper& rUnescaper):
_pszQS(pszQUERY_STRING),
_pszQSEnd(_pszQS+strlen(_pszQS)),
_pBeg(_pszQS),
_pEnd(_pszQSEnd),
_isNull(TRUE),
_rUnescaper(rUnescaper)
{
}
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::FindVarEnd
//
// Arguments: [pwc] - pointer to character at which to begin 'searching'
// for the end of the current variable assignment
//
// Synopsis: returns a pointer to the end of the variable assignment inside
// of which pwc points - this is the next '&' if there are more
// variable assignments following, or the terminating L'\0'
// of QUERY_STRING if this is the last one.
//
//--------------------------------------------------------------------------
CHAR* CQueryStringParser::FindVarEnd(CHAR* psz)
{
if (!psz)
return NULL;
while( (psz < _pszQSEnd) && *psz != '&')
psz++;
return psz;
}
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::EatChar
//
// Arguments: [pwc] - pointer to the character that is to be eaten
//
// Synopsis: If [pwc] points to the character before the null terminator
// of the string, returns NULL. Otherwise, returns [pwc]+1.
//
//--------------------------------------------------------------------------
CHAR* CQueryStringParser::EatChar(char * psz)
{
if (!psz)
return NULL;
psz++;
if (*psz)
return psz;
else
return NULL;
}
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::EatVariableName
//
// Arguments: [pwc] - pointer to the character at which to start eating
//
// Synopsis: Eat text until a '\0', whitespace, or '=' is encountered. If
// a '\0' is encountered, return NULL, otherwise return a pointer
// to the first non-blob character, where a blob character is
// defined as anything except for whitespace, '\0', and '='.
//
//--------------------------------------------------------------------------
char * CQueryStringParser::EatVariableName(char * psz)
{
if (!psz)
return NULL;
//
// Note: this limits us to Ascii for variable names, which isn't
// really a problem since we define all the variables.
//
while ( (*psz) && !isspace(*psz) && ('=' != *psz))
psz++;
if (*psz)
return psz;
return NULL;
} //EatVariableName
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::ValidateArgument
//
// Arguments: [pwc] - pointer to the character string to validate
//
// Synopsis: Returns the pointer if valid or 0 otherwise.
//
//--------------------------------------------------------------------------
char * CQueryStringParser::ValidateArgument(char * psz)
{
if ( ( 0 == psz ) || ( 0 == *psz ) )
return 0;
return psz;
}
//+-------------------------------------------------------------------------
//
// Member: CQueryStringParser::NextVar
//
// Synopsis: Extract the next command-line variable contained
// in QUERY_STRING. Returns TRUE if the next variable was
// successfully extracted, and FALSE otherwise
//
//--------------------------------------------------------------------------
BOOL CQueryStringParser::NextVar()
{
//
// set the NULL flag
//
_isNull = TRUE;
// we've hit the end of the buffer
if (!_pBeg)
return FALSE;
// find the end of the variable assignment within the buffer
_pEnd = FindVarEnd(_pBeg);
if (!_pEnd || _pEnd == _pBeg)
{
THROW( CException(MSG_WEBHITS_QS_FORMAT_INVALID));
}
//
// copy the chunk of the QUERY_STRING into the temporary buffer for
// unescaping
//
_smartBuffer.CopyTo(_pBeg, (ULONG)(_pEnd-_pBeg)); // L'\0' appended
//_rUnescaper.Unescape(_smartBuffer.GetXPtr(), _pEnd-_pBeg);
// move to the variable name
CHAR* pTemp1 = ValidateArgument( _smartBuffer.GetPointer() );
if (!pTemp1)
{
THROW( CException(MSG_WEBHITS_QS_FORMAT_INVALID));
}
CHAR* pTemp2 = EatVariableName(pTemp1);
if (!pTemp2)
{
THROW( CException(MSG_WEBHITS_QS_FORMAT_INVALID));
}
//
// Convert the variable name into a Unicode string.
//
ULONG cbToConvert = (ULONG)(pTemp2-pTemp1) ;
_rUnescaper.UnescapeAndConvertToUnicode( pTemp1, cbToConvert, _xwszVarName );
// move to the equal sign
pTemp1=ValidateArgument(pTemp2);
if (!pTemp1 || (*pTemp1 != L'='))
{
THROW( CException(MSG_WEBHITS_QS_FORMAT_INVALID));
}
pTemp1 = EatChar(pTemp1);
//
// if not a null value keep on eating away
//
if (ValidateArgument(pTemp1))
{
// move to the variable value
pTemp1 = ValidateArgument(pTemp1);
if (!pTemp1)
{
THROW( CException(MSG_WEBHITS_QS_FORMAT_INVALID));
}
// get the variable value
cbToConvert = strlen(pTemp1);
_rUnescaper.UnescapeAndConvertToUnicode( pTemp1, cbToConvert, _xwszVarValue );
//
// if we got this far, the variable had a value
//
_isNull = FALSE;
}
else
{
//
// null value - delete the string name
//
delete _xwszVarName.Acquire();
}
// advance to the next variable assignment
if (_pEnd < _pszQSEnd)
{
_pBeg = ++_pEnd;
}
else
{
_pBeg = NULL;
}
return TRUE;
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::CGetEnvVars(), public constructor
//
// Arguments: [webServer] - web server to use
// [langInfo] - language-specific info
// [rCollectVar] - ref. to a variable-retrieving object
// [rUnescaper] - ref. to unescaper object
//
// Synopsis: The constructor retrieves the Filename, VPath, QUERY_STRING,
// and Restriction.
//--------------------------------------------------------------------------
CGetEnvVars::CGetEnvVars(
CWebServer & webServer,
CLanguageInfo & langInfo,
CCollectVar& rCollectVar,
CURLUnescaper& rUnescaper ) :
_lcid(langInfo.GetUrlCodePage()),
_hiliteType(SUMMARY),
_rCollectVar( rCollectVar ),
_rUnescaper(rUnescaper),
_isBold(FALSE),
_isItalic(FALSE),
_isFixedFont(FALSE),
_ccFixedFontLine(1),
_xwc24BitColourMask(new WCHAR[8]),
_aUserParams(eMaxUserReplParams),
_langInfo( langInfo ),
_webServer( webServer ),
_dwWebHitsFileFlags( 0 ),
_dwQueryFileFlags( 0 ),
_dwTemplateFileFlags( 0 ),
_dialect( ISQLANG_V2 )
{
for ( ULONG i = 0; i < eMaxUserReplParams; i++ )
_aUserParams[i] = 0;
wcscpy(_xwc24BitColourMask.GetPointer(),L"#FF0000");
//
// We support only the GET and POST methods -- verify
//
BYTE * pszBuffer;
ULONG cbBuffer;
if ( ( strcmp( webServer.GetMethod(), "GET" ) != 0 ) &&
( strcmp( webServer.GetMethod(), "POST" ) != 0 ) )
{
webDebugOut(( DEB_ERROR, "WEBHITS: invalid REQUEST_METHOD\n" ));
THROW( CException(MSG_WEBHITS_REQUEST_METHOD_INVALID) );
}
XArray<WCHAR> xHTW;
ULONG cwc;
if ( webServer.GetCGI_PATH_INFO( xHTW, cwc ) )
{
webDebugOut(( DEB_ITRACE, "htw file: '%ws'\n", xHTW.GetPointer() ));
// Allow null.htw files to not exist, in which case the file just
// serves as a script map trigger with default formatting info.
