Remove unused locale-like code, utf8 functions and headers

This commit is contained in:
sfan5 2021-08-30 22:18:43 +02:00
parent 8f386999a6
commit 4389a15e0f
14 changed files with 5 additions and 714 deletions

@ -348,54 +348,6 @@ the engine will no longer read .png images. */
#undef _IRR_COMPILE_WITH_LIBPNG_ #undef _IRR_COMPILE_WITH_LIBPNG_
#endif #endif
//! Define _IRR_D3D_NO_SHADER_DEBUGGING to disable shader debugging in D3D9
/** If _IRR_D3D_NO_SHADER_DEBUGGING is undefined in IrrCompileConfig.h,
it is possible to debug all D3D9 shaders in VisualStudio. All shaders
(which have been generated in memory or read from archives for example) will be emitted
into a temporary file at runtime for this purpose. To debug your shaders, choose
Debug->Direct3D->StartWithDirect3DDebugging in Visual Studio, and for every shader a
file named 'irr_dbg_shader_%%.vsh' or 'irr_dbg_shader_%%.psh' will be created. Drag'n'drop
the file you want to debug into visual studio. That's it. You can now set breakpoints and
watch registers, variables etc. This works with ASM, HLSL, and both with pixel and vertex shaders.
Note that the engine will run in D3D REF for this, which is a lot slower than HAL. */
#define _IRR_D3D_NO_SHADER_DEBUGGING
#ifdef NO_IRR_D3D_NO_SHADER_DEBUGGING
#undef _IRR_D3D_NO_SHADER_DEBUGGING
#endif
//! Define _IRR_D3D_USE_LEGACY_HLSL_COMPILER to enable the old HLSL compiler in recent DX SDKs
/** This enables support for ps_1_x shaders for recent DX SDKs. Otherwise, support
for this shader model is not available anymore in SDKs after Oct2006. You need to
distribute the OCT2006_d3dx9_31_x86.cab or OCT2006_d3dx9_31_x64.cab though, in order
to provide the user with the proper DLL. That's why it's disabled by default. */
//#define _IRR_D3D_USE_LEGACY_HLSL_COMPILER
#ifdef NO_IRR_D3D_USE_LEGACY_HLSL_COMPILER
#undef _IRR_D3D_USE_LEGACY_HLSL_COMPILER
#endif
//! Define _IRR_USE_NVIDIA_PERFHUD_ to opt-in to using the nVidia PerHUD tool
/** Enable, by opting-in, to use the nVidia PerfHUD performance analysis driver
tool <http://developer.nvidia.com/object/nvperfhud_home.html>. */
#undef _IRR_USE_NVIDIA_PERFHUD_
//! Define one of the three setting for Burning's Video Software Rasterizer
/** So if we were marketing guys we could say Irrlicht has 4 Software-Rasterizers.
In a Nutshell:
All Burnings Rasterizers use 32 Bit Backbuffer, 32Bit Texture & 32 Bit Z or WBuffer,
16 Bit/32 Bit can be adjusted on a global flag.
BURNINGVIDEO_RENDERER_BEAUTIFUL
32 Bit + Vertexcolor + Lighting + Per Pixel Perspective Correct + SubPixel/SubTexel Correct +
Bilinear Texturefiltering + WBuffer
BURNINGVIDEO_RENDERER_FAST
32 Bit + Per Pixel Perspective Correct + SubPixel/SubTexel Correct + WBuffer +
Bilinear Dithering TextureFiltering + WBuffer
BURNINGVIDEO_RENDERER_ULTRA_FAST
16Bit + SubPixel/SubTexel Correct + ZBuffer
*/
//! Uncomment the following line if you want to ignore the deprecated warnings //! Uncomment the following line if you want to ignore the deprecated warnings
//#define IGNORE_DEPRECATED_WARNING //#define IGNORE_DEPRECATED_WARNING

