dprevot/php-sqlsrv
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity
php-sqlsrv/source/shared/localization.hpp

879 lines
32 KiB
C++
Raw Normal View History

added Linux specific files
2017-01-25 19:49:09 +01:00
//---------------------------------------------------------------------------------------------------------------------------------
// File: Localization.hpp
//
// Contents: Contains portable classes for localization
//
bumped up the version to 4.1
2017-01-27 02:17:12 +01:00
// Microsoft Drivers 4.1 for PHP for SQL Server
added Linux specific files
2017-01-25 19:49:09 +01:00
// Copyright(c) Microsoft Corporation
// All rights reserved.
// MIT License
// Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files(the ""Software""),
// to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense,
// and / or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions :
// The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
// THE SOFTWARE IS PROVIDED *AS IS*, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
// IN THE SOFTWARE.
//---------------------------------------------------------------------------------------------------------------------------------
#ifndef __LOCALIZATION_HPP__
#define __LOCALIZATION_HPP__
#include <time.h>
#include <assert.h>
#include "typedefs_for_linux.h"
#ifdef MPLAT_UNIX
namespace std
{
// Forward reference
class locale;
}
#endif
#define CP_UTF8 65001
#define CP_UTF16 1200
#define CP_UTF32 12000
#define CP_ACP 0 // default to ANSI code page
// This class provides allocation policies for the SystemLocale and AutoArray classes.
// This is primarily needed for the self-allocating ToUtf16/FromUtf16 methods.
// SNI needs all its allocations to use its own allocator so it would create a separate
// class that obeys this interface and provide it as a template parameter.
template< typename ArrayT >
struct ArrayTAllocator
{
static ArrayT * Alloc( size_t cch )
{
return reinterpret_cast< ArrayT * >( malloc(cch*sizeof(ArrayT)) );
}
// Realloc will free the 'old' memory if new memory was successfully allocated
// and copied to.
static ArrayT * Realloc( ArrayT * old, size_t cchNewSize )
{
return reinterpret_cast< ArrayT * >( realloc(old, cchNewSize*sizeof(ArrayT)) );
}
static void Free( ArrayT * mem )
{
free( mem );
}
};
// This is an auto_ptr-like class that is used with the SystemLocale.
// It allows for automatic freeing of the memory using the allocator policy.
// Callers would not normally use this class directly but would use one of the
// two specializations: AutoCharArray AutoWCharArray.
template< typename ArrayT, typename AllocT = ArrayTAllocator< ArrayT > >
struct AutoArray
{
size_t m_cchSize;
ArrayT * m_ptr;
AutoArray( const AutoArray & );
AutoArray & operator=( const AutoArray & );
AutoArray()
: m_cchSize( 0 ), m_ptr( NULL )
{
}
explicit AutoArray( size_t cchSize )
: m_cchSize( cchSize ), m_ptr( AllocT::Alloc(cchSize) )
{
}
virtual ~AutoArray()
{
Free();
}
void Free()
{
if ( NULL != m_ptr )
{
AllocT::Free( m_ptr );
m_ptr = NULL;
m_cchSize = 0;
}
}
bool Realloc( size_t cchSize )
{
ArrayT * newPtr = AllocT::Realloc( m_ptr, cchSize );
if ( NULL != newPtr )
{
// Safe to overwrite since Realloc freed m_ptr.