WCHAR *pwc = wcsrchr( xHTW.GetPointer(), L'/' );
if ( ( 0 != pwc ) && ( _wcsicmp( pwc+1, NullHTWFile ) ) )
_xwcsTemplateFileVPath.Set( xHTW.Acquire() );
}
//
// retrieve and parse the command line - we store in the same
// buffer in both places
//
RetrieveQueryString();
ParseQUERY_STRING();
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::ParseQUERY_STRING()
//
// Synopsis: parses QUERY_STRING for the various command-line parameters.
// This method must be called AFTER RetrieveQueryString. If a
// "crucial" variable remains unset at the end, an exception is
// is thrown.
//
//--------------------------------------------------------------------------
void CGetEnvVars::ParseQUERY_STRING()
{
webDebugOut(( DEB_ITRACE, "WEBHITS: parsing QS '%s'\n",
_xszQueryString.GetPointer() ));
CQueryStringParser QSParser( _xszQueryString.GetPointer(), _rUnescaper );
while(QSParser.NextVar())
{
XPtrST<WCHAR> xwcTempVarName;
XPtrST<WCHAR> xwcTempVarValue;
//
// acquire the variable name and value from the parser, and hand it over to SetVar
//
if (!QSParser.IsNull())
{
QSParser.GetVarName(xwcTempVarName);
QSParser.GetVarValue(xwcTempVarValue);
SetVar(xwcTempVarName,xwcTempVarValue);
}
}
VerifyQSVariablesComplete();
}
//+---------------------------------------------------------------------------
//
// Member: CGetEnvVars::IsUserParam
//
// Synopsis:
//
// Arguments: [pwcsParam] -
//
// Returns:
//
// Modifies:
//
// History: 9-10-96 srikants Created
//
// Notes:
//
//----------------------------------------------------------------------------
inline ULONG CGetEnvVars::IsUserParam( const WCHAR * pwcsParam )
{
if ( 0 == _wcsnicmp( pwcsParam, wcsUserParamPrefix, cwcUserParamPrefix) )
{
int i = _wtoi( pwcsParam+cwcUserParamPrefix );
if ( i > 0 && i <= eMaxUserReplParams )
return (ULONG) i;
else return 0;
}
else
{
return 0;
}
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::SetVar
//
// Arguments: [xwszVarName] - the name of the variable being set
// [xwszVarValue]- the value it is being set to
//
// Synopsis: Sets the command-line variable [xwszVarName] to [xwszVarValue]
// and throws if the variable name is invalid. The string
// containing the variable name is deleted here, as is the
// string containing the variable value if the latter is of no
// use and should not be stored.
//
//--------------------------------------------------------------------------
void CGetEnvVars::SetVar(XPtrST<WCHAR>& xwszVarName,
XPtrST<WCHAR>& xwszVarValue)
{
//
// create local copies of the smart pointers, ensuring deletion at the end
//
XPtrST<WCHAR> xwszLocalVarName(xwszVarName.Acquire());
XPtrST<WCHAR> xwszLocalVarValue(xwszVarValue.Acquire());
webDebugOut(( DEB_ITRACE, "var '%ws', value '%ws'\n",
xwszLocalVarName.GetPointer(),
xwszLocalVarValue.GetPointer() ));
if(_wcsicmp(xwszLocalVarName.GetPointer(),RestrictionVarName) == 0)
{
if ( 0 != _xwcsRestriction.GetPointer() )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
_xwcsRestriction.Set(xwszLocalVarValue.Acquire());
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),HiliteTypeVarName) == 0)
{
if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"full") == 0)
{
_hiliteType = FULL;
}
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),IDQFilenameVarName) == 0)
{
if ( 0 != _xwcsQueryFileVPath.GetPointer() )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
_xwcsQueryFileVPath.Set(xwszLocalVarValue.Acquire());
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),ColorVarName)==0)
{
if ( (*(xwszLocalVarValue.GetPointer()) == L'0') &&
(wcslen(xwszLocalVarValue.GetPointer()) == 8))
{
//
// a 24-bit colour spec is being passed in
//
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),xwszLocalVarValue.GetPointer()+1,
sizeof(Red24BitMask));
*(_xwc24BitColourMask.GetPointer())=L'#';
}
else if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"red")==0)
{
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),Red24BitMask,sizeof(Red24BitMask));
}
else if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"blue")==0)
{
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),Blue24BitMask,sizeof(Blue24BitMask));
}
else if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"green")==0)
{
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),Green24BitMask,sizeof(Green24BitMask));
}
else if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"black")==0)
{
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),Black24BitMask,sizeof(Black24BitMask));
}
else if (_wcsicmp(xwszLocalVarValue.GetPointer(),L"yellow")==0)
{
RtlCopyMemory(_xwc24BitColourMask.GetPointer(),Yellow24BitMask,sizeof(Yellow24BitMask));
}
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),VPathWebHitsFile) == 0)
{
if ( 0 != _xwcsWebHitsFileVPath.GetPointer() )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
_xwcsWebHitsFileVPath.Set(xwszLocalVarValue.Acquire());
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),BoldVarName) == 0)
{
_isBold=TRUE;
}
else if (_wcsicmp(xwszLocalVarName.GetPointer(),ItalicVarName)==0)
{
_isItalic=TRUE;
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(),MaxLineLength) == 0 )
{
_isFixedFont = TRUE;
if ( xwszLocalVarValue.GetPointer() )
_ccFixedFontLine = _wtoi( xwszLocalVarValue.GetPointer() );
_ccFixedFontLine = max( _ccFixedFontLine, 1 );
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(),LocaleVar) == 0 )
{
//
// Set the output and cirestriction locale info now.
//
if ( xwszLocalVarValue.GetPointer() )
{
if ( !_locale.IsNull() )
delete _locale.Acquire();
LCID lcid = GetLCIDFromString( xwszLocalVarValue.GetPointer() );
_locale.Init(1+wcslen(xwszLocalVarValue.GetPointer()));
wcscpy(_locale.GetPointer(), xwszLocalVarValue.GetPointer());
//
// The output will be generated using this codepage and for
// interpretation of the "CiRestriction".
