diff --git a/desmume/src/Makefile.am b/desmume/src/Makefile.am index cdbb6211a..5068c1c0e 100644 --- a/desmume/src/Makefile.am +++ b/desmume/src/Makefile.am @@ -44,7 +44,7 @@ libdesmume_a_SOURCES = \ PACKED.h PACKED_END.h \ utils/advanscene.cpp utils/advanscene.h \ utils/datetime.cpp utils/datetime.h \ - utils/guid.cpp utils/guid.h \ + utils/ConvertUTF.c utils/ConvertUTF.h utils/guid.cpp utils/guid.h \ utils/emufat.cpp utils/emufat.h utils/emufat_types.h \ utils/fsnitro.cpp utils/fsnitro.h \ utils/md5.cpp utils/md5.h utils/valuearray.h utils/xstring.cpp utils/xstring.h \ diff --git a/desmume/src/frontend/windows/DeSmuME.vcxproj b/desmume/src/frontend/windows/DeSmuME.vcxproj index 34c1238b6..a3956a7c1 100644 --- a/desmume/src/frontend/windows/DeSmuME.vcxproj +++ b/desmume/src/frontend/windows/DeSmuME.vcxproj @@ -105,7 +105,6 @@ - @@ -202,6 +201,7 @@ + @@ -373,7 +373,6 @@ - @@ -472,6 +471,7 @@ + diff --git a/desmume/src/frontend/windows/DeSmuME.vcxproj.filters b/desmume/src/frontend/windows/DeSmuME.vcxproj.filters index acc455362..f551f4dc0 100644 --- a/desmume/src/frontend/windows/DeSmuME.vcxproj.filters +++ b/desmume/src/frontend/windows/DeSmuME.vcxproj.filters @@ -124,12 +124,6 @@ {37e11fb9-8dec-43bd-8242-a721d69a740f} - - {fcb28aeb-7aa7-4e91-9592-692ec23b5ed4} - - - {dc595f38-b67d-4b99-93b5-d2e9628ef712} - @@ -261,6 +255,9 @@ Core\gdbstub + + Core\utils + Core\utils @@ -978,9 +975,6 @@ Windows - - Core\libretro-common\encodings - @@ -1103,6 +1097,9 @@ Core\gdbstub + + Core\utils + Core\utils @@ -1751,9 +1748,6 @@ Core\utils - - Core\libretro-common\include\encodings - diff --git a/desmume/src/frontend/windows/desmume.props b/desmume/src/frontend/windows/desmume.props index a3293272e..8e778e61e 100644 --- a/desmume/src/frontend/windows/desmume.props +++ b/desmume/src/frontend/windows/desmume.props @@ -91,9 +91,6 @@ _CRT_SECURE_NO_DEPRECATE;WIN32;HAVE_ZLIB_DEFLATE;HAVE_LIBZ;NOMINMAX;%(PreprocessorDefinitions) - - HAVE_UTF8;%(PreprocessorDefinitions) - _DEBUG;%(PreprocessorDefinitions) RELEASE;NDEBUG;%(PreprocessorDefinitions) diff --git a/desmume/src/movie.cpp b/desmume/src/movie.cpp index d1d8a7b39..acc5cc427 100644 --- a/desmume/src/movie.cpp +++ b/desmume/src/movie.cpp @@ -26,6 +26,7 @@ #include "utils/guid.h" #include "utils/xstring.h" #include "utils/datetime.h" +#include "utils/ConvertUTF.h" #include "MMU.h" #include "NDSSystem.h" @@ -39,8 +40,6 @@ #include "path.h" #include "emufile.h" -#include "encodings/utf.h" - using namespace std; bool freshMovie = false; //True when a movie loads, false when movie is altered. Used to determine if a movie has been altered since opening bool autoMovieBackup = true; @@ -301,14 +300,17 @@ int MovieData::dump(EMUFILE* fp, bool binary) fp->fprintf("bootFromFirmware %d\n", CommonSettings.BootFromFirmware?1:0); } else { - u8 fwNicknameUTF8[MAX_FW_NICKNAME_LENGTH*4]; + UTF8 fwNicknameUTF8[MAX_FW_NICKNAME_LENGTH*4]; + UTF8 *fwNicknameUTF8Start = fwNicknameUTF8; + const UTF16 *fwNicknameUTF16 = (const UTF16 *)CommonSettings.fw_config.nickname; memset(fwNicknameUTF8, 0, sizeof(fwNicknameUTF8)); - size_t junk; - utf16_conv_utf8(fwNicknameUTF8,&junk,CommonSettings.fw_config.nickname,CommonSettings.fw_config.nickname_len); + ConvertUTF16toUTF8(&fwNicknameUTF16, fwNicknameUTF16 + CommonSettings.fw_config.nickname_len, &fwNicknameUTF8Start, fwNicknameUTF8Start + sizeof(fwNicknameUTF8), strictConversion); - u8 fwMessageUTF8[MAX_FW_MESSAGE_LENGTH*4]; + UTF8 fwMessageUTF8[MAX_FW_MESSAGE_LENGTH*4]; + UTF8 *fwMessageUTF8Start = fwMessageUTF8; + const UTF16 *fwMessageUTF16 = (const UTF16 *)CommonSettings.fw_config.message; memset(fwMessageUTF8, 0, sizeof(fwMessageUTF8)); - utf16_conv_utf8(fwMessageUTF8,&junk,CommonSettings.fw_config.message,CommonSettings.fw_config.message_len); + ConvertUTF16toUTF8(&fwMessageUTF16, fwMessageUTF16 + CommonSettings.fw_config.message_len, &fwMessageUTF8Start, fwMessageUTF8Start + sizeof(fwMessageUTF8), strictConversion); fp->fprintf("firmNickname %s\n", fwNicknameUTF8); fp->fprintf("firmMessage %s\n", fwMessageUTF8); diff --git a/desmume/src/utils/ConvertUTF.c b/desmume/src/utils/ConvertUTF.c new file mode 100644 index 000000000..9b3deebd6 --- /dev/null +++ b/desmume/src/utils/ConvertUTF.c @@ -0,0 +1,539 @@ +/* + * Copyright 2001-2004 Unicode, Inc. + * + * Disclaimer + * + * This source code is provided as is by Unicode, Inc. No claims are + * made as to fitness for any particular purpose. No warranties of any + * kind are expressed or implied. The recipient agrees to determine + * applicability of information provided. If this file has been + * purchased on magnetic or optical media from Unicode, Inc., the + * sole remedy