Amebis/wxWidgets
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fixed compilation in non-Unicode build; fixed bug with buffer overrun in wxMBConvUTF8::MB2WC()

Browse Source 
git-svn-id: https://svn.wxwidgets.org/svn/wx/wxWidgets/trunk@33109 c3d73ce0-8a6f-49c7-b76d-6d57e0e08775
This commit is contained in:
Vadim Zeitlin
2005-03-27 22:20:44 +00:00
parent 172541f64b
commit 3698ae7195
1 changed files with 49 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

93
src/common/strconv.cpp
View File

@@ -281,7 +281,7 @@ const wxWCharBuffer wxMBConv::cMB2WC(const char *szString, size_t nStringLen, si
//success - return actual length and the buffer //success - return actual length and the buffer
*pOutSize = nActualLength; *pOutSize = nActualLength;
return theBuffer; return theBuffer;
} }
const wxCharBuffer wxMBConv::cWC2MB(const wchar_t *szString, size_t nStringLen, size_t* pOutSize) const const wxCharBuffer wxMBConv::cWC2MB(const wchar_t *szString, size_t nStringLen, size_t* pOutSize) const
@@ -316,7 +316,7 @@ const wxCharBuffer wxMBConv::cWC2MB(const wchar_t *szString, size_t nStringLen,
//Increase the actual length (+1 for current null character) //Increase the actual length (+1 for current null character)
nActualLength += nLen + 1; nActualLength += nLen + 1;
//if buffer too big, realloc the buffer //if buffer too big, realloc the buffer
if (nActualLength > (nCurrentSize+1)) if (nActualLength > (nCurrentSize+1))
{ {
@@ -343,7 +343,7 @@ const wxCharBuffer wxMBConv::cWC2MB(const wchar_t *szString, size_t nStringLen,
//success - return actual length and the buffer //success - return actual length and the buffer
*pOutSize = nActualLength; *pOutSize = nActualLength;
return theBuffer; return theBuffer;
} }
// ---------------------------------------------------------------------------- // ----------------------------------------------------------------------------
@@ -364,7 +364,7 @@ size_t wxMBConvLibc::WC2MB(char *buf, const wchar_t *psz, size_t n) const
// wxConvBrokenFileNames is made for GTK2 in Unicode mode when // wxConvBrokenFileNames is made for GTK2 in Unicode mode when
// files are accidentally written in an encoding which is not // files are accidentally written in an encoding which is not
// the system encoding. Typically, the system encoding will be // the system encoding. Typically, the system encoding will be
// UTF8 but there might be files stored in ISO8859-1 on disk. // UTF8 but there might be files stored in ISO8859-1 on disk.
// ---------------------------------------------------------------------------- // ----------------------------------------------------------------------------
class wxConvBrokenFileNames: public wxMBConvLibc class wxConvBrokenFileNames: public wxMBConvLibc
@@ -405,7 +405,7 @@ size_t wxConvBrokenFileNames::WC2MB(char *outputBuf, const wchar_t *psz, size_t
} }
// ---------------------------------------------------------------------------- // ----------------------------------------------------------------------------
// UTF-7 // UTF-7
// ---------------------------------------------------------------------------- // ----------------------------------------------------------------------------