//
_langInfo.SetRestrictionLocale( lcid );
}
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(), CodepageVar) == 0)
{
if (xwszLocalVarValue.GetPointer() )
{
if ( !_codepage.IsNull() )
delete _codepage.Acquire();
_codepage.Init(1+wcslen(xwszLocalVarValue.GetPointer()));
wcscpy(_codepage.GetPointer(), xwszLocalVarValue.GetPointer());
}
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(), DialectVar) == 0)
{
if ( xwszLocalVarValue.GetPointer() )
{
ULONG d = (ULONG) _wtoi( xwszLocalVarValue.GetPointer() );
if ( d > ISQLANG_V2 || d < ISQLANG_V1 )
THROW( CException( MSG_WEBHITS_INVALID_DIALECT ) );
_dialect = d;
}
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(),BeginHiliteVar) == 0 )
{
if ( 0 != _xwcsBeginHiliteTag.GetPointer() )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
#if 0 // security hole -- we can't display random html from users
_xwcsBeginHiliteTag.Set(xwszLocalVarValue.Acquire());
#endif
}
else if ( _wcsicmp(xwszLocalVarName.GetPointer(),EndHiliteVar) == 0 )
{
if ( 0 != _xwcsEndHiliteTag.GetPointer() )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
#if 0 // security hole -- we can't display random html from users
_xwcsEndHiliteTag.Set(xwszLocalVarValue.Acquire());
#endif
}
else
{
ULONG nUserParam = IsUserParam( xwszLocalVarName.GetPointer() );
if ( nUserParam > 0 )
{
if ( 0 != _aUserParams[nUserParam-1] )
THROW( CException( MSG_WEBHITS_DUPLICATE_PARAMETER ) );
_aUserParams[nUserParam-1] = xwszLocalVarValue.Acquire();
}
else
{
//
// the variable name is not recognized - throw. We don't need to
// delete the strings as they are contained in smart pointers
//
webDebugOut((DEB_ERROR,"WEBHITS: bad variable name:%ws\n",
xwszLocalVarName.GetPointer() ));
THROW(CException(MSG_WEBHITS_VARNAME_INVALID));
}
}
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::VerifyQSVariablesComplete()
//
// Synopsis: Checks whether all the "crucial" variables that were to be
// set as part of QUERY_STRING have been set, and throws an
// exception otherwise.
//
//--------------------------------------------------------------------------
void CGetEnvVars::VerifyQSVariablesComplete()
{
if(_xwcsRestriction.IsNull())
{
webDebugOut(( DEB_ERROR,
"WEBHITS: incomplete variable set read from QS\n" ));
THROW(CException(MSG_WEBHITS_INVALID_QUERY));
}
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::RetrieveCONTENT_LENGTH
//
// Synopsis: retrieves and returns the value of CONTENT_LENGTH,
// and throws if unable to retrieve
//
//--------------------------------------------------------------------------
int CGetEnvVars::RetrieveCONTENT_LENGTH()
{
if(!_rCollectVar.GetEnvVar("CONTENT_LENGTH"))
{
webDebugOut(( DEB_ERROR, "WEBHITS: failed to get CONTENT_LENGTH\n" ));
return 0;
THROW( CException(MSG_WEBHITS_CONTENT_LENGTH_INVALID) );
}
else
{
//
// get the number of bytes
//
return _wtoi(_rCollectVar.GetVarValue());
}
}
//+-------------------------------------------------------------------------
//
// Member: CGetEnvVars::RetrieveQueryString
//
// Synopsis: retrieves the value of QUERY_STRING, setting _xwcsQueryString
// to point to it, and throws otherwise
//
//--------------------------------------------------------------------------
void CGetEnvVars::RetrieveQueryString()
{
if(!_rCollectVar.GetEnvVar("QUERY_STRING"))
{
webDebugOut(( DEB_ERROR, "WEBHITS: failed to get QS\n" ));
THROW( CException(MSG_WEBHITS_INVALID_QUERY) );
}
else
{
int cbToCopy = _rCollectVar.GetMultiByteStrLen();
//
// we do not unescape QUERY_STRING to preserve delimiting information
//
_xszQueryString.Init(cbToCopy+1);
RtlCopyMemory( _xszQueryString.GetPointer(),
_rCollectVar.GetMultiByteStr(),
cbToCopy );
_xszQueryString[cbToCopy] = 0;
}
}
//+-------------------------------------------------------------------------
//
// function GetLCID
//
// Synopsis: returns the locale in HTTP_ACCEPT_LANGUAGE, or if that one is
// is not set, the one obtained from GetSystemDefaultLCID()
//
//--------------------------------------------------------------------------
LCID GetLCID( CWebServer & webServer )
{
webDebugOut(( DEB_ITRACE,
"WEBHITS: Getting HTTP_ACCEPT_LANGUAGE variable\n" ));
XArray<WCHAR> xBuffer;
ULONG cwcBuffer;
BOOL fOK = webServer.GetCGIVariableW( L"HTTP_ACCEPT_LANGUAGE",
xBuffer,
cwcBuffer );
if ( !fOK )
{
LCID locale = GetSystemDefaultLCID();
return locale;
}
LCID lcid = GetLCIDFromString( xBuffer.GetPointer() );
if ( 0xFFFFFFFF == lcid )
lcid = GetSystemDefaultLCID();
return lcid;
}
//+---------------------------------------------------------------------------
//
// Function: SetCodePageForCRunTimes
//
// Synopsis: Set the appropriate code page for the c-runtimes so that
// swprintf, putchar, etc correctly translate the unicode into
// the appropriate multi-byte sequence.
//
// Arguments: [codePage] - Code page of the client.
//
// History: 9-06-96 srikants Created
//
//----------------------------------------------------------------------------
UINT CLanguageInfo::SetCodePageForCRunTimes( UINT codePage )
{
char szCodePage[20];
sprintf( szCodePage,".%d", codePage );
char * p = setlocale( LC_ALL, szCodePage );
if ( 0 == p )
{
webDebugOut(( DEB_WARN,
"Could not set code page for %d\n. Going to system default",
codePage ));
LCID lcid = GetSystemDefaultLCID();
codePage = LocaleToCodepage( lcid );
sprintf( szCodePage,".%d", codePage );
char * p = setlocale( LC_ALL, szCodePage );
}
return codePage;
}
//+-------------------------------------------------------------------------
//
// Member: CSortQueryHits::Init()
//
// Synopsis: Initialization function for CSortQueryHits (need to do this
// as a HitIter& is needed to fully construct this class).
// creates array of positions sorted in order of occurrence
// within the document (note that a position may occur several
// times).
//
//--------------------------------------------------------------------------
void CSortQueryHits::Init()
{
Win4Assert(0 == _aPosition);
_positionCount = CountPositions();
if (0 != _positionCount)
{
_aPosition = new Position[_positionCount];
int iPosition=0;
for (BOOL fOk=_rHitIter.FirstHit();fOk;fOk=_rHitIter.NextHit())
{
int posInHit=_rHitIter.GetPositionCount();
for (int i=0;i<posInHit;i++)
_aPosition[iPosition++] = _rHitIter.GetPosition(i);
}
Win4Assert(iPosition == _positionCount);
qsort(_aPosition,_positionCount,sizeof(Position),&ComparePositions);
}
}
//+-------------------------------------------------------------------------
//
// Member: CSortQueryHits::CountPositions()
//
// Synopsis: Count the total number of positions across all hits returned
// by ISearch.
