Source code of Windows XP (NT5)
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.
 
 
 
 
 
 

1435 lines
37 KiB

/**********************************************************************/
/** Microsoft Windows NT **/
/** Copyright(c) Microsoft Corp., 1994 **/
/**********************************************************************/
/*
string.cxx
This module contains a light weight string class
FILE HISTORY:
Johnl 15-Aug-1994 Created
MuraliK 27-Feb-1995 Modified to be a standalone module with buffer.
MuraliK 2-June-1995 Made into separate library
*/
#include "precomp.hxx"
//
// Normal includes only for this module to be active
//
# include <opt_time.h>
extern "C" {
# include <nt.h>
# include <ntrtl.h>
# include <nturtl.h>
# include <windows.h>
};
# include "dbgutil.h"
# include <string.hxx>
# include <auxctrs.h>
# include <tchar.h>
# include <mbstring.h>
//
// String globals
//
typedef UCHAR * ( __cdecl * PFNSTRCASE ) ( UCHAR * );
typedef INT ( __cdecl * PFNSTRNICMP ) ( const UCHAR *, const UCHAR *, size_t );
typedef INT ( __cdecl * PFNSTRICMP ) ( const UCHAR *, const UCHAR * );
typedef size_t ( __cdecl * PFNSTRLEN ) ( const UCHAR * );
typedef UCHAR * (__cdecl * PFNSTRRCHR) (const UCHAR *, UINT);
PFNSTRCASE g_pfnStrupr = _mbsupr;
PFNSTRCASE g_pfnStrlwr = _mbslwr;
PFNSTRNICMP g_pfnStrnicmp = _mbsnicmp;
PFNSTRICMP g_pfnStricmp = _mbsicmp;
PFNSTRLEN g_pfnStrlen = _mbslen;
PFNSTRRCHR g_pfnStrrchr = _mbsrchr;
BOOL g_fFavorDBCS = FALSE;
#define UTF8_HACK_KEY "System\\CurrentControlSet\\Services\\InetInfo\\Parameters"
#define UTF8_HACK_VALUE "FavorDBCS"
//
// Private Definations
//
//
// When appending data, this is the extra amount we request to avoid
// reallocations
//
#define STR_SLOP 128
//
// Converts a value between zero and fifteen to the appropriate hex digit
//
#define HEXDIGIT( nDigit ) \
(TCHAR)((nDigit) > 9 ? \
(nDigit) - 10 + 'A' \
: (nDigit) + '0')
//
// Converts a single hex digit to its decimal equivalent
//
#define TOHEX( ch ) \
((ch) > '9' ? \
(ch) >= 'a' ? \
(ch) - 'a' + 10 : \
(ch) - 'A' + 10 \
: (ch) - '0')
/*******************************************************************
NAME: STR::STR
SYNOPSIS: Construct a string object
ENTRY: Optional object initializer
NOTES: If the object is not valid (i.e. !IsValid()) then GetLastError
should be called.
The object is guaranteed to construct successfully if nothing
or NULL is passed as the initializer.
********************************************************************/
// Inlined in string.hxx
VOID
STR::AuxInit( const BYTE * pInit )
{
BOOL fRet;
if ( pInit )
{
INT cbCopy = (::strlen( (const CHAR * ) pInit ) + 1) * sizeof(CHAR);
fRet = Resize( cbCopy );
if ( fRet ) {
CopyMemory( QueryPtr(), pInit, cbCopy );
m_cchLen = (cbCopy)/sizeof(CHAR) - 1;
} else {
BUFFER::SetValid( FALSE);
}
} else {
*((CHAR *) QueryPtr()) = '\0';
m_cchLen = 0;
}
return;
} // STR::AuxInit()
/*******************************************************************
NAME: STR::AuxAppend
SYNOPSIS: Appends the string onto this one.
ENTRY: Object to append
********************************************************************/
BOOL STR::AuxAppend( const BYTE * pStr, UINT cbStr, BOOL fAddSlop )
{
DBG_ASSERT( pStr != NULL );
UINT cbThis = QueryCB();
//
// Only resize when we have to. When we do resize, we tack on
// some extra space to avoid extra reallocations.