m_ptr = newPtr;
m_cchSize = cchSize;
return true;
}
return false;
}
ArrayT * Detach()
{
ArrayT * oldPtr = m_ptr;
m_ptr = NULL;
m_cchSize = 0;
return oldPtr;
}
void UpdateSize()
{
if ( NULL == m_ptr )
{
m_cchSize = 0;
}
else
{
// XPLAT_ODBC_TODO VSTS 819733 MPlat: Reconcile std c++ usage between platforms
// Should use char_traits<ArrayT>::length
ArrayT * end = m_ptr;
while ( (ArrayT)0 != *end++ )
;
// Want the null terminator included
m_cchSize = end - m_ptr;
}
}
};
class SystemLocale
{
public:
// -----------------------------------------------------------------------
// Public Static Functions
#ifdef MPLAT_UNIX
static const SystemLocale & Singleton();
#else
// Windows returns by value since this is an empty class
static const SystemLocale Singleton();
#endif
#ifdef MPLAT_UNIX
int GetResourcePath( char * buffer, size_t cchBuffer ) const;
static const int MINS_PER_HOUR = 60;
static const int MINS_PER_DAY = 24 * MINS_PER_HOUR;
// Returns the bias between the supplied utc and local times.
// utc = local + bias
static int BiasInMinutes( const struct tm & utc, const struct tm & local )
{
int bias = 0;
if ( utc.tm_mon != local.tm_mon )
{
// Offset crosses month boundary so one of two must be first day of month
if ( 1 == utc.tm_mday )
bias += MINS_PER_DAY;
else
{
assert( 1 == local.tm_mday );
bias -= MINS_PER_DAY;
}
}
else
{
bias += MINS_PER_DAY * (utc.tm_mday - local.tm_mday);
}
bias += MINS_PER_HOUR * (utc.tm_hour - local.tm_hour);
bias += (utc.tm_min - local.tm_min);
// Round based on diff in secs, in case utc/local straddle a day with leap seconds
int secs_diff = (utc.tm_sec - local.tm_sec);
if ( 29 < secs_diff )
++bias;
else if ( secs_diff < -29 )
--bias;
return bias;
}
// Returns both standard and daylight savings biases for the current year
// utc = local + bias
// Both might be equal if DST is not honored
// If platform doesn't know if bias is DST or standard (ie. unknown)
// then standard time is assumed.
// Note that applying current year's biases to dates from other years may result
// in incorrect time adjustments since regions change their rules over time.
// The current SNAC driver code uses this approach as well so we are doing this
// to preserve consistent behavior. If SNAC changes to lookup the offsets that
// were effective for a given date then we should update our logic here as well.
static DWORD TimeZoneBiases( int * stdInMinutes, int * dstInMinutes )
{
struct tm local, utc;
// Find current year
time_t now = time( NULL );
if ( (time_t)(-1) == now || NULL == localtime_r(&now, &local) )
return ERROR_INVALID_DATA;
// Find bias for first of each month until both STD and DST are found
// Possible perf improvements (can wait until perf tests indicate a need):
// Just use Dec 21 and Jun 21 (near the two soltices)
// Or calc once and cache (must be thread safe)
bool foundUNK = false;
bool foundSTD = false;
bool foundDST = false;
int std_bias = 0;
int dst_bias = 0;
local.tm_mday = 1;
for ( int mon = 0; mon < 12; ++mon )
{
local.tm_mon = mon;
if ( (time_t)(-1) == (now = mktime(&local)) || NULL == gmtime_r(&now, &utc) )
return ERROR_INVALID_DATA;
if ( 0 < local.tm_isdst )
{
if ( !foundDST )
{
dst_bias = BiasInMinutes( utc, local );
foundDST = true;
if ( foundSTD )
break; // Done checking when both STD & DST are found
}
}
else
{
// Time is STD or unknown, put in STD
if ( !foundSTD )
{
std_bias = BiasInMinutes( utc, local );
if ( local.tm_isdst < 0 )
foundUNK = true;
else
{
foundSTD = true;
if ( foundDST )
break; // Done checking when both STD and DST are found
}
}
}
}
// At least one of STD, DST, or unknown must have been set
assert( foundSTD || foundDST || foundUNK );
// For zones that don't observe DST (somewhat common),
// report DST bias as the same as STD
if ( !foundDST )
dst_bias = std_bias;
// For zones that ONLY observe DST (extremely rare if at all),
// report STD bias as the same as DST
if ( !foundSTD && !foundUNK )
std_bias = dst_bias;
*stdInMinutes = std_bias;
*dstInMinutes = dst_bias;
return ERROR_SUCCESS;
}
#endif
static DWORD CurrentLocalTime( LPSYSTEMTIME pTime );
// Multi-byte UTF8 code points start with '11xx xxxx'
static bool IsUtf8LeadByte( BYTE utf8 )
{
return (0xC0 == (utf8 & 0xC0));
}
// Maximum number of storage units (char or WCHAR)
// for a code page (e.g. UTF16 == 2 for surrogates)
static UINT MaxCharCchSize( UINT codepage );
// Inspects the byte at start, and returns the start
// of the next code point (possibly multiple bytes later).