//
//--------------------------------------------------------------------------
int CSortQueryHits::CountPositions()
{
int count=0;
for (BOOL fOk = _rHitIter.FirstHit();fOk;fOk=_rHitIter.NextHit() )
{
count +=_rHitIter.GetPositionCount();
webDebugOut(( DEB_ITRACE, "Count = %d\n", count ));
}
return count;
}
//+-------------------------------------------------------------------------
//
// Member: CLinkQueryHits::CLinkQueryHits, public constructor
//
// Arguments: [rInternalQuery] - DbRestriction
// [rGetEnvVars] - object containing CGI env. variables
// [rHttpOutput] - HTTP output object
// [cmsReadTimeout] - Read timeout for IFilter on the doc
// [lockSingleThreadedFilter] - lock for single-threaded filters
// [propertyList] - Properties to webhit
// [ulDisplayScript] - Flags for displaying scripts
//
//--------------------------------------------------------------------------
CLinkQueryHits::CLinkQueryHits(
CInternalQuery & rInternalQuery,
CGetEnvVars & rGetEnvVars,
PHttpFullOutput& rHttpOutput,
DWORD cmsReadTimeout,
CReleasableLock & lockSingleThreadedFilter,
CEmptyPropertyList & propertyList,
ULONG ulDisplayScript ) :
_document( (WCHAR*) rGetEnvVars.GetWebHitsFilePPath(),
BOGUS_RANK,
rInternalQuery.GetISearchRef(),
cmsReadTimeout,
lockSingleThreadedFilter,
propertyList,
ulDisplayScript ),
_rGetEnvVars(rGetEnvVars),
_rInternalQuery(rInternalQuery),
_rHttpOutput(rHttpOutput),
_HitIter(),
_sortedHits(_HitIter),
_currentOffset(0),
_posIndex(0),
_tagCount(0),
_ccOutputBuffer(0)
{
//
// initialize the iterator
//
_HitIter.Init(&_document);
_sortedHits.Init();
//
// set the total count of positions
//
_posCount = _sortedHits.GetPositionCount();
//
// if there are any positions, initialize the "next position" data members
// to the first one
//
if (_posCount > 0)
{
Position nextPos = _sortedHits.GetPosition(0);
_nextBegOffset = nextPos.GetBegOffset();
_nextEndOffset = nextPos.GetEndOffset();
}
}
//+-------------------------------------------------------------------------
//
// Member: CLinkQueryHits::IsSeparatedBySpaces
//
// Arguments: [startOffset] - beginning offset
// [endOffset] - ending offset
//
// Synopsis: returns TRUE if the positions in the current paragraph
// determined by [startOffset] and [endOffset] are separated
// by whitespace characters.
//
// Note: There is a maximum 'allowed' separation, beyond which
// whitespace is considered significant.
//
//--------------------------------------------------------------------------
unsigned const ccSignificantWhitespace = 20;
BOOL CLinkQueryHits::IsSeparatedBySpaces(int startOffset, int endOffset)
{
Win4Assert( startOffset <= endOffset );
if (startOffset > endOffset)
return FALSE;
//
// Small buffer for calls to GetStringTypeW
//
WORD awCharType[ccSignificantWhitespace];
int len = endOffset - startOffset;
if ( 0 == len )
return TRUE;
if ( len > ccSignificantWhitespace )
return FALSE;
const WCHAR* pStart = _document.GetPointerToOffset(startOffset);
//
// Check for whitespace
//
if ( !GetStringTypeW( CT_CTYPE1, pStart, len, awCharType ) )
{
webDebugOut(( DEB_ERROR, "GetStringType returned %d\n", GetLastError() ));
return FALSE;
}
//
// Only blanks are legal 'spaces'
//
for ( int i = 0; i < len; i++ )
{
if ( 0 == (awCharType[i] & C1_BLANK) )
return FALSE;
}
return TRUE;
}
//+-------------------------------------------------------------------------
//
// Member: CLinkQueryHits::InsertLinks
//
// Synopsis: Hit highlight the document by inserting linked HTML tags
//
//--------------------------------------------------------------------------
void CLinkQueryHits::InsertLinks()
{
if ( !_rHttpOutput.IsTemplateFilePresent() )
_rHttpOutput.TagPosition(-1);
_rHttpOutput.OutputFullHeader();
//
// Determine if this is a mainly text document.
//
if ( _rGetEnvVars.IsFixedFont() )
_rHttpOutput.OutputPreformattedTag();
ULONG eofOffset = _document.GetEOFOffset();
if (_sortedHits.GetPositionCount() == 0)
{
//
// There are no hits. Just render the whole text.
//
WCHAR* pText = _document.GetWritablePointerToOffset(0);
_rHttpOutput.OutputHttpText(pText,eofOffset);
}
else
{
BOOL openTag = FALSE;
while(_currentOffset < (int) eofOffset)
{
WCHAR* pText = _document.GetWritablePointerToOffset(_currentOffset);
const ULONG cSectionSize = _nextBegOffset - _currentOffset;
if ( cSectionSize > 0 )
{
_rHttpOutput.OutputHttpText(pText,cSectionSize);
_currentOffset +=cSectionSize;
if ( (int) eofOffset == _currentOffset )
break;
pText+=cSectionSize;
}
const ULONG cHiliteSize = _nextEndOffset - _nextBegOffset;
if ( cHiliteSize > 0 )
{
//
// display the "<<" tag - if a tag is not already open
//
if (!openTag)
{
_rHttpOutput.OutputLeftTag(_tagCount-1);
_rHttpOutput.TagPosition(_tagCount);
openTag = TRUE;
}
//
// display the highlited position text
//
_rHttpOutput.OutputHilite(pText, cHiliteSize);
_currentOffset += cHiliteSize;
Win4Assert(_currentOffset == _nextEndOffset);
}
//
// get the next distinct position
//
BOOL existsNextPosition = MoveToNextDifferentPosition();
//
// display the ">>" tag unless separated only by spaces or
// last tag in doc in which case we have to make it point to the
// top
if ( !existsNextPosition )
{
if ( openTag )
{
_rHttpOutput.OutputRightTag(-1);
openTag = FALSE;
}
_nextBegOffset = _nextEndOffset = eofOffset;
}
else if ( openTag && !IsSeparatedBySpaces(_currentOffset,_nextBegOffset) )
{
_rHttpOutput.OutputRightTag(_tagCount+1);
_tagCount++;
openTag = FALSE;
}
}
}
}
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::CExtractedHit, public constructor
//
// Arguments: [rDocument] - ref. to document being hilited
// [rHit] - ref. to hit being hilited
// [rOutput] - ref. to Http Output object
// [cwcMargin] - number of chars. to be printed before and
// after the hit
// [cwcSeparation]- maximum number of characters that may
// separate consecutive positions before
// truncation occurs
// [cwcDelim] - the number of characters to print before and
// after a position in the case of truncation
//--------------------------------------------------------------------------
CExtractedHit::CExtractedHit( CDocument& rDocument,
Hit& rHit,
PHttpOutput& rOutput,
int cwcMargin,
int cwcSeparation,
int cwcDelim ):
_rDocument(rDocument),
_rHit(rHit),
_cwcMargin(cwcMargin),
_rOutput(rOutput),
_cwcSeparation(cwcSeparation),
_cwcDelim(cwcDelim),
_cwcOutputBuffer(0)
{
SortHit();
}
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::SortHit()
//
// Synopsis: Sort the positions in the hit in the order in which they occur
// in the document
//
//--------------------------------------------------------------------------
void CExtractedHit::SortHit()
{
qsort( _rHit._aPos,
_rHit.GetPositionCount(),
sizeof(Position),
&ComparePositions );
}
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::DisplayPosition
//
// Arguments: [rPos] - ref. to position being displayed
//
// Synopsis: Display the highlighted position (i.e. JUST the position)
//
//--------------------------------------------------------------------------