@ -1,7 +0,0 @@
// Copyright (C) 2002-2012 Nikolaus Gebhardt
// This file is part of the "Irrlicht Engine".
// For conditions of distribution and use, see copyright notice in irrlicht.h
// replaced by template
#include "CMeshBuffer.h"

@ -1,7 +0,0 @@
// Copyright (C) 2002-2012 Nikolaus Gebhardt
// This file is part of the "Irrlicht Engine".
// For conditions of distribution and use, see copyright notice in irrlicht.h
// replaced by template
#include "CMeshBuffer.h"

@ -1,56 +0,0 @@
// Copyright (C) 2002-2012 Nikolaus Gebhardt
// This file is part of the "Irrlicht Engine".
// For conditions of distribution and use, see copyright notice in irrlicht.h
#ifndef __S_PARTICLE_H_INCLUDED__
#define __S_PARTICLE_H_INCLUDED__
#include "vector3d.h"
#include "dimension2d.h"
#include "SColor.h"
namespace irr
{
namespace scene
{
//! Struct for holding particle data
struct SParticle
{
//! Position of the particle
core::vector3df pos;
//! Direction and speed of the particle
core::vector3df vector;
//! Start life time of the particle
u32 startTime;
//! End life time of the particle
u32 endTime;
//! Current color of the particle
video::SColor color;
//! Original color of the particle.
/** That's the color of the particle it had when it was emitted. */
video::SColor startColor;
//! Original direction and speed of the particle.
/** The direction and speed the particle had when it was emitted. */
core::vector3df startVector;
//! Scale of the particle.
/** The current scale of the particle. */
core::dimension2df size;
//! Original scale of the particle.
/** The scale of the particle when it was emitted. */
core::dimension2df startSize;
};
} // end namespace scene
} // end namespace irr
#endif

@ -19,59 +19,6 @@ enum E_INDEX_TYPE
}; };
/*
//! vertex index used by the Irrlicht engine.
template <class T>
struct SSpecificVertexIndex
{
T Index;
//! default constructor
SSpecificVertexIndex() {}
//! constructor
SSpecificVertexIndex(u32 _index) :Index(_index) {}
bool operator==(const SSpecificVertexIndex& other) const
{
return (Index == other.Index);
}
bool operator!=(const SSpecificVertexIndex& other) const
{
return (Index != other.Index);
}
bool operator<(const SSpecificVertexIndex& other) const
{
return (Index < other.Index);
}
SSpecificVertexIndex operator+(const u32& other) const
{
return SSpecificVertexIndex(Index + other);
}
operator const u32() const
{
return (const u32)Index;
}
E_INDEX_TYPE getType() const
{
if (sizeof(T)==sizeof(u16))
return video::EIT_16BIT;
return video::EIT_32BIT;
}
};
//typedef SSpecificVertexIndex<u16> SVertexIndex;
typedef u32 SVertexIndex;
*/
} // end namespace video } // end namespace video
} // end namespace irr } // end namespace irr