// If NULL or start points at null terminator, than start is returned.
// If start points at a dangling UTF8 trail byte, then (start+1) is
// returned since we can't know how large this code point is.
static char * NextChar( UINT codepage, const char * start );
#ifdef MPLAT_UNIX
// This version is for non-null terminated strings.
// Last ptr will be one past end of buffer.
static char * NextChar( UINT codepage, const char * start, size_t cchBytesLeft );
#endif
// Given the start byte, how many total bytes are expected for
// this code point. If start is a UTF8 trail byte, then 1 is returned.
static UINT CchExpectedNextChar( UINT codepage, BYTE start )
{
if ( 0 == (start & (char)0x80) )
return 1; // ASCII
else if ( CP_UTF8 == codepage )
return IsUtf8LeadByte(start) ? CchUtf8CodePt(start) : 1;
else if ( IsDBCSLeadByteEx(codepage, start) )
return 2;
else
return 1;
}
// Returns the number of bytes that need to be trimmed to avoid splitting
// a multi-byte code point sequence at the end of the buffer.
// Returns zero if a trailing UTF8 code value is found but no
// matching lead byte was found for it (ie. invalid, dangling trail byte).
_Ret_range_(0, cchBuffer) static UINT TrimPartialCodePt( UINT codepage, _In_count_(cchBuffer) const BYTE * buffer, size_t cchBuffer )
{
if ( 0 == cchBuffer )
return 0;
if ( CP_UTF8 == codepage )
{
return TrimPartialUtf8CodePt( buffer, cchBuffer );
}
else
{
size_t i = cchBuffer;
for ( ; 0 < i; --i )
{
if ( !IsDBCSLeadByteEx( codepage, buffer[i-1] ) )
break;
}
// If odd, then last byte is truly a lead byte so return 1 byte to trim
return ((cchBuffer-i) & 1) ? 1 : 0;
}
}
// For all transcoding functions
// Returns zero on error. Do not call GetLastError() since that is not portable (pErrorCode has result of GetLastError()).
// pHasDataLoss will be true if an unrecognized code point was encountered in the source and a default output instead.
// Replaces calls to MultiByteToWideChar and WideCharToMultiByte
// Transcode between a code page and UTF16
static size_t ToUtf16( UINT srcCodePage, const char * src, SSIZE_T cchSrc,
__out_ecount_opt(cchDest) WCHAR * dest, size_t cchDest,
DWORD * pErrorCode = NULL );
static size_t ToUtf16Strict( UINT srcCodePage, const char * src, SSIZE_T cchSrc,
__out_ecount_opt(cchDest) WCHAR * dest, size_t cchDest,
DWORD * pErrorCode = NULL );
static size_t FromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc,
__out_ecount_opt(cchDest) char * dest, size_t cchDest,
bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL );
static size_t FromUtf16Strict(UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc,
__out_ecount_opt(cchDest) char * dest, size_t cchDest,
bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL);
// Allocates destination buffer to match required size
// Template is used so call can provide allocation policy
// Used instead of the Windows API pattern of calling with zero dest buffer size to find
// required buffer size, followed by second call with newly allocated buffer.