void CExtractedHit::DisplayPosition(const Position& rPos)
{
const WCHAR* pText = _rDocument.GetPointerToOffset(rPos.GetBegOffset());
_rOutput.OutputHilite(pText,rPos.GetLength());
}
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::ExtractHit()
//
// Synopsis: Extract the hit - i.e. display all of the positions and
// the associated preamble/postamble text
//
//--------------------------------------------------------------------------
void CExtractedHit::ExtractHit()
{
//
// make sure that we are not dealing with a null hit - i.e. a hit composed
// entirely of null positions
if (_rHit.IsNullHit())
return;
// stores the number of positions in the hit
int cPositions = _rHit.GetPositionCount();
//
// introduce new paragraph
//
_rOutput.OutputParaTag();
_rOutput.OutputEllipsis();
//
// go through displaying each position
//
//
// find the first non-null position in the hit
//
int firstRealPos = EatNullPositions();
//
// display the preamble
//
PrintPreamble(_rHit.GetPos(firstRealPos),_cwcMargin);
//
// display the positions and the stuff in between the positions
//
for (int i=firstRealPos;i < cPositions; i++)
{
DisplayPosition(_rHit.GetPos(i));
// the stuff between consecutive positions
if (i != cPositions - 1)
{
//
// guard against the case where multiple identical positions are
// returned as part of the same hit
//
if (_rHit.GetPos(i).GetBegOffset() <
_rHit.GetPos(i+1).GetBegOffset())
{
PrintBtwPositions(_rHit.GetPos(i),_rHit.GetPos(i+1));
}
else
{
while ( ( i != ( cPositions - 1 ) ) &&
( _rHit.GetPos(i).GetBegOffset() ==
_rHit.GetPos(i+1).GetBegOffset() ) )
i++;
}
}
}
//
// display the postamble
//
PrintPostamble(_rHit.GetPos(cPositions-1),_cwcMargin);
_rOutput.OutputEllipsis();
_rOutput.OutputHRULE();
} //ExtractHit
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::ComputeDistance
//
// Arguments: [rStartPos] - ref. to starting position
// [rEndPos] - ref. to end position
//
// Synopsis: Compute the distance in characters between the start and
// end positions
//
//--------------------------------------------------------------------------
ULONG CExtractedHit::ComputeDistance (const Position& rStartPos,
const Position& rEndPos)
{
return rEndPos.GetBegOffset() - rStartPos.GetEndOffset();
}
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::PrintPreamble
//
// Arguments: [rStartPosition] - ref. to start position
// [cwcDist] - maximum number of characters to print
//
//
// Synopsis: Prints the context text preceding the hit - the number of
// characters printed depends on cwcDist. The break is made
// at a word boundary
//
//--------------------------------------------------------------------------
void CExtractedHit::PrintPreamble(const Position& rStartPosition,int cwcDist)
{
WCHAR* pBeg = _rDocument.GetWritablePointerToOffset(
rStartPosition.GetBegOffset()-cwcDist);
const WCHAR* pEnd = _rDocument.GetPointerToOffset(
rStartPosition.GetBegOffset());
// Boundary case - if the beginning of the document, don't skip any thing
if ( pBeg != _rDocument.GetPointerToOffset(0) )
{
while( (pBeg < pEnd) &&
!iswspace(*pBeg) &&
!(UNICODE_PARAGRAPH_SEPARATOR == *pBeg))
pBeg++;
}
Win4Assert(pBeg <= pEnd);
_rOutput.OutputHttpText(pBeg, CiPtrToUlong( pEnd-pBeg ));
} //PrintPreamble
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::PrintPostamble
//
// Arguments: [rStartPosition] - ref. to start position
// [cwcDist] - maximum number of characters to be printed
//
// Synopsis: Print the context text following the hit - the number of
// characters printed depends on cwcDist. The break is made
// at a word boundary
//
//--------------------------------------------------------------------------
void CExtractedHit::PrintPostamble(const Position& rStartPosition, int cwcDist)
{
WCHAR* pBeg = _rDocument.GetWritablePointerToOffset(
rStartPosition.GetEndOffset());
const WCHAR* pEnd = _rDocument.GetPointerToOffset(
rStartPosition.GetEndOffset()+cwcDist);
while( (pEnd > pBeg) &&
!iswspace(*pEnd) &&
!(UNICODE_PARAGRAPH_SEPARATOR == *pEnd) )
pEnd--;
Win4Assert(pEnd >= pBeg);
_rOutput.OutputHttpText( pBeg, CiPtrToUlong( pEnd-pBeg ));
} //PrintPostamble
//+-------------------------------------------------------------------------
//
// Member: CExtractedHit::PrintBtwPositions
//
// Arguments: [rStartPosition] - ref. to start position
// [rEndPosition] - ref. to end position
//
// Synopsis: Display the text between the two positions, breaking at word
// boundaries. If the positions are separated by a distance
// greated than _cwcSeparation, then the text in-between is
// truncated, and up to _cwcDelim characters after the first and
// before the second position are printed, separated by an
// ellipsis
//
//--------------------------------------------------------------------------
void CExtractedHit::PrintBtwPositions( const Position& rStartPosition,
const Position& rEndPosition)
{
long dist;
if ((dist=ComputeDistance(rStartPosition,rEndPosition)) > _cwcSeparation)
{
PrintPostamble(rStartPosition,_cwcDelim);
_rOutput.OutputEllipsis();
PrintPreamble(rEndPosition,_cwcDelim);
}
else
{
WCHAR* pText = _rDocument.GetWritablePointerToOffset(
rStartPosition.GetEndOffset() );
_rOutput.OutputHttpText(pText,dist);
}
} //PrintBtwPositions
//+-------------------------------------------------------------------------
//
// Member: CExtractedHits::ExtractHits
//
// Arguments: same as CExtractHit constructor, except rHitIter - a hit
// iterator replaces the hit reference
//
// Synopsis: Class functor to encapsulate the action of highliting the
// hits in rHitIter. A temporary CExtractHit object is created
// for each hit.
//
//--------------------------------------------------------------------------
CExtractHits::CExtractHits(CDocument& rDocument,
HitIter& rHitIter,
PHttpOutput& rOutput,
int cwcMargin,
int cwcDelim,
int cwcSeparation )
{
for (BOOL fOK=rHitIter.FirstHit(); fOK; fOK = rHitIter.NextHit())
{
CExtractedHit extractedHit( rDocument,
rHitIter.GetHit(),
rOutput,
cwcMargin,
cwcDelim,
cwcSeparation );
extractedHit.ExtractHit();
}
} //CExtractHits
//+-------------------------------------------------------------------------
//
// Member: PHttpOutput::PHttpOutput - public constructor
//
// Arguments: [webServer] -- The web server to write to
// [langInfo] -- Language information
//
//--------------------------------------------------------------------------
PHttpOutput::PHttpOutput(
CWebServer & webServer,
CLanguageInfo & langInfo ):
_hasPrintedHeader(FALSE),
_xwc24BitColourMask(new WCHAR[8]),
_isBold(FALSE),
_isItalic(FALSE),
_isInPreformat(FALSE),
_newLine(FALSE),
_cwcOutputBuffer(0),
_pGetEnvVars(NULL),
_pTemplate(NULL),
_fUseHiliteTags(FALSE),
_cwcBeginHiliteTag(0),
_cwcEndHiliteTag(0),
_webServer( webServer ),
_langInfo( langInfo )
{
wcscpy(_xwc24BitColourMask.GetPointer(),L"#FF0000");
_mbStr.Init( MAX_OUTPUT_BUF );
}
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::Init
//
// Synopsis: Initializes the output generation class with the environment
// variables object and the template object.