@ -31,96 +31,6 @@ namespace scene
**/ **/
const c8* const ALLOW_ZWRITE_ON_TRANSPARENT = "Allow_ZWrite_On_Transparent"; const c8* const ALLOW_ZWRITE_ON_TRANSPARENT = "Allow_ZWrite_On_Transparent";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** Was used for changing the texture path of the built-in csm loader like this:
\code
SceneManager->getParameters()->setAttribute(scene::CSM_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const CSM_TEXTURE_PATH = "CSM_TexturePath";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** Was used for changing the texture path of the built-in lmts loader like this:
\code
SceneManager->getParameters()->setAttribute(scene::LMTS_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const LMTS_TEXTURE_PATH = "LMTS_TexturePath";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** Was used for changing the texture path of the built-in MY3D loader like this:
\code
SceneManager->getParameters()->setAttribute(scene::MY3D_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const MY3D_TEXTURE_PATH = "MY3D_TexturePath";
//! Name of the parameter specifying the COLLADA mesh loading mode
/**
Specifies if the COLLADA loader should create instances of the models, lights and
cameras when loading COLLADA meshes. By default, this is set to false. If this is
set to true, the ISceneManager::getMesh() method will only return a pointer to a
dummy mesh and create instances of all meshes and lights and cameras in the collada
file by itself. Example:
\code
SceneManager->getParameters()->setAttribute(scene::COLLADA_CREATE_SCENE_INSTANCES, true);
\endcode
*/
const c8* const COLLADA_CREATE_SCENE_INSTANCES = "COLLADA_CreateSceneInstances";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** This path is prefixed to the file names defined in the Deled file when loading
textures. This allows to alter the paths for a specific project setting.
Use it like this:
\code
SceneManager->getStringParameters()->setAttribute(scene::DMF_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const DMF_TEXTURE_PATH = "DMF_TexturePath";
//! Name of the parameter for preserving DMF textures dir structure with built-in DMF loader.
/** If this parameter is set to true, the texture directory defined in the Deled file
is ignored, and only the texture name is used to find the proper file. Otherwise, the
texture path is also used, which allows to use a nicer media layout.
Use it like this:
\code
//this way you won't use this setting (default)
SceneManager->getParameters()->setAttribute(scene::DMF_IGNORE_MATERIALS_DIRS, false);
\endcode
\code
//this way you'll use this setting
SceneManager->getParameters()->setAttribute(scene::DMF_IGNORE_MATERIALS_DIRS, true);
\endcode
**/
const c8* const DMF_IGNORE_MATERIALS_DIRS = "DMF_IgnoreMaterialsDir";
//! Name of the parameter for setting reference value of alpha in transparent materials.
/** Use it like this:
\code
//this way you'll set alpha ref to 0.1
SceneManager->getParameters()->setAttribute(scene::DMF_ALPHA_CHANNEL_REF, 0.1);
\endcode
**/
const c8* const DMF_ALPHA_CHANNEL_REF = "DMF_AlphaRef";
//! Name of the parameter for choose to flip or not tga files.
/** Use it like this:
\code
//this way you'll choose to flip alpha textures
SceneManager->getParameters()->setAttribute(scene::DMF_FLIP_ALPHA_TEXTURES, true);
\endcode
**/
const c8* const DMF_FLIP_ALPHA_TEXTURES = "DMF_FlipAlpha";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** Was used for changing the texture path of the built-in obj loader like this:
\code
SceneManager->getParameters()->setAttribute(scene::OBJ_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const OBJ_TEXTURE_PATH = "OBJ_TexturePath";
//! Flag to avoid loading group structures in .obj files //! Flag to avoid loading group structures in .obj files
/** Use it like this: /** Use it like this:
\code \code
@ -147,19 +57,6 @@ namespace scene
**/ **/
const c8* const B3D_LOADER_IGNORE_MIPMAP_FLAG = "B3D_IgnoreMipmapFlag"; const c8* const B3D_LOADER_IGNORE_MIPMAP_FLAG = "B3D_IgnoreMipmapFlag";
//! Deprecated, use IMeshLoader::getMeshTextureLoader()->setTexturePath instead.
/** Was used for changing the texture path of the built-in b3d loader like this:
\code
SceneManager->getParameters()->setAttribute(scene::B3D_TEXTURE_PATH, "path/to/your/textures");
\endcode
**/
const c8* const B3D_TEXTURE_PATH = "B3D_TexturePath";
//! Flag set as parameter when the scene manager is used as editor
/** In this way special animators like deletion animators can be stopped from
deleting scene nodes for example */
const c8* const IRR_SCENE_MANAGER_IS_EDITOR = "IRR_Editor";
//! Name of the parameter for setting the length of debug normals. //! Name of the parameter for setting the length of debug normals.
/** Use it like this: /** Use it like this:
\code \code