//
// Note: QuerySize returns the requested size of the string buffer,
// *not* the strlen of the buffer
//
AcIncrement( CacStringAppend);
if ( QuerySize() < cbThis + cbStr + sizeof(CHAR) )
{
if ( !Resize( cbThis + cbStr + (fAddSlop ? STR_SLOP : sizeof(CHAR) )) )
return FALSE;
}
// copy the exact string and append a null character
memcpy( (BYTE *) QueryPtr() + cbThis,
pStr,
cbStr);
m_cchLen += cbStr/sizeof(CHAR);
*((CHAR *) QueryPtr() + m_cchLen) = '\0'; // append an explicit null char
return TRUE;
} // STR::AuxAppend()
#if 0
// STR::SetLen() is inlined now
BOOL
STR::SetLen( IN DWORD cchLen)
/*++
Truncates the length of the string stored in this buffer
to specified value.
--*/
{
if ( cchLen >= QuerySize()) {
// the buffer itself is not sufficient for this length. return error.
return ( FALSE);
}
// null terminate the string at specified location
*((CHAR *) QueryPtr() + cchLen) = '\0';
m_cchLen = cchLen;
return ( TRUE);
} // STR::SetLen()
#endif // 0
/*******************************************************************
NAME: STR::LoadString
SYNOPSIS: Loads a string resource from this module's string table
or from the system string table
ENTRY: dwResID - System error or module string ID
lpszModuleName - name of the module from which to load.
If NULL, then load the string from system table.
********************************************************************/
BOOL STR::LoadString( IN DWORD dwResID,
IN LPCTSTR lpszModuleName, // Optional
IN DWORD dwLangID // Optional
)
{
BOOL fReturn = FALSE;
INT cch;
//
// If lpszModuleName is NULL, load the string from system's string table.
//
if ( lpszModuleName == NULL) {
BYTE * pchBuff = NULL;
//
// Call the appropriate function so we don't have to do the Unicode
// conversion
//
cch = ::FormatMessageA( FORMAT_MESSAGE_ALLOCATE_BUFFER |
FORMAT_MESSAGE_IGNORE_INSERTS |
FORMAT_MESSAGE_MAX_WIDTH_MASK |
FORMAT_MESSAGE_FROM_SYSTEM,
NULL,
dwResID,
dwLangID,
(LPSTR) &pchBuff,
1024,
NULL );
if ( cch ) {
fReturn = Copy( (LPCSTR) pchBuff, cch );
}
//
// Free the buffer FormatMessage allocated
//
if ( cch )
{
::LocalFree( (VOID*) pchBuff );
}
} else {
CHAR ach[STR_MAX_RES_SIZE];
cch = ::LoadStringA( GetModuleHandle( lpszModuleName),
dwResID,
(CHAR *) ach,
sizeof(ach));
if ( cch )
{
fReturn = Copy( (LPSTR) ach, cch );
}
}
return ( fReturn);
} // STR::LoadString()
BOOL STR::LoadString( IN DWORD dwResID,
IN HMODULE hModule
)
{
DBG_ASSERT( hModule != NULL );
BOOL fReturn = FALSE;
INT cch;
CHAR ach[STR_MAX_RES_SIZE];
cch = ::LoadStringA(hModule,
dwResID,
(CHAR *) ach,
sizeof(ach));
if ( cch ) {
fReturn = Copy( (LPSTR) ach, cch );
}
return ( fReturn);
} // STR::LoadString()
BOOL
STR::FormatString(
IN DWORD dwResID,
IN LPCTSTR apszInsertParams[],
IN LPCTSTR lpszModuleName,
IN DWORD cbMaxMsg
)
{
DWORD cch;
LPSTR pchBuff;
BOOL fRet = FALSE;
cch = ::FormatMessageA( FORMAT_MESSAGE_ALLOCATE_BUFFER |
FORMAT_MESSAGE_ARGUMENT_ARRAY |
FORMAT_MESSAGE_FROM_HMODULE,
GetModuleHandle( lpszModuleName ),
dwResID,
0,
(LPSTR) &pchBuff,
cbMaxMsg * sizeof(WCHAR),
(va_list *) apszInsertParams );
if ( cch )
{
fRet = Copy( (LPCSTR) pchBuff, cch );