//
// Arguments: [pGetEnvVars] - Pointer to the object that has the relevant
// environment variables.
// [pTemplate] - [OPTIONAL] - Pointer to the template object.
// If non-zero, this will be used to generate the output for
// hit-highlighter.
//
// History: 9-09-96 srikants Created
//
//----------------------------------------------------------------------------
void PHttpOutput::Init(CGetEnvVars* pGetEnvVars, CWebhitsTemplate * pTemplate )
{
Win4Assert( 0 != pGetEnvVars );
delete[] _xwc24BitColourMask.Acquire();
_pGetEnvVars = pGetEnvVars;
_xwc24BitColourMask.Set(_pGetEnvVars->GetColour().Acquire());
_isItalic = _pGetEnvVars->GetItalic();
_isBold = _pGetEnvVars->GetBold();
_ccCurrLine = 0;
_ccMaxLine = _pGetEnvVars->GetFixedFontLineLen();
_pTemplate = pTemplate;
_fUseHiliteTags = _pGetEnvVars->GetBeginHiliteTag() &&
_pGetEnvVars->GetEndHiliteTag();
if ( _fUseHiliteTags )
{
_cwcBeginHiliteTag = wcslen( _pGetEnvVars->GetBeginHiliteTag() );
_cwcEndHiliteTag = wcslen( _pGetEnvVars->GetEndHiliteTag() );
}
}
#define WCHAR_COUNT(x) ( (sizeof(x)/sizeof(WCHAR))-1 )
//+-------------------------------------------------------------------------
//
// Member: PHttpOutput::OutputHtmlHeader()
//
// Synopsis: Output the HTML header
//
//--------------------------------------------------------------------------
void PHttpOutput::OutputHTMLHeader()
{
if ( 0 == _pTemplate || !_pTemplate->DoesHeaderExist() )
{
WCHAR* pwszTemp;
static const WCHAR wszHdr1[] = L"HTTP/1.0 200 OK\r\nContent-Type: text/html\r\n\r\n<HTML>\n<HEAD>\n";
static const ULONG ccHdr1 = WCHAR_COUNT(wszHdr1);
OutputHttp( wszHdr1, ccHdr1 );
static const WCHAR wszHdr2[] = L"<TITLE>Query Results</TITLE>\n</HEAD>\n";
static const ULONG ccHdr2 = WCHAR_COUNT(wszHdr2);
OutputHttp( wszHdr2, ccHdr2 );
static const WCHAR wszHdr3[] = L"<H2>\"";
static const ULONG ccHdr3 = WCHAR_COUNT(wszHdr3);
OutputHttp( wszHdr3, ccHdr3 );
pwszTemp = (WCHAR*) _pGetEnvVars->GetRestriction();
OutputHttpText(pwszTemp, wcslen(pwszTemp));
static const WCHAR wszHdr4[] = L"\" in </H2>\n";
static const ULONG ccHdr4 = WCHAR_COUNT( wszHdr4 );
OutputHttp( wszHdr4, ccHdr4 );
static const WCHAR wszHdr5[] = L"<H2><a href=\"";
static const ULONG ccHdr5 = WCHAR_COUNT(wszHdr5);
OutputHttp( wszHdr5, ccHdr5 );
pwszTemp = (WCHAR*) _pGetEnvVars->GetWebHitsFileVPath();
OutputHttp(pwszTemp,wcslen(pwszTemp));
static const WCHAR wszHdr6[] = L"\">";
static const ULONG ccHdr6 = WCHAR_COUNT(wszHdr6);
OutputHttp( wszHdr6, ccHdr6 );
OutputHttpText(pwszTemp,wcslen(pwszTemp));
static const WCHAR wszHdr7[] = L"</a> </H2><P><HR>\n<BODY>";
static const ULONG ccHdr7 = WCHAR_COUNT(wszHdr7);
OutputHttp( wszHdr7, ccHdr7 );
}
else
{
static const WCHAR wszHdr8[] = L"HTTP/1.0 200 OK\r\nContent-Type: text/html\r\n\r\n";
static const ULONG ccHdr8 = WCHAR_COUNT(wszHdr8);
OutputHttp( wszHdr8, ccHdr8 );
CVirtualString str;
_pTemplate->GetWTXFile().GetHeader( str, _pTemplate->GetVariableSet() );
OutputHttp( str.Get(), str.StrLen() );
}
_hasPrintedHeader = TRUE;
} //OutputHTMLHeader
//+-------------------------------------------------------------------------
//
// Member: PHttpOutput::OutputHtmlFooter
//
// Synopsis: Output HTML footer
//
//--------------------------------------------------------------------------
void PHttpOutput::OutputHTMLFooter()
{
if ( _isInPreformat )
{
static const WCHAR wszTag1[] = L"</pre>";
static const ULONG ccTag1 = WCHAR_COUNT( wszTag1 );
OutputHttp( wszTag1, ccTag1 );
}
if ( 0 == _pTemplate || !_pTemplate->DoesFooterExist() )
{
static const WCHAR wszTag2[] = L"</BODY>\n </HTML>";
static const ULONG ccTag2 = WCHAR_COUNT( wszTag2 );
OutputHttp( wszTag2, ccTag2 );
}
else
{
CVirtualString str;
_pTemplate->GetWTXFile().GetFooter( str,
_pTemplate->GetVariableSet() );
OutputHttp( str.Get(), str.StrLen() );
}
} //OutputHTMLFooter
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputHilite()
//
// Arguments: [pwszBuffer] - pointer to the buffer to be output highlited
// [cwcBuffLength] - number of characters to print from the buff.