@ -194,9 +194,9 @@ static inline io::path mergeFilename(const io::path& path, const io::path& filen
//! some standard function ( to remove dependencies ) //! some standard function ( to remove dependencies )
inline s32 isdigit(s32 c) { return c >= '0' && c <= '9'; } inline bool isdigit(s32 c) { return c >= '0' && c <= '9'; }
inline s32 isspace(s32 c) { return c == ' ' || c == '\f' || c == '\n' || c == '\r' || c == '\t' || c == '\v'; } inline bool isspace(s32 c) { return c == ' ' || c == '\f' || c == '\n' || c == '\r' || c == '\t' || c == '\v'; }
inline s32 isupper(s32 c) { return c >= 'A' && c <= 'Z'; } inline bool isupper(s32 c) { return c >= 'A' && c <= 'Z'; }
} // end namespace core } // end namespace core

@ -12,17 +12,6 @@ namespace irr
{ {
namespace core namespace core
{ {
//! Selection of characters which count as decimal point in fast_atof
//! By default Irrlicht considers "." as the decimal point in numbers.
//! But sometimes you might run into situations where floats were written in
//! a local format with another decimal point like ",".
//! Best solution is usually to fix those cases by converting the input.
//! But if you don't have that choice you can set this to ".,".
//! WARNING: This is not thread-safe, so don't change while there's a chance
//! of another thread using fast_atof functions at the same time.
// TODO: This should probably also be used in irr::core::string, but
// the float-to-string code used there has to be rewritten first.
IRRLICHT_API extern irr::core::stringc LOCALE_DECIMAL_POINTS;
#define IRR_ATOF_TABLE_SIZE 17 #define IRR_ATOF_TABLE_SIZE 17
// we write [IRR_ATOF_TABLE_SIZE] here instead of [] to work around a swig bug // we write [IRR_ATOF_TABLE_SIZE] here instead of [] to work around a swig bug
@ -328,7 +317,7 @@ inline const char* fast_atof_move(const char* in, f32& result)
f32 value = strtof10(in, &in); f32 value = strtof10(in, &in);
if ( LOCALE_DECIMAL_POINTS.findFirst(*in) >= 0 ) if ( *in == '.' )
{ {
const char* afterDecimal = ++in; const char* afterDecimal = ++in;
const f32 decimal = strtof10(in, &afterDecimal); const f32 decimal = strtof10(in, &afterDecimal);

@ -38,29 +38,11 @@ template <typename T, typename TAlloc = irrAllocator<T> >
class string; class string;
static size_t multibyteToWString(string<wchar_t>& destination, const char* source, u32 sourceSize); static size_t multibyteToWString(string<wchar_t>& destination, const char* source, u32 sourceSize);
static size_t wStringToMultibyte(string<c8>& destination, const wchar_t* source, u32 sourceSize); static size_t wStringToMultibyte(string<c8>& destination, const wchar_t* source, u32 sourceSize);
inline s32 isdigit(s32 c); inline bool isdigit(s32 c);
enum eLocaleID
{
IRR_LOCALE_ANSI = 0,
IRR_LOCALE_GERMAN = 1
};
static eLocaleID locale_current = IRR_LOCALE_ANSI;
static inline void locale_set ( eLocaleID id )
{
locale_current = id;
}
//! Returns a character converted to lower case //! Returns a character converted to lower case
static inline u32 locale_lower ( u32 x ) static inline u32 locale_lower ( u32 x )
{ {
switch ( locale_current )
{
case IRR_LOCALE_GERMAN:
case IRR_LOCALE_ANSI:
break;
}
// ansi // ansi
return x >= 'A' && x <= 'Z' ? x + 0x20 : x; return x >= 'A' && x <= 'Z' ? x + 0x20 : x;
} }
@ -68,29 +50,10 @@ static inline u32 locale_lower ( u32 x )
//! Returns a character converted to upper case //! Returns a character converted to upper case
static inline u32 locale_upper ( u32 x ) static inline u32 locale_upper ( u32 x )
{ {
switch ( locale_current )
{
case IRR_LOCALE_GERMAN:
case IRR_LOCALE_ANSI:
break;
}
// ansi // ansi
return x >= 'a' && x <= 'z' ? x + ( 'A' - 'a' ) : x; return x >= 'a' && x <= 'z' ? x + ( 'A' - 'a' ) : x;
} }
//! Convert this utf-8-encoded string to the platform's wchar.
/** The resulting string is always NULL-terminated and well-formed.
\param len The size of the output buffer in bytes.
*/
IRRLICHT_API void utf8ToWchar(const char *in, wchar_t *out, const u64 len);
//! Convert this wchar string to utf-8.
/** The resulting string is always NULL-terminated and well-formed.
\param len The size of the output buffer in bytes.
*/
IRRLICHT_API void wcharToUtf8(const wchar_t *in, char *out, const u64 len);
template <typename T, typename TAlloc> template <typename T, typename TAlloc>
class string class string