// Implementation (C) 2004 Fredrik Roubert // Implementation (C) 2004 Fredrik Roubert
@@ -615,6 +615,8 @@ size_t wxMBConvUTF7::WC2MB(char *buf, const wchar_t *psz, size_t n) const
static wxUint32 utf8_max[]= static wxUint32 utf8_max[]=
{ 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff, 0xffffffff }; { 0x7f, 0x7ff, 0xffff, 0x1fffff, 0x3ffffff, 0x7fffffff, 0xffffffff };
// boundaries of the private use area we use to (temporarily) remap invalid
// characters invalid in a UTF-8 encoded string
const wxUint32 wxUnicodePUA = 0x100000; const wxUint32 wxUnicodePUA = 0x100000;
const wxUint32 wxUnicodePUAEnd = wxUnicodePUA + 256; const wxUint32 wxUnicodePUAEnd = wxUnicodePUA + 256;
@@ -718,26 +720,23 @@ size_t wxMBConvUTF8::MB2WC(wchar_t *buf, const char *psz, size_t n) const
#endif #endif
} }
} }
else else if (m_options & MAP_INVALID_UTF8_TO_OCTAL)
if (m_options & MAP_INVALID_UTF8_TO_OCTAL)
{ {
while (opsz < psz && (!buf || len < n)) while (opsz < psz && (!buf || len < n))
{ {
wchar_t str[6]; if ( buf && len + 3 < n )
wxSnprintf( str, 5, L"\\%o", (int) (unsigned char) *opsz ); {
if (buf) unsigned char n = *opsz;
*buf++ = str[0]; *buf++ = L'\\';
if (buf) *buf++ = L'0' + n / 0100;
*buf++ = str[1]; *buf++ = L'0' + (n % 0100) / 010;
if (buf) *buf++ = L'0' + n % 010;
*buf++ = str[2]; }
if (buf)
*buf++ = str[3];
opsz++; opsz++;
len += 4; len += 4;
} }
} }
else else // MAP_INVALID_UTF8_NOT
{ {
return (size_t)-1; return (size_t)-1;
} }
@@ -749,6 +748,11 @@ size_t wxMBConvUTF8::MB2WC(wchar_t *buf, const char *psz, size_t n) const
return len; return len;
} }
static inline bool isoctal(wchar_t wch)
{
return L'0' <= wch && wch <= L'7';
}
size_t wxMBConvUTF8::WC2MB(char *buf, const wchar_t *psz, size_t n) const size_t wxMBConvUTF8::WC2MB(char *buf, const wchar_t *psz, size_t n) const
{ {
size_t len = 0; size_t len = 0;
@@ -763,26 +767,26 @@ size_t wxMBConvUTF8::WC2MB(char *buf, const wchar_t *psz, size_t n) const
#else #else
cc=(*psz++) & 0x7fffffff; cc=(*psz++) & 0x7fffffff;
#endif #endif
if ((m_options & MAP_INVALID_UTF8_TO_PUA)
&& cc >= wxUnicodePUA && cc < wxUnicodePUAEnd) if ( (m_options & MAP_INVALID_UTF8_TO_PUA)
&& cc >= wxUnicodePUA && cc < wxUnicodePUAEnd )
{ {
if (buf) if (buf)
*buf++ = (char)(cc - wxUnicodePUA); *buf++ = (char)(cc - wxUnicodePUA);
len++; len++;
} }
else else if ( (m_options & MAP_INVALID_UTF8_TO_OCTAL) &&
if ((m_options & MAP_INVALID_UTF8_TO_OCTAL) cc == L'\\' &&
&& cc == L'\\') isoctal(psz[0]) && isoctal(psz[1]) && isoctal(psz[2]) )
{ {
wchar_t str[4];
str[0] = *psz; psz++;
str[1] = *psz; psz++;
str[2] = *psz; psz++;
str[3] = 0;
int octal;
wxSscanf( str, L"%o", &octal );
if (buf) if (buf)
*buf++ = (char) octal; {
*buf++ = (char) (psz[0] - L'0')*0100 +
(psz[1] - L'0')*010 +
(psz[2] - L'0');
}
psz += 3;
len++; len++;
} }
else else
@@ -810,7 +814,8 @@ size_t wxMBConvUTF8::WC2MB(char *buf, const wchar_t *psz, size_t n) const
} }
} }
if (buf && (len<n)) *buf = 0; if (buf && (len<n))
*buf = 0;
return len; return len;
} }
@@ -1448,7 +1453,7 @@ size_t wxMBConv_iconv::MB2WC(wchar_t *buf, const char *psz, size_t n) const
// as MB<->WC conversion would fail "randomly". // as MB<->WC conversion would fail "randomly".