//
// Synopsis: Output the buffer in highlited form
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputHilite(const WCHAR* pwszBuffer, ULONG cwcBuffLength)
{
if ( !_fUseHiliteTags )
{
WCHAR wcsColourCodeEnd[] = L"</font>";
WCHAR wcsColourTag[50];
swprintf(wcsColourTag,L"<font color=\"%s\">",_xwc24BitColourMask.GetPointer());
static const WCHAR wcsBoldBegin[]=L"<B>";
static const cwcBoldBegin = wcslen( wcsBoldBegin );
static const WCHAR wcsItalicBegin[]=L"<em>";
static const cwcItalicBegin = wcslen( wcsItalicBegin );
static const WCHAR wcsBoldEnd[]=L"</B>";
static const cwcBoldEnd = wcslen(wcsBoldEnd);
static const WCHAR wcsItalicEnd[]=L"</em>";
static const cwcItalicEnd=wcslen(wcsItalicEnd);
if (_isBold)
{
OutputHttp( wcsBoldBegin,cwcBoldBegin);
}
if (_isItalic)
{
OutputHttp( wcsItalicBegin, cwcItalicBegin);
}
OutputHttp(wcsColourTag,wcslen(wcsColourTag));
OutputHttp(pwszBuffer,cwcBuffLength);
OutputHttp(wcsColourCodeEnd,wcslen(wcsColourCodeEnd));
if (_isItalic)
{
OutputHttp( wcsItalicEnd, cwcItalicEnd );
}
if (_isBold)
{
OutputHttp( wcsBoldEnd, cwcBoldEnd );
}
}
else
{
OutputHttp( _pGetEnvVars->GetBeginHiliteTag(), _cwcBeginHiliteTag );
OutputHttp( pwszBuffer,cwcBuffLength );
OutputHttp( _pGetEnvVars->GetEndHiliteTag(), _cwcEndHiliteTag );
}
} //OutputHilite
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputLeftTag()
//
// Arguments: [tagParam] - integer to be used in setting destination tag
//
// Synopsis: Output the "<<" tag, making it refer to "Tag[tagParam]"
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputLeftTag(int tagParam)
{
_cwcOutputBuffer = swprintf(_wcOutputBuffer,L"%s\"%s%d\"%s",
L"<a href=",L"#CiTag",tagParam,L">&lt;&lt;</a>");
OutputHttp(_wcOutputBuffer,_cwcOutputBuffer);
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputRightTag()
//
// Arguments: [tagParam] - integer to be used in setting destination tag
//
// Synopsis: Output the ">>" tag, making it refer to "Tag[tagParam]"
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputRightTag(int tagParam)
{
_cwcOutputBuffer = swprintf(_wcOutputBuffer, L"%s\"%s%d\"%s",L"<a href=",
L"#CiTag", tagParam, L">&gt;&gt;</a>&nbsp;");
OutputHttp(_wcOutputBuffer,_cwcOutputBuffer);
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputEllipsis()
//
// Synopsis: Output the ellipsis that separates the truncated text between
// two consecutive positions
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputEllipsis()
{
OutputHttp(L" ... ",5);
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputFullHeader()
//
// Synopsis: Output the additional header information specific to the
// full hit-highliting
//
//+-----------------------------------------------------------------------
void PHttpFullOutput::OutputFullHeader()
{
if ( 0 == _pTemplate || !_pTemplate->DoesHeaderExist() )
{
_cwcOutputBuffer = swprintf(_wcOutputBuffer,
L"%s\"%s\"%s %s %s\"%s\"%s",
L"<h3><b> <font color=",
L"#FF0000",
L">",
L"&lt;&lt; </font> takes you to the previous hit. ",
L"<font color=",L"#FF0000",
L"> &gt;&gt; </font> takes you to the next hit.</b></h3><P>\n");
OutputHttp(_wcOutputBuffer,_cwcOutputBuffer);
_cwcOutputBuffer = swprintf(_wcOutputBuffer, L"%s\"%s\"%s %s",
L"<b>Click <a href=",L"#CiTag0",
L"> &gt;&gt; </a> to go to the first hit</b>\n", L"<HR>\n" );
OutputHttp(_wcOutputBuffer, _cwcOutputBuffer);
}
} //OutputFullHeader
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::WriteToStdout
//
// Synopsis:
//
// Arguments: [pwcsBuffer] - the buffer
// [cLength] - count of wide characters
//
// History: 11-15-96 srikants Created
//
//----------------------------------------------------------------------------
void PHttpOutput::WriteToStdout( WCHAR const * pwcsBuffer, ULONG cLength )
{
if ( 0 != cLength )
_vsResult.StrCat( pwcsBuffer, cLength );
} //WriteToStdout
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::Flush
//
// Synopsis: Flushes the buffer to the web server
//
// History: 10-10-97 dlee Created
//
//----------------------------------------------------------------------------
void PHttpOutput::Flush()
{
if ( 0 != _vsResult.StrLen() )
{
DWORD cbToWrite = WideCharToXArrayMultiByte(
_vsResult.Get(),
_vsResult.StrLen(),
_langInfo.GetOutputCodePage(),
_mbStr );
if ( 0 != cbToWrite )
_webServer.RawWriteClient( _mbStr.Get(), cbToWrite );
_vsResult.Empty();
}
} //Flush
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::OutputPreformattedTag
//
// Synopsis: Outputs the pre-formatted tag if not already emitted.
//
//----------------------------------------------------------------------------
void PHttpOutput::OutputPreformattedTag()
{
if ( !_isInPreformat )
{
static WCHAR wszTag[] = L"<pre>";
static const len = ( sizeof(wszTag)/sizeof(WCHAR) ) - 1;
WriteToStdout( wszTag, len );
_isInPreformat = TRUE;
}
}
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::OutputPreFormatRawText
//
// Synopsis: Outputs the text for pre-formatted type.
//
// Arguments: [pwcsBuffer] -
// [cLength] -
//
// History: 11-15-96 srikants Created
//
//----------------------------------------------------------------------------
void PHttpOutput::OutputPreFormatRawText( const WCHAR * pwcsBuffer,
ULONG cLength )
{
Win4Assert( _isInPreformat );
WCHAR const * pwcsCurrLineBegin = pwcsBuffer;
for ( ULONG i=0, cwcToWrite = 0;
i< cLength;
i++ )
{
if ( IsNewLine( pwcsBuffer[i] ) )
{
//
// Empty out the current line - as much as has been
// accumulated.
//
WriteToStdout( pwcsCurrLineBegin, cwcToWrite );
WriteNewline();
if ( pwcsBuffer[i] == '\r' && pwcsBuffer[i+1] == '\n' )
i++;
_ccCurrLine = cwcToWrite = 0;
pwcsCurrLineBegin = pwcsBuffer+i+1; // Position at the next character
}
else
{
//
// Include this character in the current line.
//
cwcToWrite++;
//
// _ccCurrLine is cumulative from multiple invocations.
//
_ccCurrLine++;
if ( _ccCurrLine > _ccMaxLine && iswspace( pwcsBuffer[i] ) )
{
//
// Write out the line and force a line feed.
//
WriteToStdout( pwcsCurrLineBegin, cwcToWrite );
WriteNewline();
pwcsCurrLineBegin += cwcToWrite;
Win4Assert( pwcsCurrLineBegin == pwcsBuffer+i+1 );
_ccCurrLine = cwcToWrite = 0;
}
}
}
WriteToStdout( pwcsCurrLineBegin, cwcToWrite );
} //OutputPreFormatRawText
void PHttpOutput::WriteNewline()
{
static WCHAR wcNewLine[] = L"\r\n";
static const len = (sizeof(wcNewLine)/sizeof(WCHAR))-1;
WriteToStdout( wcNewLine, len );
}
void PHttpOutput::WriteBreakTag()
{
static WCHAR wcBreakTag[] = L"<BR>";
static const len = (sizeof(wcBreakTag)/sizeof(WCHAR)) - 1;
WriteToStdout( wcBreakTag, len );
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputHttp
//
// Arguments: [pwcsBuffer] - pointer to buffer containing text
// [cLength] - the number of characters to print
// [fRawText] - if TRUE, convert cr, lf and paragraph
// mark to <BR> or newline
//
// Synopsis: Encapsulates the output operation - for now, writes to stdout
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputHttp( const WCHAR* pwcsBuffer, ULONG cLength,
BOOL fRawText )
{
if ( !fRawText )
{
//
// This is HTML header or footer or formatting. No need to
// introduce <BR> tags for new-lines and paragraphs.