::LocalFree( (VOID*) pchBuff );
}
/* INTRINSA suppress = uninitialized */
return fRet;
}
/*******************************************************************
NAME: STR::Escape
SYNOPSIS: Replaces non-ASCII characters with their hex equivalent
NOTES:
HISTORY:
Johnl 17-Aug-1994 Created
********************************************************************/
BOOL STR::Escape( VOID )
{
CHAR * pch = QueryStr();
int i = 0;
CHAR ch;
DBG_ASSERT( pch );
while ( ch = pch[i] )
{
//
// Escape characters that are in the non-printable range
// but ignore CR and LF
//
if ( (((ch >= 0) && (ch <= 32)) ||
((ch >= 128) && (ch <= 159))||
(ch == '%') || (ch == '?') || (ch == '+') || (ch == '&') ||
(ch == '#')) &&
!(ch == '\n' || ch == '\r') )
{
if ( !Resize( QuerySize() + 2 * sizeof(CHAR) ))
return FALSE;
//
// Resize can change the base pointer
//
pch = QueryStr();
//
// Insert the escape character
//
pch[i] = '%';
//
// Insert a space for the two hex digits (memory can overlap)
//
/* INTRINSA suppress = uninitialized */
::memmove( &pch[i+3],
&pch[i+1],
(::strlen( &pch[i+1] ) + 1) * sizeof(CHAR));
//
// Convert the low then the high character to hex
//
UINT nDigit = (UINT)(ch % 16);
pch[i+2] = HEXDIGIT( nDigit );
ch /= 16;
nDigit = (UINT)(ch % 16);
pch[i+1] = HEXDIGIT( nDigit );
i += 3;
}
else
i++;
}
m_cchLen = ::strlen( QueryStr()); // to be safe recalc the new length
return TRUE;
} // STR::Escape()
/*******************************************************************
NAME: STR::EscapeSpaces
SYNOPSIS: Replaces all spaces with their hex equivalent
NOTES:
HISTORY:
Johnl 17-Aug-1994 Created
********************************************************************/
BOOL STR::EscapeSpaces( VOID )
{
CHAR * pch = QueryStr();
CHAR * pchTmp;
int i = 0;
DBG_ASSERT( pch );
while ( pchTmp = strchr( pch + i, ' ' ))
{
i = DIFF( pchTmp - QueryStr() );
if ( !Resize( QuerySize() + 2 * sizeof(CHAR) ))
return FALSE;
//
// Resize can change the base pointer
//
pch = QueryStr();
//
// Insert the escape character
//
pch[i] = '%';
//
// Insert a space for the two hex digits (memory can overlap)
//
::memmove( &pch[i+3],
&pch[i+1],
(::strlen( &pch[i+1] ) + 1) * sizeof(CHAR));
//
// This routine only replaces spaces
//
pch[i+1] = '2';
pch[i+2] = '0';
}
//
// If i is zero then no spaces were found
//
if ( i != 0 )
{
m_cchLen = ::strlen( QueryStr()); // to be safe recalc the new length
}
return TRUE;
} // STR::EscapeSpaces()
/*******************************************************************
NAME: STR::Unescape
SYNOPSIS: Replaces hex escapes with the Latin-1 equivalent
NOTES: This is a Unicode only method
HISTORY:
Johnl 17-Aug-1994 Created
********************************************************************/
BOOL STR::Unescape( VOID )
{
CHAR *pScan;
CHAR *pDest;
CHAR *pNextScan;
wchar_t wch;
DWORD dwLen;
BOOL fChanged = FALSE;
pDest = pScan = strchr( QueryStr(), '%');
while (pScan)
{
if ( (pScan[1] == 'u' || pScan[1] == 'U') &&
::isxdigit( (UCHAR)pScan[2] ) &&
::isxdigit( (UCHAR)pScan[3] ) &&
::isxdigit( (UCHAR)pScan[4] ) &&
::isxdigit( (UCHAR)pScan[5] ) )
{