for any claim will be exchange of defective media + * within 90 days of receipt. + * + * Limitations on Rights to Redistribute This Code + * + * Unicode, Inc. hereby grants the right to freely use the information + * supplied in this file in the creation of products supporting the + * Unicode Standard, and to make copies of this file in any form + * for internal or external distribution as long as this notice + * remains attached. + */ + +/* --------------------------------------------------------------------- + + Conversions between UTF32, UTF-16, and UTF-8. Source code file. + Author: Mark E. Davis, 1994. + Rev History: Rick McGowan, fixes & updates May 2001. + Sept 2001: fixed const & error conditions per + mods suggested by S. Parent & A. Lillich. + June 2002: Tim Dodd added detection and handling of incomplete + source sequences, enhanced error detection, added casts + to eliminate compiler warnings. + July 2003: slight mods to back out aggressive FFFE detection. + Jan 2004: updated switches in from-UTF8 conversions. + Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions. + + See the header file "ConvertUTF.h" for complete documentation. + +------------------------------------------------------------------------ */ + + +#include "ConvertUTF.h" +#ifdef CVTUTF_DEBUG +#include +#endif + +static const int halfShift = 10; /* used for shifting by 10 bits */ + +static const UTF32 halfBase = 0x0010000UL; +static const UTF32 halfMask = 0x3FFUL; + +#define UNI_SUR_HIGH_START (UTF32)0xD800 +#define UNI_SUR_HIGH_END (UTF32)0xDBFF +#define UNI_SUR_LOW_START (UTF32)0xDC00 +#define UNI_SUR_LOW_END (UTF32)0xDFFF +#define false 0 +#define true 1 + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF32toUTF16 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF32* source = *sourceStart; + UTF16* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch; + if (target >= targetEnd) { + result = targetExhausted; break; + } + ch = *source++; + if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */ + /* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + *target++ = (UTF16)ch; /* normal case */ + } + } else if (ch > UNI_MAX_LEGAL_UTF32) { + if (flags == strictConversion) { + result = sourceIllegal; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + /* target is a character in range 0xFFFF - 0x10FFFF. */ + if (target + 1 >= targetEnd) { + --source; /* Back up source pointer! */ + result = targetExhausted; break; + } + ch -= halfBase; + *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START); + *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START); + } + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF16toUTF32 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF16* source = *sourceStart; + UTF32* target = *targetStart; + UTF32 ch, ch2; + while (source < sourceEnd) { + const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */ + ch = *source++; + /* If we have a surrogate pair, convert to UTF32 first. */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) { + /* If the 16 bits following the high surrogate are in the source buffer... */ + if (source < sourceEnd) { + ch2 = *source; + /* If it's a low surrogate, convert to UTF32. */ + if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) { + ch = ((ch - UNI_SUR_HIGH_START) << halfShift) + + (ch2 - UNI_SUR_LOW_START) + halfBase; + ++source; + } else if (flags == strictConversion) { /* it's an unpaired high surrogate */ + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } else { /* We don't have the 16 bits following the high surrogate. */ + --source; /* return to the high surrogate */ + result = sourceExhausted; + break; + } + } else if (flags == strictConversion) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } + if (target >= targetEnd) { + source = oldSource; /* Back up source pointer! */ + result = targetExhausted; break; + } + *target++ = ch; + } + *sourceStart = source; + *targetStart = target; +#ifdef CVTUTF_DEBUG +if (result == sourceIllegal) { + fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2); + fflush(stderr); +} +#endif + return result; +} + +/* --------------------------------------------------------------------- */ + +/* + * Index into the table below with the first byte of a UTF-8 sequence to + * get the number of trailing bytes that are supposed to follow it. + * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is + * left as-is for anyone who may want to do such conversion, which was + * allowed in earlier algorithms. + */ +static const char trailingBytesForUTF8[256] = { + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 +}; + +/* + * Magic values subtracted from a buffer