@ -155,9 +155,6 @@
#include "SMaterial.h" #include "SMaterial.h"
#include "SMesh.h" #include "SMesh.h"
#include "SMeshBuffer.h" #include "SMeshBuffer.h"
#include "SMeshBufferLightMap.h"
#include "SMeshBufferTangents.h"
#include "SParticle.h"
#include "SSharedMeshBuffer.h" #include "SSharedMeshBuffer.h"
#include "SSkinMeshBuffer.h" #include "SSkinMeshBuffer.h"
#include "SVertexIndex.h" #include "SVertexIndex.h"

@ -207,7 +207,6 @@ add_library(IRROTHEROBJ OBJECT
os.cpp os.cpp
leakHunter.cpp leakHunter.cpp
CProfiler.cpp CProfiler.cpp
utf8.cpp
) )
if(ANDROID) if(ANDROID)

@ -50,8 +50,6 @@
#include "CSceneCollisionManager.h" #include "CSceneCollisionManager.h"
#include <locale.h>
namespace irr namespace irr
{ {
namespace scene namespace scene

@ -127,7 +127,6 @@ namespace irr
namespace core namespace core
{ {
const matrix4 IdentityMatrix(matrix4::EM4CONST_IDENTITY); const matrix4 IdentityMatrix(matrix4::EM4CONST_IDENTITY);
irr::core::stringc LOCALE_DECIMAL_POINTS(".");
} }
namespace video namespace video