wch = TOHEX(pScan[2]) * 4096 + TOHEX(pScan[3]) * 256;
wch += TOHEX(pScan[4]) * 16 + TOHEX(pScan[5]);
dwLen = WideCharToMultiByte( CP_ACP,
0,
&wch,
1,
(LPSTR) pDest,
2,
NULL,
NULL );
pDest += dwLen;
pScan += 6;
fChanged = TRUE;
}
else if ( ::isxdigit( (UCHAR)pScan[1] ) && // WinSE 4944
::isxdigit( (UCHAR)pScan[2] ))
{
*pDest = TOHEX(pScan[1]) * 16 + TOHEX(pScan[2]);
pDest ++;
pScan += 3;
fChanged = TRUE;
}
else // Not an escaped char, just a '%'
{
if (fChanged)
*pDest = *pScan;
pDest++;
pScan++;
}
//
// Copy all the information between this and the next escaped char
//
pNextScan = strchr( pScan, '%');
if (fChanged) // pScan!=pDest, so we have to copy the char's
{
if (!pNextScan) // That was the last '%' in the string
{
::memmove( pDest,
pScan,
(::strlen( pScan ) + 1) * sizeof(CHAR)); // +1 to copy '\0'
}
else // There is another '%', and it is not back to back with this one
if (dwLen = DIFF(pNextScan - pScan))
{
::memmove( pDest,
pScan,
dwLen * sizeof(CHAR));
pDest += dwLen;
}
}
pScan = pNextScan;
}
if ( fChanged )
{
m_cchLen = ::strlen( QueryStr()); // for safety recalc the length
}
return TRUE;
}
BOOL
STR::CopyToBuffer( WCHAR * lpszBuffer, LPDWORD lpcch) const
/*++
Description:
Copies the string into the WCHAR buffer passed in if the buffer
is sufficient to hold the translated string.
If the buffer is small, the function returns small and sets *lpcch
to contain the required number of characters.
Arguments:
lpszBuffer pointer to WCHAR buffer which on return contains
the UNICODE version of string on success.
lpcch pointer to DWORD containing the length of the buffer.
If *lpcch == 0 then the function returns TRUE with
the count of characters required stored in *lpcch.
Also in this case lpszBuffer is not affected.
Returns:
TRUE on success.
FALSE on failure. Use GetLastError() for further details.
History:
MuraliK 11-30-94
--*/
{
BOOL fReturn = TRUE;
if ( lpcch == NULL) {
SetLastError( ERROR_INVALID_PARAMETER);
return ( FALSE);
}
if ( *lpcch == 0) {
//
// Inquiring the size of buffer alone
//
*lpcch = QueryCCH() + 1; // add one character for terminating null
} else {
//
// Copy after conversion from ANSI to Unicode
//
int iRet;
iRet = MultiByteToWideChar( CP_ACP, MB_PRECOMPOSED,
QueryStrA(), QueryCCH() + 1,
lpszBuffer, (int )*lpcch);
if ( iRet == 0 || iRet != (int ) *lpcch) {
//
// Error in conversion.
//
fReturn = FALSE;
}
}
return ( fReturn);
} // STR::CopyToBuffer()
BOOL
STR::CopyToBuffer( CHAR * lpszBuffer, LPDWORD lpcch) const
/*++
Description:
Copies the string into the CHAR buffer passed in if the buffer
is sufficient to hold the translated string.
If the buffer is small, the function returns small and sets *lpcch
to contain the required number of characters.
Arguments:
lpszBuffer pointer to CHAR buffer which on return contains
the string on success.
lpcch pointer to DWORD containing the length of the buffer.
If *lpcch == 0 then the function returns TRUE with
the count of characters required stored in *lpcch.
Also in this case lpszBuffer is not affected.
Returns:
TRUE on success.
FALSE on failure. Use GetLastError() for further details.