template< typename AllocT >
static size_t ToUtf16( UINT srcCodePage, const char * src, SSIZE_T cchSrc, __deref_out_ecount(1) WCHAR ** dest, DWORD * pErrorCode = NULL );
template< typename AllocT >
static size_t ToUtf16Strict( UINT srcCodePage, const char * src, SSIZE_T cchSrc, __deref_out_ecount(1) WCHAR ** dest, DWORD * pErrorCode = NULL );
template< typename AllocT >
static size_t FromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, __deref_out_ecount(1) char ** dest, bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL );
template< typename AllocT >
static size_t FromUtf16Strict(UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, __deref_out_ecount(1) char ** dest, bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL);
// -----------------------------------------------------------------------
// Public Member Functions
#ifndef TIME_ZONE_ID_UNKNOWN
#define TIME_ZONE_ID_UNKNOWN 0
#define TIME_ZONE_ID_STANDARD 1
#define TIME_ZONE_ID_DAYLIGHT 2
#endif
// pTZInfo, if supplied, holds one of the above defined values
DWORD CurrentTimeZoneBias( LONG * offsetInMinutes, DWORD * pTZInfo = NULL ) const;
// The Ansi code page, always UTF8 for Linux
UINT AnsiCP() const;
// Used for files (e.g. returns 437 on US Windows, UTF8 for Linux)
UINT OemCP() const;
// Returns UTF-16LE for all platforms (LE == Little Endian)
UINT WideCP() const
{
return CP_UTF16;
}
// Performs case folding to lower case using the current system locale
// Replaces calls to LCMapStringA
size_t ToLower( const char * src, SSIZE_T cchSrc, __out_ecount_opt(cchDest) char * dest, size_t cchDest, DWORD * pErrorCode = NULL ) const;
#ifndef CSTR_ERROR
#define CSTR_ERROR 0 // compare failed
#define CSTR_LESS_THAN 1 // string 1 less than string 2
#define CSTR_EQUAL 2 // string 1 equal to string 2
#define CSTR_GREATER_THAN 3 // string 1 greater than string 2
#endif
// String comparison using the rules of the current system locale.
// Replaces calls to CompareString
// Ignoring width (Bing for "Full Width Characters") has no affect on Linux
// Return value is one of the above defined values.
// On error, pErrorCode has result of GetLastError() (do not call GetLastError directly since it isn't portable).
int Compare( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode = NULL ) const;
int CompareIgnoreCase( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode = NULL ) const;
int CompareIgnoreWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode = NULL ) const;
int CompareIgnoreCaseAndWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode = NULL ) const;
private:
// Prevent copying.
// Also prevents misuse of return from Singleton() method.
// Since return types are different on Windows vs Linux,
// callers should not cache the result of Singleton().
SystemLocale( const SystemLocale & );
SystemLocale & operator=( const SystemLocale & );
#ifdef MPLAT_UNIX
// MPLAT_UNIX ----------------------------------------------------------------
std::locale * m_pLocale;
explicit SystemLocale( const char * localeName );
~SystemLocale();
static UINT ExpandSpecialCP( UINT codepage )
{
// Convert CP_ACP, CP_OEM to CP_UTF8
return (codepage < 2 ? CP_UTF8 : codepage);
}
// MPLAT_UNIX ----------------------------------------------------------------
#else
// !MPLAT_UNIX ---------------------------------------------------------------
SystemLocale() {}
static size_t ReturnCchResult( SSIZE_T cch, DWORD * pErrorCode )
{
if ( cch < 0 )
{
cch = 0;
}
if ( NULL != pErrorCode )
{
*pErrorCode = (0 == cch ? GetLastError() : ERROR_SUCCESS);
}