@ -1,380 +0,0 @@
// Copyright (C) 2014 Lauri Kasanen
// This file is part of the "Irrlicht Engine". The UTF-8 functions are from physfs,
// under the zlib license, reproduced below.
#include "irrTypes.h"
#include "irrString.h"
namespace irr
{
namespace core
{
/*
Copyright (c) 2001-2011 Ryan C. Gordon and others.
This software is provided 'as-is', without any express or implied warranty.
In no event will the authors be held liable for any damages arising from
the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software in a
product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
Ryan C. Gordon <icculus@icculus.org>
*/
/*
* From rfc3629, the UTF-8 spec:
* http://www.ietf.org/rfc/rfc3629.txt
*
* Char. number range | UTF-8 octet sequence
* (hexadecimal) | (binary)
* --------------------+---------------------------------------------
* 0000 0000-0000 007F | 0xxxxxxx
* 0000 0080-0000 07FF | 110xxxxx 10xxxxxx
* 0000 0800-0000 FFFF | 1110xxxx 10xxxxxx 10xxxxxx
* 0001 0000-0010 FFFF | 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
*/
/*
* This may not be the best value, but it's one that isn't represented
* in Unicode (0x10FFFF is the largest codepoint value). We return this
* value from utf8codepoint() if there's bogus bits in the
* stream. utf8codepoint() will turn this value into something
* reasonable (like a question mark), for text that wants to try to recover,
* whereas utf8valid() will use the value to determine if a string has bad
* bits.
*/
#define UNICODE_BOGUS_CHAR_VALUE 0xFFFFFFFF
/*
* This is the codepoint we currently return when there was bogus bits in a
* UTF-8 string. May not fly in Asian locales?
*/
#define UNICODE_BOGUS_CHAR_CODEPOINT '?'
static u32 utf8codepoint(const char **_str)
{
const char *str = *_str;
u32 retval = 0;
u32 octet = (u32) ((u8) *str);
u32 octet2, octet3, octet4;
if (octet == 0) /* null terminator, end of string. */
return 0;
else if (octet < 128) /* one octet char: 0 to 127 */
{
(*_str)++; /* skip to next possible start of codepoint. */
return(octet);
} /* else if */
else if ((octet > 127) && (octet < 192)) /* bad (starts with 10xxxxxx). */
{
/*
* Apparently each of these is supposed to be flagged as a bogus
* char, instead of just resyncing to the next valid codepoint.
*/
(*_str)++; /* skip to next possible start of codepoint. */
return UNICODE_BOGUS_CHAR_VALUE;
} /* else if */
else if (octet < 224) /* two octets */
{
octet -= (128+64);
octet2 = (u32) ((u8) *(++str));
if ((octet2 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
*_str += 2; /* skip to next possible start of codepoint. */
retval = ((octet << 6) | (octet2 - 128));
if ((retval >= 0x80) && (retval <= 0x7FF))
return retval;
} /* else if */
else if (octet < 240) /* three octets */
{
octet -= (128+64+32);
octet2 = (u32) ((u8) *(++str));
if ((octet2 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet3 = (u32) ((u8) *(++str));
if ((octet3 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
*_str += 3; /* skip to next possible start of codepoint. */
retval = ( ((octet << 12)) | ((octet2-128) << 6) | ((octet3-128)) );
/* There are seven "UTF-16 surrogates" that are illegal in UTF-8. */
switch (retval)
{
case 0xD800:
case 0xDB7F:
case 0xDB80:
case 0xDBFF:
case 0xDC00:
case 0xDF80:
case 0xDFFF:
return UNICODE_BOGUS_CHAR_VALUE;
} /* switch */
/* 0xFFFE and 0xFFFF are illegal, too, so we check them at the edge. */
if ((retval >= 0x800) && (retval <= 0xFFFD))
return retval;
} /* else if */
else if (octet < 248) /* four octets */
{
octet -= (128+64+32+16);
octet2 = (u32) ((u8) *(++str));
if ((octet2 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet3 = (u32) ((u8) *(++str));
if ((octet3 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet4 = (u32) ((u8) *(++str));
if ((octet4 & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
*_str += 4; /* skip to next possible start of codepoint. */
retval = ( ((octet << 18)) | ((octet2 - 128) << 12) |
((octet3 - 128) << 6) | ((octet4 - 128)) );