History:
MuraliK 20-Nov-1996
--*/
{
BOOL fReturn = TRUE;
if ( lpcch == NULL) {
SetLastError( ERROR_INVALID_PARAMETER);
return ( FALSE);
}
register DWORD cch = QueryCCH() + 1;
if ( (*lpcch >= cch) && ( NULL != lpszBuffer)) {
DBG_ASSERT( lpszBuffer);
CopyMemory( lpszBuffer, QueryStrA(), cch);
} else {
DBG_ASSERT( (NULL == lpszBuffer) || (*lpcch < cch));
SetLastError( ERROR_INSUFFICIENT_BUFFER);
fReturn = FALSE;
}
*lpcch = cch;
return ( fReturn);
} // STR::CopyToBuffer()
BOOL
STR::SafeCopy( const CHAR * pchInit )
{
DWORD cchLen = 0;
char cFirstByte = '\0';
BOOL bReturn = TRUE;
if ( QueryPtr() ) {
cFirstByte = *(QueryStr());
cchLen = m_cchLen;
*(QueryStr()) = '\0';
m_cchLen = 0;
}
if (pchInit != NULL) {
bReturn = AuxAppend( (const BYTE *) pchInit, ::strlen( pchInit ), FALSE );
if (!bReturn && QueryPtr()) {
*(QueryStr()) = cFirstByte;
m_cchLen = cchLen;
}
}
return bReturn;
}
/*******************************************************************
NAME: ::CollapseWhite
SYNOPSIS: Collapses white space starting at the passed pointer.
RETURNS: Returns a pointer to the next chunk of white space or the
end of the string.
NOTES: This is a Unicode only method
HISTORY:
Johnl 24-Aug-1994 Created
********************************************************************/
WCHAR * CollapseWhite( WCHAR * pch )
{
LPWSTR pchStart = pch;
while ( ISWHITE( *pch ) )
pch++;
::memmove( pchStart,
pch,
DIFF(pch - pchStart) );
while ( *pch && !ISWHITE( *pch ))
pch++;
return pch;
} // CollapseWhite()
//
// Private constants.
//
#define ACTION_NOTHING 0x00000000
#define ACTION_EMIT_CH 0x00010000
#define ACTION_EMIT_DOT_CH 0x00020000
#define ACTION_EMIT_DOT_DOT_CH 0x00030000
#define ACTION_BACKUP 0x00040000
#define ACTION_MASK 0xFFFF0000
//
// Private globals.
//
INT p_StateTable[16] =
{
// state 0
0 , // other
0 , // "."
4 , // EOS
1 , // "\"
// state 1
0 , // other
2 , // "."
4 , // EOS
1 , // "\"
// state 2
0 , // other
3 , // "."
4 , // EOS
1 , // "\"
// state 3
0 , // other
0 , // "."
4 , // EOS
1 // "\"
};
INT p_ActionTable[16] =
{
// state 0
ACTION_EMIT_CH, // other
ACTION_EMIT_CH, // "."
ACTION_EMIT_CH, // EOS
ACTION_EMIT_CH, // "\"
// state 1
ACTION_EMIT_CH, // other
ACTION_NOTHING, // "."
ACTION_EMIT_CH, // EOS
ACTION_NOTHING, // "\"
// state 2
ACTION_EMIT_DOT_CH, // other
ACTION_NOTHING, // "."
ACTION_EMIT_CH, // EOS
ACTION_NOTHING, // "\"
// state 3
ACTION_EMIT_DOT_DOT_CH, // other
ACTION_EMIT_DOT_DOT_CH, // "."
ACTION_BACKUP, // EOS
ACTION_BACKUP // "\"
};
// since max states = 4, we calculat the index by multiplying with 4.
# define IndexFromState( st) ( (st) * 4)
// the following table provides the index for various ISA Latin1 characters
// in the incoming URL.
// It assumes that the URL is ISO Latin1 == ASCII
INT p_rgIndexForChar[] = {
2, // null char
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 1 thru 10
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 11 thru 20
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 21 thru 30
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 31 thru 40
0, 0, 0, 0, 0, 1, 3, 0, 0, 0, // 41 thru 50 46 = '.' 47 = '/'
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 51 thru 60
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 61 thru 70
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 71 thru 80
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 81 thru 90
0, 3, 0, 0, 0, 0, 0, 0, 0, 0, // 91 thru 100 92 = '\\'
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 101 thru 110
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 111 thru 120
0, 0, 0, 0, 0, 0, 0, 0 // 121 thru 128
};
#define IS_UTF8_TRAILBYTE(ch) (((ch) & 0xc0) == 0x80)
/*******************************************************************
NAME: IsUTF8URL
ENTRY: pszPath - The path to sanitize.