return static_cast<size_t>(cch);
}
static int CompareWithFlags( DWORD flags, const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode = NULL );
static size_t FastAsciiMultiByteToWideChar
(
UINT CodePage,
__in_ecount(cch) const char *pch, // IN | source string
SSIZE_T cch, // IN | count of characters or -1
__out_ecount_opt(cwch) PWCHAR pwch, // IN | Result string
size_t cwch, // IN | count of wchars of result buffer or 0
DWORD* pErrorCode, // OUT | optional pointer to return error code
bool bStrict = false // IN | Return error if invalid chars in src
);
static size_t FastAsciiWideCharToMultiByte
(
UINT CodePage,
const WCHAR *pwch, // IN | source string
SSIZE_T cwch, // IN | count of characters or -1
__out_bcount(cch) char *pch, // IN | Result string
size_t cch, // IN | Length of result buffer or 0
BOOL *pfDataLoss, // OUT | True if there was data loss during CP conversion
DWORD *pErrorCode // OUT | optional pointer to return error code
);
// !MPLAT_UNIX ---------------------------------------------------------------
#endif
// Returns the number of bytes this UTF8 code point expects
static UINT CchUtf8CodePt( BYTE codept )
{
assert( IsUtf8LeadByte(codept) );
// Initial byte of utf8 sequence indicates its length
// 110x xxxx = 2 bytes
// 1110 xxxx = 3 bytes
// 1111 0xxx = 4 bytes
// 1111 10xx = 5 bytes, future Unicode extension not covered by this logic
// 1111 110x = 6 bytes, future Unicode extension not covered by this logic
UINT expected_size = (0xC0 == (codept & 0xE0)) ? 2 : (0xE0 == (codept & 0xF0)) ? 3 : 4;
// Verify constraints
assert( 4 == MaxCharCchSize(CP_UTF8) );
return expected_size;
}
// Returns the number of bytes that need to be trimmed to avoid splitting
// a UTF8 code point sequence at the end of the buffer.
// Returns zero for ASCII.
// Also returns zero if a trailing UTF8 code value is found but no
// matching lead byte was found for it (ie. invalid, dangling trail byte).
static UINT TrimPartialUtf8CodePt( const BYTE * buffer, size_t cchBuffer )
{
if ( 0 == cchBuffer )
return 0;
if ( 0 == (buffer[cchBuffer-1] & 0x80) )
{
// Last char is ASCII so no trim needed
return 0;
}
// Last char is non-initial byte of multibyte utf8 sequence
// Need to determine if it is the last (ie. no trim need)
UINT cchMax = MaxCharCchSize( CP_UTF8 );
for ( UINT i = 1; 0 < cchBuffer && i <= cchMax; --cchBuffer, ++i )
{
if ( IsUtf8LeadByte(buffer[cchBuffer-1]) )
{
// Found initial byte, verify size of sequence
UINT cchExpected = CchUtf8CodePt( buffer[cchBuffer-1] );
if ( i == cchExpected )
return 0; // utf8 sequence is complete so no trim needed
else
{
assert( i <= cchBuffer );
return i; // trim the incomplete sequence
}
}
}
// Did not find initial utf8 byte so trim nothing
return 0;
}
};
// Convenience wrapper for converting from UTF16 into a newly
// allocated char[]. Class behaves like auto_ptr (will free in dtor,
// but has Release method so caller can take ownership of memory).
template< typename AllocT = ArrayTAllocator< char > >
struct AutoCharArray : public AutoArray< char, AllocT >
{
size_t AllocConvertFromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL )
{
char * converted = NULL;
size_t cchCvt = SystemLocale::FromUtf16< AllocT >( destCodePage, src, cchSrc, &converted, pHasDataLoss, pErrorCode );
if ( 0 < cchCvt )
{
this->Free();
this->m_ptr = converted;
this->m_cchSize = cchCvt;
}
return cchCvt;
}
};
// Convenience wrapper for converting to UTF16 into a newly
// allocated WCHAR[]. Class behaves like auto_ptr (will free in dtor,
// but has Release method so caller can take ownership of memory).