//
WriteToStdout( pwcsBuffer, cLength );
return;
}
else if ( _isInPreformat )
{
//
// We can emit cr-lf as cr-lf. No need to convert to <BR> tags.
// However, we have to respect the max-line length.
//
OutputPreFormatRawText( pwcsBuffer, cLength );
return;
}
//
// We must output raw text that is not pre-formatted. We have to
// convert the newlines, paragraph separators to BreakTags.
//
WCHAR const * pwcsCurrLineBegin = pwcsBuffer;
for ( ULONG i=0, cwcToWrite = 0;
i< cLength;
i++ )
{
if ( IsNewLine( pwcsBuffer[i] ) )
{
WriteToStdout( pwcsCurrLineBegin, cwcToWrite );
WriteBreakTag();
if ( pwcsBuffer[i] == '\r' && pwcsBuffer[i+1] == '\n' )
i++;
cwcToWrite = 0;
pwcsCurrLineBegin = pwcsBuffer+i+1;
}
else
{
cwcToWrite++;
}
}
WriteToStdout( pwcsCurrLineBegin, cwcToWrite );
} //OutputHttp
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputErrorHeader()
//
// Synopsis: Output an "ERROR" header
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputErrorHeader()
{
static const WCHAR wszHdr[] = L"HTTP/1.0 200 OK \r\nContent-Type: text/html\r\n\r\n<HTML>\n<BODY>\n";
static const ULONG ccHdr = WCHAR_COUNT( wszHdr );
OutputHttp( wszHdr, ccHdr );
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::OutputErrorMessage
//
// Synopsis: Send error message to stdout / web page
//
// Arguments: [pwcsBuffer] -- Error message
// [ccBuffer] -- Size in characters of [pwcsBuffer]
//
// History: 31-Jul-97 KyleP Added header
//
//+-----------------------------------------------------------------------
void PHttpOutput::OutputErrorMessage( WCHAR * pwcsBuffer, ULONG ccBuffer )
{
static const WCHAR wszTag1[] = L"<p><h3><center>";
static const ULONG ccTag1 = WCHAR_COUNT( wszTag1 );
OutputHttp( wszTag1, ccTag1 );
OutputHttpText( pwcsBuffer, ccBuffer );
static const WCHAR wszTag2[] = L"</center></h3><BR>";
static const ULONG ccTag2 = WCHAR_COUNT( wszTag2 );
OutputHttp( wszTag2, ccTag2 );
}
//+-----------------------------------------------------------------------
//
// Member: PHttpOutput::TagPosition
//
// Arguments: [tagParam] - output a <NAME="Tag[tagParam]"> tag
//
// Synopsis: Tag the current position
//
//+-----------------------------------------------------------------------
void PHttpOutput::TagPosition(int tagParam)
{
_cwcOutputBuffer= swprintf(_wcOutputBuffer,L"%s\"%s%d\"%s",
L"<a NAME=",L"CiTag",tagParam,L"> </a>");
OutputHttp(_wcOutputBuffer,_cwcOutputBuffer);
}
//+---------------------------------------------------------------------------
//
// Member: PHttpOutput::OutputHttpText
//
// Synopsis: Outputs the given data as "text" and not as html formatting.
//
// Arguments: [pwcsBuffer] - buffer to be output
// [cchLength] - length of pwcsBuffer
//
// Notes: Any UNICODE_PARAGRAPH_SEPARATOR characters in the buffer
// need to be preserved for use by OutputHttp
//
// History: 9-09-96 srikants Created
//
//----------------------------------------------------------------------------
void PHttpOutput::OutputHttpText( WCHAR * pwcsBuffer, ULONG cchLength )
{
if ( cchLength > 0 )
{
//
// NOTE - HTMLEscapeW expects the string to be NULL terminated. It
// does not accept a length field. HTMLEscapeW is one of the most
// frequently executed routines in idq.dll and so changing it to
// optionally take in a length parameter may affect performance.
// That is why I have chosen to overwrite the current buffer with
// a NULL at cchLength and restore it after the escaping - srikants.
//
const WCHAR wcTemp = pwcsBuffer[cchLength];
pwcsBuffer[cchLength] = 0;
WCHAR * pwcsTmpBuf = pwcsBuffer;
WCHAR * pwcsParaMark = wcschr( pwcsTmpBuf, UNICODE_PARAGRAPH_SEPARATOR );
while ( 0 != pwcsParaMark )
{
_escapedStr.Empty();
*pwcsParaMark = 0;
HTMLEscapeW( pwcsTmpBuf, _escapedStr, _langInfo.GetOutputCodePage() );
_escapedStr.CharCat( UNICODE_PARAGRAPH_SEPARATOR );
OutputHttp( _escapedStr.Get(), _escapedStr.StrLen(), TRUE );
*pwcsParaMark = UNICODE_PARAGRAPH_SEPARATOR;
pwcsTmpBuf = pwcsParaMark + 1;
pwcsParaMark = wcschr( pwcsTmpBuf, UNICODE_PARAGRAPH_SEPARATOR );
}
if (*pwcsTmpBuf)
{
_escapedStr.Empty();
HTMLEscapeW( pwcsTmpBuf, _escapedStr, _langInfo.GetOutputCodePage() );
OutputHttp( _escapedStr.Get(), _escapedStr.StrLen(), TRUE );
}
pwcsBuffer[cchLength] = wcTemp;
}
} //OutputHttpText
//+---------------------------------------------------------------------------
//
// Member: CWebhitsTemplate::CWebhitsTemplate
//
// Synopsis: Constructor of the webhits template file. It takes the
// environment variables object and reads in the htx file,
// processes it. Also, adds the appropriate replaceable
// parameters.
//
// Arguments: [envVars] - Environment variables to use
// [codePage] - Codepage to use for the template
//
// History: 9-09-96 srikants Created
//
//----------------------------------------------------------------------------
CWebhitsTemplate::CWebhitsTemplate(
CGetEnvVars const & envVars,
ULONG codePage )
: _wtxFile( envVars.GetTemplateFileVPath(),
envVars.GetTemplateFilePPath(),
codePage )
{
//
// Add the replacable parameters to the variable set.
//
_variableSet.AddParam( wcsParamCiUrl, envVars.GetWebHitsFileVPath() );
_variableSet.AddParam( wcsParamRestriction, envVars.GetRestriction() );
if (envVars.GetLocale())
_variableSet.AddParam( wcsLocale, envVars.GetLocale() );
if (envVars.GetCodepage())
_variableSet.AddParam( wcsCodepage, envVars.GetCodepage() );
//
// Add user definable parameters.
//
for ( ULONG i = 0; i < CGetEnvVars::eMaxUserReplParams; i++ )
{
if ( envVars.GetUserParam(i+1) )
{
_variableSet.AddParam( awcsUserParamNames[i],
envVars.GetUserParam(i+1) );
}
}
//
// Parse the htx file.
//
_wtxFile.ParseFile( _variableSet );
} //CWebhitsTemplate