HISTORY:
atsusk 06-Jan-1998 Created.
********************************************************************/
BOOL IsUTF8URL(CHAR * pszPath)
{
CHAR ch;
if ( g_fFavorDBCS )
{
return ( MultiByteToWideChar( CP_ACP,
MB_ERR_INVALID_CHARS,
pszPath,
-1,
NULL,
0) == 0);
}
while (ch = *pszPath++) {
if (ch & 0x80) {
wchar_t wch;
int iLen;
BOOL bDefault = FALSE;
char chTrail1;
char chTrail2;
chTrail1 = *pszPath++;
if (chTrail1) {
chTrail2 = *pszPath;
} else {
chTrail2 = 0;
}
if ( ((ch & 0xF0) == 0xE0) &&
IS_UTF8_TRAILBYTE(chTrail1) &&
IS_UTF8_TRAILBYTE(chTrail2) ) {
// handle three byte case
// 1110xxxx 10xxxxxx 10xxxxxx
wch = (wchar_t) (((ch & 0x0f) << 12) |
((chTrail1 & 0x3f) << 6) |
(chTrail2 & 0x3f));
pszPath++;
} else
if ( ((ch & 0xE0) == 0xC0) &&
IS_UTF8_TRAILBYTE(chTrail1) ) {
// handle two byte case
// 110xxxxx 10xxxxxx
wch = (wchar_t) (((ch & 0x1f) << 6) | (chTrail1 & 0x3f));
} else
return FALSE;
iLen = WideCharToMultiByte( CP_ACP,
0,
&wch,
1,
NULL,
0,
NULL,
&bDefault );
if (bDefault == TRUE || iLen == 0 || iLen > 2)
return FALSE;
}
}
return TRUE;
} // IsUTF8URL()
/*******************************************************************
NAME: CanonURL
SYNOPSIS: Sanitizes a path by removing bogus path elements.
As expected, "/./" entries are simply removed, and
"/../" entries are removed along with the previous
path element.
To maintain compatibility with URL path semantics
additional transformations are required. All backward
slashes "\\" are converted to forward slashes. Any
repeated forward slashes (such as "///") are mapped to
single backslashes.
A state table (see the p_StateTable global at the
beginning of this file) is used to perform most of
the transformations. The table's rows are indexed
by current state, and the columns are indexed by
the current character's "class" (either slash, dot,
NULL, or other). Each entry in the table consists
of the new state tagged with an action to perform.
See the ACTION_* constants for the valid action
codes.
ENTRY: pszPath - The path to sanitize.
fIsDBCSLocale - Indicates the server is in a
locale that uses DBCS.
HISTORY:
KeithMo 07-Sep-1994 Created.
MuraliK 28-Apr-1995 Adopted this for symbolic paths
********************************************************************/
INT
CanonURL(
CHAR * pszPath,
BOOL fIsDBCSLocale
)
{
UCHAR * pszSrc;
UCHAR * pszDest;
DWORD ch;
INT index;
BOOL fDBCS = FALSE;
DWORD cchMultiByte = 0;
DBG_ASSERT( pszPath != NULL );
//
// Always look for UTF8 except when DBCS characters are detected
//
BOOL fScanForUTF8 = IsUTF8URL(pszPath);
// If fScanForUTF8 is true, this URL is UTF8. don't recognize DBCS.
if (fIsDBCSLocale && fScanForUTF8) {
fIsDBCSLocale = FALSE;
}
//
// Start our scan at the first character
//
pszSrc = pszDest = (UCHAR *) pszPath;
//
// State 0 is the initial state.
//
index = 0; // State = 0
//
// Loop until we enter state 4 (the final, accepting state).
//
do {
//
// Grab the next character from the path and compute its
// next state. While we're at it, map any forward
// slashes to backward slashes.