template< typename AllocT = ArrayTAllocator< WCHAR > >
struct AutoWCharArray : public AutoArray< WCHAR, AllocT >
{
size_t AllocConvertToUtf16( UINT destCodePage, const char * src, SSIZE_T cchSrc, bool * pHasDataLoss = NULL, DWORD * pErrorCode = NULL )
{
WCHAR * converted = NULL;
size_t cchCvt = SystemLocale::ToUtf16< AllocT >( destCodePage, src, cchSrc, &converted, pErrorCode );
if ( 0 < cchCvt )
{
this->Free();
this->m_ptr = converted;
this->m_cchSize = cchCvt;
}
return cchCvt;
}
};
// ---------------------------------------------------------------------------
// Inlines that vary by platform
#if defined(MPLAT_UNIX)
// MPLAT_UNIX ----------------------------------------------------------------
#include "globalization.h"
inline UINT SystemLocale::AnsiCP() const
{
return CP_UTF8;
}
inline UINT SystemLocale::OemCP() const
{
return CP_UTF8;
}
inline UINT SystemLocale::MaxCharCchSize( UINT codepage )
{
codepage = ExpandSpecialCP( codepage );
switch ( codepage )
{
case CP_UTF8:
return 4;
case 932:
case 936:
case 949:
case 950:
case CP_UTF16:
return 2;
default:
return 1;
}
}
inline int SystemLocale::CompareIgnoreWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
// XPLAT_ODBC_TODO: VSTS 806013 MPLAT: Support IgnoreWidth for SNI string comparisons
return Compare( left, cchLeft, right, cchRight, pErrorCode );
}
inline int SystemLocale::CompareIgnoreCaseAndWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
// XPLAT_ODBC_TODO: VSTS 806013 MPLAT: Support IgnoreWidth for SNI string comparisons
return CompareIgnoreCase( left, cchLeft, right, cchRight, pErrorCode );
}
template< typename AllocT >
inline size_t SystemLocale::ToUtf16( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR ** dest, DWORD * pErrorCode )
{
srcCodePage = ExpandSpecialCP( srcCodePage );
EncodingConverter cvt( CP_UTF16, srcCodePage );
if ( !cvt.Initialize() )
{
if ( NULL != pErrorCode )
*pErrorCode = ERROR_INVALID_PARAMETER;
return 0;
}
size_t cchSrcActual = (cchSrc < 0 ? (1+strlen(src)) : cchSrc);
bool hasLoss;
return cvt.Convert< WCHAR, char, AllocT >( dest, src, cchSrcActual, false, &hasLoss, pErrorCode );
}
template< typename AllocT >
inline size_t SystemLocale::ToUtf16Strict( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR ** dest, DWORD * pErrorCode )
{
srcCodePage = ExpandSpecialCP( srcCodePage );
EncodingConverter cvt( CP_UTF16, srcCodePage );
if ( !cvt.Initialize() )
{
if ( NULL != pErrorCode )
*pErrorCode = ERROR_INVALID_PARAMETER;
return 0;
}
size_t cchSrcActual = (cchSrc < 0 ? (1+strlen(src)) : cchSrc);
bool hasLoss;
return cvt.Convert< WCHAR, char, AllocT >( dest, src, cchSrcActual, true, &hasLoss, pErrorCode );
}
template< typename AllocT >
inline size_t SystemLocale::FromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, char ** dest, bool * pHasDataLoss, DWORD * pErrorCode )
{
destCodePage = ExpandSpecialCP( destCodePage );
EncodingConverter cvt( destCodePage, CP_UTF16 );
if ( !cvt.Initialize() )
{
if ( NULL != pErrorCode )
*pErrorCode = ERROR_INVALID_PARAMETER;
return 0;
}
size_t cchSrcActual = (cchSrc < 0 ? (1+mplat_wcslen(src)) : cchSrc);
bool hasLoss;
return cvt.Convert< char, WCHAR, AllocT >( dest, src, cchSrcActual, false, &hasLoss, pErrorCode );
}
// MPLAT_UNIX ----------------------------------------------------------------
#else
// ! MPLAT_UNIX ----------------------------------------------------------------
inline const SystemLocale SystemLocale::Singleton()
{
// On Windows, Localization is an empty class so creation of this
// should be optimized away. Empty classes have a sizeof 1 so there's
// something to take the address of.