if ((retval >= 0x10000) && (retval <= 0x10FFFF))
return retval;
} /* else if */
/*
* Five and six octet sequences became illegal in rfc3629.
* We throw the codepoint away, but parse them to make sure we move
* ahead the right number of bytes and don't overflow the buffer.
*/
else if (octet < 252) /* five octets */
{
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
*_str += 5; /* skip to next possible start of codepoint. */
return UNICODE_BOGUS_CHAR_VALUE;
} /* else if */
else /* six octets */
{
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
octet = (u32) ((u8) *(++str));
if ((octet & (128+64)) != 128) /* Format isn't 10xxxxxx? */
return UNICODE_BOGUS_CHAR_VALUE;
*_str += 6; /* skip to next possible start of codepoint. */
return UNICODE_BOGUS_CHAR_VALUE;
} /* else if */
return UNICODE_BOGUS_CHAR_VALUE;
} /* utf8codepoint */
static void PHYSFS_utf8ToUcs4(const char *src, u32 *dst, u64 len)
{
len -= sizeof (u32); /* save room for null char. */
while (len >= sizeof (u32))
{
u32 cp = utf8codepoint(&src);
if (cp == 0)
break;
else if (cp == UNICODE_BOGUS_CHAR_VALUE)
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
*(dst++) = cp;
len -= sizeof (u32);
} /* while */
*dst = 0;
} /* PHYSFS_utf8ToUcs4 */
static void PHYSFS_utf8ToUcs2(const char *src, u16 *dst, u64 len)
{
len -= sizeof (u16); /* save room for null char. */
while (len >= sizeof (u16))
{
u32 cp = utf8codepoint(&src);
if (cp == 0)
break;
else if (cp == UNICODE_BOGUS_CHAR_VALUE)
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
/* !!! BLUESKY: UTF-16 surrogates? */
if (cp > 0xFFFF)
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
*(dst++) = cp;
len -= sizeof (u16);
} /* while */
*dst = 0;
} /* PHYSFS_utf8ToUcs2 */
static void utf8fromcodepoint(u32 cp, char **_dst, u64 *_len)
{
char *dst = *_dst;
u64 len = *_len;
if (len == 0)
return;
if (cp > 0x10FFFF)
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
else if ((cp == 0xFFFE) || (cp == 0xFFFF)) /* illegal values. */
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
else
{
/* There are seven "UTF-16 surrogates" that are illegal in UTF-8. */
switch (cp)
{
case 0xD800:
case 0xDB7F:
case 0xDB80:
case 0xDBFF:
case 0xDC00:
case 0xDF80:
case 0xDFFF:
cp = UNICODE_BOGUS_CHAR_CODEPOINT;
} /* switch */
} /* else */
/* Do the encoding... */
if (cp < 0x80)
{
*(dst++) = (char) cp;
len--;
} /* if */
else if (cp < 0x800)
{
if (len < 2)
len = 0;
else
{
*(dst++) = (char) ((cp >> 6) | 128 | 64);
*(dst++) = (char) (cp & 0x3F) | 128;
len -= 2;
} /* else */
} /* else if */
else if (cp < 0x10000)
{
if (len < 3)
len = 0;
else
{
*(dst++) = (char) ((cp >> 12) | 128 | 64 | 32);
*(dst++) = (char) ((cp >> 6) & 0x3F) | 128;
*(dst++) = (char) (cp & 0x3F) | 128;
len -= 3;
} /* else */
} /* else if */
else
{
if (len < 4)
len = 0;
else
{
*(dst++) = (char) ((cp >> 18) | 128 | 64 | 32 | 16);
*(dst++) = (char) ((cp >> 12) & 0x3F) | 128;
*(dst++) = (char) ((cp >> 6) & 0x3F) | 128;
*(dst++) = (char) (cp & 0x3F) | 128;
len -= 4;
} /* else if */
} /* else */
*_dst = dst;
*_len = len;
} /* utf8fromcodepoint */
#define UTF8FROMTYPE(typ, src, dst, len) \
if (len == 0) return; \
len--; \
while (len) \
{ \
const u32 cp = (u32) ((typ) (*(src++))); \
if (cp == 0) break; \
utf8fromcodepoint(cp, &dst, &len); \
} \
*dst = '\0'; \
static void PHYSFS_utf8FromUcs4(const u32 *src, char *dst, u64 len)
{
UTF8FROMTYPE(u32, src, dst, len);
} /* PHYSFS_utf8FromUcs4 */
static void PHYSFS_utf8FromUcs2(const u16 *src, char *dst, u64 len)
{
UTF8FROMTYPE(u64, src, dst, len);
} /* PHYSFS_utf8FromUcs4 */
#undef UTF8FROMTYPE
void utf8ToWchar(const char *in, wchar_t *out, const u64 len)
{
#ifdef _WIN32
PHYSFS_utf8ToUcs2(in, (u16 *) out, len);
#else
PHYSFS_utf8ToUcs4(in, (u32 *) out, len);
#endif
}
void wcharToUtf8(const wchar_t *in, char *out, const u64 len)
{
#ifdef _WIN32
PHYSFS_utf8FromUcs2((const u16 *) in, out, len);
#else
PHYSFS_utf8FromUcs4((const u32 *) in, out, len);
#endif
}
} // end namespace core
} // end namespace irr