//
index = IndexFromState( p_StateTable[index]); // 4 = # states
ch = (DWORD ) *pszSrc++;
//
// If this is a DBCS trailing byte - skip it
//
if ( !fIsDBCSLocale )
{
index += (( ch >= 0x80) ? 0 : p_rgIndexForChar[ch]);
}
else
{
if ( fDBCS )
{
//
// If this is a 0 terminator, we need to set next
// state accordingly
//
if ( ch == 0 )
{
index += p_rgIndexForChar[ ch ];
}
//
// fDBCS == TRUE means this byte was a trail byte.
// index is implicitly set to zero.
//
fDBCS = FALSE;
}
else
{
index += (( ch >= 0x80) ? 0 : p_rgIndexForChar[ch]);
if ( IsDBCSLeadByte( (UCHAR)ch ) )
{
//
// This is a lead byte, so the next is a trail.
//
fDBCS = TRUE;
}
}
}
//
// Interesting UTF8 characters always have the top bit set
//
if ( (ch & 0x80) && fScanForUTF8 )
{
wchar_t wch;
UCHAR mbstr[2];
//
// This is a UTF8 character, convert it here.
// index is implicitly set to zero.
//
if ( cchMultiByte < 2 )
{
char chTrail1;
char chTrail2;
chTrail1 = *pszSrc;
if (chTrail1) {
chTrail2 = *(pszSrc+1);
} else {
chTrail2 = 0;
}
wch = 0;
if ((ch & 0xf0) == 0xe0)
{
// handle three byte case
// 1110xxxx 10xxxxxx 10xxxxxx
wch = (wchar_t) (((ch & 0x0f) << 12) |
((chTrail1 & 0x3f) << 6) |
(chTrail2 & 0x3f));
cchMultiByte = WideCharToMultiByte( CP_ACP,
0,
&wch,
1,
(LPSTR) mbstr,
2,
NULL,
NULL );
ch = mbstr[0];
pszSrc += (3 - cchMultiByte);
// WinSE 12843: Security Fix, Index should be updated for this character
index += (( ch >= 0x80) ? 0 : p_rgIndexForChar[ch]);
} else if ((ch & 0xe0) == 0xc0)
{
// handle two byte case
// 110xxxxx 10xxxxxx
wch = (wchar_t) (((ch & 0x1f) << 6) | (chTrail1 & 0x3f));
cchMultiByte = WideCharToMultiByte( CP_ACP,
0,
&wch,
1,
(LPSTR) mbstr,
2,
NULL,
NULL );
ch = mbstr[0];
pszSrc += (2 - cchMultiByte);
// WinSE 12843: Security Fix, Index should be updated for this character
index += (( ch >= 0x80) ? 0 : p_rgIndexForChar[ch]);
}
} else {
//
// get ready to emit 2nd byte of converted character
//
ch = mbstr[1];
cchMultiByte = 0;
}
}
//
// Perform the action associated with the state.
//
switch( p_ActionTable[index] )
{
case ACTION_EMIT_DOT_DOT_CH :
*pszDest++ = '.';
/* fall through */
case ACTION_EMIT_DOT_CH :
*pszDest++ = '.';
/* fall through */
case ACTION_EMIT_CH :
*pszDest++ = (CHAR ) ch;
/* fall through */
case ACTION_NOTHING :
break;
case ACTION_BACKUP :
if( (pszDest > ( (UCHAR *) pszPath + 1 ) ) && (*pszPath == '/'))
{
pszDest--;
DBG_ASSERT( *pszDest == '/' );
*pszDest = '\0';
pszDest = (UCHAR *) strrchr( pszPath, '/') + 1;
}
*pszDest = '\0';
break;
default :
DBG_ASSERT( !"Invalid action code in state table!" );
index = IndexFromState(0) + 2; // move to invalid state
DBG_ASSERT( p_StateTable[index] == 4);
*pszDest++ = '\0';
break;
}
} while( p_StateTable[index] != 4 );
//
// point to terminating nul
//
if (p_ActionTable[index] == ACTION_EMIT_CH) {
pszDest--;
}
DBG_ASSERT(*pszDest == '\0' && pszDest > (UCHAR*) pszPath);
return DIFF(pszDest - (UCHAR*)pszPath);
} // CanonURL()
DWORD
InitializeStringFunctions(
VOID
)
/*++
Initializes the string function pointers depending on the system code page.