C_ASSERT( 1 == sizeof(SystemLocale) );
return SystemLocale();
}
inline DWORD SystemLocale::CurrentTimeZoneBias( LONG * offsetInMinutes, DWORD * pTZInfo ) const
{
TIME_ZONE_INFORMATION tzi;
DWORD tzInfo;
if ( NULL == offsetInMinutes )
return ERROR_INVALID_PARAMETER;
else if ( TIME_ZONE_ID_INVALID == (tzInfo = GetTimeZoneInformation(&tzi)) )
return GetLastError();
else
{
*offsetInMinutes = tzi.Bias;
if ( NULL != pTZInfo )
*pTZInfo = tzInfo;
return ERROR_SUCCESS;
}
}
inline DWORD SystemLocale::CurrentLocalTime( LPSYSTEMTIME pTime )
{
GetLocalTime( pTime );
return ERROR_SUCCESS;
}
inline UINT SystemLocale::AnsiCP() const
{
return GetACP();
}
inline UINT SystemLocale::OemCP() const
{
return GetOEMCP();
}
inline UINT SystemLocale::MaxCharCchSize( UINT codepage )
{
CPINFO cpinfo;
BOOL rc = GetCPInfo( codepage, &cpinfo );
return (rc ? cpinfo.MaxCharSize : 0);
}
inline size_t SystemLocale::ToLower( const char * src, SSIZE_T cchSrc, char * dest, size_t cchDest, DWORD * pErrorCode ) const
{
// Windows API takes 'int' sized parameters
if ( cchSrc < -1 || 0x7FFFFFF < cchSrc || 0x7FFFFFF < cchDest )
{
if ( NULL != pErrorCode )
*pErrorCode = ERROR_INVALID_PARAMETER;
return 0;
}
OACR_WARNING_PUSH
OACR_WARNING_DISABLE(SYSTEM_LOCALE_MISUSE , " INTERNATIONALIZATION BASELINE AT KATMAI RTM. FUTURE ANALYSIS INTENDED. ")
OACR_WARNING_DISABLE(ANSI_APICALL, " Keeping the ANSI API for now. ")
int cch = LCMapStringA(
LOCALE_SYSTEM_DEFAULT,
LCMAP_LOWERCASE,
src,
(int)cchSrc,
dest,
(int)cchDest );
OACR_WARNING_POP
return ReturnCchResult( cch, pErrorCode );
}
inline int SystemLocale::CompareWithFlags( DWORD flags, const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode )
{
// Windows API takes 'int' sized parameters
if ( cchLeft < -1 || 0x7FFFFFF < cchLeft || cchRight < -1 || 0x7FFFFFF < cchRight )
{
if ( NULL != pErrorCode )
*pErrorCode = ERROR_INVALID_PARAMETER;
return 0;
}
OACR_WARNING_PUSH
OACR_WARNING_DISABLE(SYSTEM_LOCALE_MISUSE , " INTERNATIONALIZATION BASELINE AT KATMAI RTM. FUTURE ANALYSIS INTENDED. ")
int cmp = CompareStringA( LOCALE_SYSTEM_DEFAULT, flags, left, (int)cchLeft, right, (int)cchRight );
OACR_WARNING_POP
if ( NULL != pErrorCode )
{
*pErrorCode = (CSTR_ERROR == cmp ? GetLastError() : ERROR_SUCCESS);
}
return cmp;
}
inline int SystemLocale::Compare( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
return CompareWithFlags( 0, left, cchLeft, right, cchRight, pErrorCode );
}
inline int SystemLocale::CompareIgnoreCase( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
return CompareWithFlags( NORM_IGNORECASE, left, cchLeft, right, cchRight, pErrorCode );
}
inline int SystemLocale::CompareIgnoreCaseAndWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
return CompareWithFlags( NORM_IGNORECASE|NORM_IGNOREWIDTH, left, cchLeft, right, cchRight, pErrorCode );
}
inline int SystemLocale::CompareIgnoreWidth( const char * left, SSIZE_T cchLeft, const char * right, SSIZE_T cchRight, DWORD * pErrorCode ) const