wxMutexLocker lock(wxConstCast(this, wxMBConv_iconv)->m_iconvMutex); wxMutexLocker lock(wxConstCast(this, wxMBConv_iconv)->m_iconvMutex);
#endif #endif
size_t inbuf = strlen(psz); size_t inbuf = strlen(psz);
size_t outbuf = n * SIZEOF_WCHAR_T; size_t outbuf = n * SIZEOF_WCHAR_T;
size_t res, cres; size_t res, cres;
@@ -1510,7 +1515,7 @@ size_t wxMBConv_iconv::WC2MB(char *buf, const wchar_t *psz, size_t n) const
// NB: explained in MB2WC // NB: explained in MB2WC
wxMutexLocker lock(wxConstCast(this, wxMBConv_iconv)->m_iconvMutex); wxMutexLocker lock(wxConstCast(this, wxMBConv_iconv)->m_iconvMutex);
#endif #endif
size_t inbuf = wxWcslen(psz) * SIZEOF_WCHAR_T; size_t inbuf = wxWcslen(psz) * SIZEOF_WCHAR_T;
size_t outbuf = n; size_t outbuf = n;
size_t res, cres; size_t res, cres;
@@ -2083,9 +2088,9 @@ public:
#if SIZEOF_WCHAR_T == 4 #if SIZEOF_WCHAR_T == 4
UniChar* szUniCharBuffer = new UniChar[nOutSize]; UniChar* szUniCharBuffer = new UniChar[nOutSize];
#endif #endif
CFStringGetCharacters(theString, theRange, szUniCharBuffer); CFStringGetCharacters(theString, theRange, szUniCharBuffer);
CFRelease(theString); CFRelease(theString);
szUniCharBuffer[nOutLength] = '\0' ; szUniCharBuffer[nOutLength] = '\0' ;
@@ -2095,14 +2100,14 @@ public:
converter.MB2WC(szOut, (const char*)szUniCharBuffer , nOutSize ) ; converter.MB2WC(szOut, (const char*)szUniCharBuffer , nOutSize ) ;
delete[] szUniCharBuffer; delete[] szUniCharBuffer;
#endif #endif
return nOutLength; return nOutLength;
} }
size_t WC2MB(char *szOut, const wchar_t *szUnConv, size_t nOutSize) const size_t WC2MB(char *szOut, const wchar_t *szUnConv, size_t nOutSize) const
{ {
wxASSERT(szUnConv); wxASSERT(szUnConv);
size_t nRealOutSize; size_t nRealOutSize;
size_t nBufSize = wxWcslen(szUnConv); size_t nBufSize = wxWcslen(szUnConv);
UniChar* szUniBuffer = (UniChar*) szUnConv; UniChar* szUniBuffer = (UniChar*) szUnConv;
@@ -2130,7 +2135,7 @@ public:
{ {
if (szOut != NULL) if (szOut != NULL)
CFStringGetCharacters(theString, CFRangeMake(0, nOutSize - 1), (UniChar*) szOut); CFStringGetCharacters(theString, CFRangeMake(0, nOutSize - 1), (UniChar*) szOut);
nRealOutSize = CFStringGetLength(theString) + 1; nRealOutSize = CFStringGetLength(theString) + 1;
} }
else else
@@ -2143,7 +2148,7 @@ public:
//0 tells CFString to return NULL if it meets such a character //0 tells CFString to return NULL if it meets such a character
false, //not an external representation false, //not an external representation
(UInt8*) szOut, (UInt8*) szOut,
nOutSize, nOutSize,
(CFIndex*) &nRealOutSize (CFIndex*) &nRealOutSize
); );
} }
@@ -2159,7 +2164,7 @@ public:
bool IsOk() const bool IsOk() const
{ {
return m_encoding != kCFStringEncodingInvalidId && return m_encoding != kCFStringEncodingInvalidId &&
CFStringIsEncodingAvailable(m_encoding); CFStringIsEncodingAvailable(m_encoding);
} }
@@ -2297,7 +2302,7 @@ public:
if ( buf && res < n) if ( buf && res < n)
{ {
buf[res] = 0; buf[res] = 0;
//we need to double-trip to verify it didn't insert any ? in place //we need to double-trip to verify it didn't insert any ? in place
//of bogus characters //of bogus characters
wxWCharBuffer wcBuf(n); wxWCharBuffer wcBuf(n);
@@ -2536,7 +2541,7 @@ wxMBConv *wxCSConv::DoCreate() const
#if defined(__WXMAC__) #if defined(__WXMAC__)
{ {
// leave UTF16 and UTF32 to the built-ins of wx // leave UTF16 and UTF32 to the built-ins of wx
if ( m_name || ( m_encoding < wxFONTENCODING_UTF16BE || if ( m_name || ( m_encoding < wxFONTENCODING_UTF16BE ||
( m_encoding >= wxFONTENCODING_MACMIN && m_encoding <= wxFONTENCODING_MACMAX ) ) ) ( m_encoding >= wxFONTENCODING_MACMIN && m_encoding <= wxFONTENCODING_MACMAX ) ) )
{ {