If the code page doesn't have multi-byte characters, then pointers
resolve to regular single byte functions. Otherwise, they resolve to more
expense multi-byte functions.
Arguments:
None
Returns:
0 if successful, else Win32 Error
--*/
{
CPINFO CodePageInfo;
BOOL bRet;
HKEY hKey;
DWORD dwRet;
bRet = GetCPInfo( CP_ACP, &CodePageInfo );
if ( bRet && CodePageInfo.MaxCharSize == 1 )
{
g_pfnStrlwr = (PFNSTRCASE) _strlwr;
g_pfnStrupr = (PFNSTRCASE) _strupr;
g_pfnStrnicmp = (PFNSTRNICMP) _strnicmp;
g_pfnStricmp = (PFNSTRICMP) _stricmp;
g_pfnStrlen = (PFNSTRLEN) strlen;
g_pfnStrrchr = (PFNSTRRCHR) strrchr;
}
//
// Do we need to hack for Korean?
//
dwRet = RegOpenKeyEx( HKEY_LOCAL_MACHINE,
UTF8_HACK_KEY,
0,
KEY_READ,
&hKey );
if ( dwRet == ERROR_SUCCESS )
{
DWORD dwValue = 0;
DWORD cbValue = sizeof( dwValue );
dwRet = RegQueryValueEx( hKey,
UTF8_HACK_VALUE,
NULL,
NULL,
(LPBYTE) &dwValue,
&cbValue );
if ( dwRet == ERROR_SUCCESS )
{
g_fFavorDBCS = !!dwValue;
}
DBG_REQUIRE( RegCloseKey( hKey ) == ERROR_SUCCESS );
}
return ERROR_SUCCESS;
}
UCHAR *
IISstrupr(
UCHAR * pszString
)
/*++
Wrapper for strupr() call.
Arguments:
pszString - String to uppercase
Returns:
Pointer to string uppercased
--*/
{
DBG_ASSERT( g_pfnStrupr != NULL );
return g_pfnStrupr( pszString );
}
UCHAR *
IISstrlwr(
UCHAR * pszString
)
/*++
Wrapper for strlwr() call.
Arguments:
pszString - String to lowercase
Returns:
Pointer to string lowercased
--*/
{
DBG_ASSERT( g_pfnStrlwr != NULL );
return g_pfnStrlwr( pszString );
}
size_t
IISstrlen(
UCHAR * pszString
)
/*++
Wrapper for strlen() call.
Arguments:
pszString - String to check
Returns:
Length of string
--*/
{
DBG_ASSERT( g_pfnStrlen != NULL );
return g_pfnStrlen( pszString );
}
INT
IISstrnicmp(
UCHAR * pszString1,
UCHAR * pszString2,
size_t size
)
/*++
Wrapper for strnicmp() call.
Arguments:
pszString1 - String1
pszString2 - String2
size - # characters to compare upto
Returns:
0 if equal, -1 if pszString1 < pszString2, else 1
--*/
{
DBG_ASSERT( g_pfnStrnicmp != NULL );
return g_pfnStrnicmp( pszString1, pszString2, size );
}
INT
IISstricmp(
UCHAR * pszString1,
UCHAR * pszString2
)
/*++
Wrapper for stricmp() call.
Arguments:
pszString1 - String1
pszString2 - String2
Returns:
0 if equal, -1 if pszString1 < pszString2, else 1
--*/
{
DBG_ASSERT( g_pfnStricmp != NULL );
return g_pfnStricmp( pszString1, pszString2 );
}
// like strncpy, but doesn't pad the end of the string with zeroes, which
// is expensive when `source' is short and `count' is large
char *
IISstrncpy(
char * dest,
const char * source,
size_t count)
{
char *start = dest;
while (count && (*dest++ = *source++)) /* copy string */
count--;
if (count) /* append one zero */
*dest = '\0';
return(start);
}
UCHAR *
IISstrrchr(
const UCHAR * pszString,
UINT c
)
/*++
Wrapper for strrchr() call.
Arguments:
pszString - String
c - Character to find.
Returns:
pointer to the char or NULL.
--*/
{
DBG_ASSERT( g_pfnStrrchr != NULL );
return g_pfnStrrchr( pszString, c );
}