{
return CompareWithFlags( NORM_IGNOREWIDTH, left, cchLeft, right, cchRight, pErrorCode );
}
inline char * SystemLocale::NextChar( UINT codepage, const char * start )
{
return CharNextExA( (WORD)codepage, start, 0 );
}
inline size_t SystemLocale::ToUtf16( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR * dest, size_t cchDest, DWORD * pErrorCode )
{
return FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, dest, cchDest, pErrorCode );
}
inline size_t SystemLocale::ToUtf16Strict( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR * dest, size_t cchDest, DWORD * pErrorCode )
{
return FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, dest, cchDest, pErrorCode, true );
}
inline size_t SystemLocale::FromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, char * dest, size_t cchDest, bool * pHasDataLoss, DWORD * pErrorCode )
{
BOOL dataloss = FALSE;
size_t cchCvt = FastAsciiWideCharToMultiByte( destCodePage, src, cchSrc, dest, cchDest, &dataloss, pErrorCode );
if ( NULL != pHasDataLoss )
{
*pHasDataLoss = (FALSE != dataloss);
}
return cchCvt;
}
template< typename AllocT >
inline size_t SystemLocale::ToUtf16( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR ** dest, DWORD * pErrorCode )
{
size_t cchCvt = FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, NULL, 0, pErrorCode );
if ( 0 < cchCvt )
{
AutoArray< WCHAR, AllocT > newDestBuffer( cchCvt );
cchCvt = FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, newDestBuffer.m_ptr, cchCvt, pErrorCode );
if ( 0 < cchCvt )
*dest = newDestBuffer.Detach();
}
return cchCvt;
}
template< typename AllocT >
inline size_t SystemLocale::ToUtf16Strict( UINT srcCodePage, const char * src, SSIZE_T cchSrc, WCHAR ** dest, DWORD * pErrorCode )
{
size_t cchCvt = FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, NULL, 0, pErrorCode, true );
if ( 0 < cchCvt )
{
AutoArray< WCHAR, AllocT > newDestBuffer( cchCvt );
cchCvt = FastAsciiMultiByteToWideChar( srcCodePage, src, cchSrc, newDestBuffer.m_ptr, cchCvt, pErrorCode, true );
if ( 0 < cchCvt )
*dest = newDestBuffer.Detach();
}
return cchCvt;
}
template< typename AllocT >
inline size_t SystemLocale::FromUtf16( UINT destCodePage, const WCHAR * src, SSIZE_T cchSrc, char ** dest, bool * pHasDataLoss, DWORD * pErrorCode )
{
BOOL dataloss = FALSE;
size_t cchCvt = FastAsciiWideCharToMultiByte( destCodePage, src, cchSrc, NULL, 0, &dataloss, pErrorCode );
if ( 0 < cchCvt )
{
AutoArray< char, AllocT > newDestBuffer( cchCvt );
cchCvt = FastAsciiWideCharToMultiByte( destCodePage, src, cchSrc, newDestBuffer.m_ptr, cchCvt, &dataloss, pErrorCode );
if ( 0 < cchCvt )
*dest = newDestBuffer.Detach();
}
if ( NULL != pHasDataLoss )
{
*pHasDataLoss = (FALSE != dataloss);
}
return cchCvt;
}
// ! MPLAT_UNIX ----------------------------------------------------------------
#endif
#endif // __LOCALIZATION_HPP__
Reference in a new issue Copy permalink