value during UTF8 conversion. + * This table contains as many values as there might be trailing bytes + * in a UTF-8 sequence. + */ +static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, + 0x03C82080UL, 0xFA082080UL, 0x82082080UL }; + +/* + * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed + * into the first byte, depending on how many bytes follow. There are + * as many entries in this table as there are UTF-8 sequence types. + * (I.e., one byte sequence, two byte... etc.). Remember that sequencs + * for *legal* UTF-8 will be 4 or fewer bytes total. + */ +static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC }; + +/* --------------------------------------------------------------------- */ + +/* The interface converts a whole buffer to avoid function-call overhead. + * Constants have been gathered. Loops & conditionals have been removed as + * much as possible for efficiency, in favor of drop-through switches. + * (See "Note A" at the bottom of the file for equivalent code.) + * If your compiler supports it, the "isLegalUTF8" call can be turned + * into an inline function. + */ + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF16toUTF8 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF16* source = *sourceStart; + UTF8* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch; + unsigned short bytesToWrite = 0; + const UTF32 byteMask = 0xBF; + const UTF32 byteMark = 0x80; + const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */ + ch = *source++; + /* If we have a surrogate pair, convert to UTF32 first. */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) { + /* If the 16 bits following the high surrogate are in the source buffer... */ + if (source < sourceEnd) { + UTF32 ch2 = *source; + /* If it's a low surrogate, convert to UTF32. */ + if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) { + ch = ((ch - UNI_SUR_HIGH_START) << halfShift) + + (ch2 - UNI_SUR_LOW_START) + halfBase; + ++source; + } else if (flags == strictConversion) { /* it's an unpaired high surrogate */ + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } else { /* We don't have the 16 bits following the high surrogate. */ + --source; /* return to the high surrogate */ + result = sourceExhausted; + break; + } + } else if (flags == strictConversion) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } + /* Figure out how many bytes the result will require */ + if (ch < (UTF32)0x80) { bytesToWrite = 1; + } else if (ch < (UTF32)0x800) { bytesToWrite = 2; + } else if (ch < (UTF32)0x10000) { bytesToWrite = 3; + } else if (ch < (UTF32)0x110000) { bytesToWrite = 4; + } else { bytesToWrite = 3; + ch = UNI_REPLACEMENT_CHAR; + } + + target += bytesToWrite; + if (target > targetEnd) { + source = oldSource; /* Back up source pointer! */ + target -= bytesToWrite; result = targetExhausted; break; + } + switch (bytesToWrite) { /* note: everything falls through. */ + case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 1: *--target = (UTF8)(ch | firstByteMark[bytesToWrite]); + } + target += bytesToWrite; + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +/* + * Utility routine to tell whether a sequence of bytes is legal UTF-8. + * This must be called with the length pre-determined by the first byte. + * If not calling this from ConvertUTF8to*, then the length can be set by: + * length = trailingBytesForUTF8[*source]+1; + * and the sequence is illegal right away if there aren't that many bytes + * available. + * If presented with a length > 4, this returns false. The Unicode + * definition of UTF-8 goes up to 4-byte sequences. + */ + +static Boolean isLegalUTF8(const UTF8 *source, int length) { + UTF8 a; + const UTF8 *srcptr = source+length; + switch (length) { + default: return false; + /* Everything else falls through when "true"... */ + case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false; + case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false; + case 2: if ((a = (*--srcptr)) > 0xBF) return false; + + switch (*source) { + /* no fall-through in this inner switch */ + case 0xE0: if (a < 0xA0) return false; break; + case 0xED: if (a > 0x9F) return false; break; + case 0xF0: if (a < 0x90) return false; break; + case 0xF4: if (a > 0x8F) return false; break; + default: if (a < 0x80) return false; + } + + case 1: if (*source >= 0x80 && *source < 0xC2) return false; + } + if (*source > 0xF4) return false; + return true; +} + +/* --------------------------------------------------------------------- */ + +/* + * Exported function to return whether a UTF-8 sequence is legal or not. + * This is not used here; it's just exported. + */ +Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) { + int length = trailingBytesForUTF8[*source]+1; + if (source+length > sourceEnd) { + return false; + } + return isLegalUTF8(source, length); +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF8toUTF16 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF8* source = *sourceStart; + UTF16* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch = 0; + unsigned short extraBytesToRead = trailingBytesForUTF8[*source]; + if (source + extraBytesToRead >= sourceEnd) { + result = sourceExhausted; break; + } + /* Do this check whether lenient or strict */ + if (! isLegalUTF8(source, extraBytesToRead+1)) { + result = sourceIllegal; + break; + } + /* + * The cases all fall through. See "Note A" below. + */ + switch (extraBytesToRead) { + case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + case 3: ch += *source++; ch <<= 6; + case 2: ch += *source++; ch <<= 6; + case 1: ch += *source++; ch <<= 6; + case 0: ch += *source++; + } + ch -= offsetsFromUTF8[extraBytesToRead]; + + if (target >= targetEnd) { + source -= (extraBytesToRead+1); /* Back up source pointer! */ + result = targetExhausted; break; + } + if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */ + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + *target++ = (UTF16)ch; /* normal case */ + } + } else if (ch > UNI_MAX_UTF16) { + if (flags == strictConversion) { + result = sourceIllegal; + source -= (extraBytesToRead+1); /* return to the start */ + break; /* Bail out; shouldn't continue */ + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + /* target is a character in range 0xFFFF - 0x10FFFF. */ + if (target + 1 >= targetEnd) { + source -= (extraBytesToRead+1); /* Back up source pointer! */ + result = targetExhausted; break; + } + ch -= halfBase; + *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START); + *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START); + } + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF32toUTF8 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF32* source = *sourceStart; + UTF8* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch; + unsigned short bytesToWrite = 0; + const UTF32 byteMask = 0xBF; + const UTF32 byteMark = 0x80; + ch = *source++; + if (flags == strictConversion ) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + --source; /* return to the illegal value itself */ + result = sourceIllegal; + break; + } + } + /* + * Figure out how many bytes the result will require. Turn any + * illegally large UTF32 things (> Plane 17) into replacement chars. + */ + if (ch < (UTF32)0x80) { bytesToWrite = 1; + } else if (ch < (UTF32)0x800) { bytesToWrite = 2; + } else if (ch < (UTF32)0x10000) { bytesToWrite = 3; + } else if (ch <= UNI_MAX_LEGAL_UTF32) { bytesToWrite = 4; + } else { bytesToWrite = 3; + ch = UNI_REPLACEMENT_CHAR; + result = sourceIllegal; + } + + target += bytesToWrite; + if (target > targetEnd) { + --source; /* Back up source pointer! */ + target -= bytesToWrite; result = targetExhausted; break; + } + switch (bytesToWrite) { /* note: everything falls through. */ + case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; + case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]); + } + target += bytesToWrite; + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- */ + +ConversionResult ConvertUTF8toUTF32 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) { + ConversionResult result = conversionOK; + const UTF8* source = *sourceStart; + UTF32* target = *targetStart; + while (source < sourceEnd) { + UTF32 ch = 0; + unsigned short extraBytesToRead = trailingBytesForUTF8[*source]; + if (source + extraBytesToRead >= sourceEnd) { + result = sourceExhausted; break; + } + /* Do this check whether lenient or strict */ + if (! isLegalUTF8(source, extraBytesToRead+1)) { + result = sourceIllegal; + break; + } + /* + * The cases all fall through. See "Note A" below. + */ + switch (extraBytesToRead) { + case 5: ch += *source++; ch <<= 6; + case 4: ch += *source++; ch <<= 6; + case 3: ch += *source++; ch <<= 6; + case 2: ch += *source++; ch <<= 6; + case 1: ch += *source++; ch <<= 6; + case 0: ch += *source++; + } + ch -= offsetsFromUTF8[extraBytesToRead]; + + if (target >= targetEnd) { + source -= (extraBytesToRead+1); /* Back up the source pointer! */ + result = targetExhausted; break; + } + if (ch <= UNI_MAX_LEGAL_UTF32) { + /* + * UTF-16 surrogate values are illegal in UTF-32, and anything + * over Plane 17 (> 0x10FFFF) is illegal. + */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the illegal value itself */ + result = sourceIllegal; + break; + } else { + *target++ = UNI_REPLACEMENT_CHAR; + } + } else { + *target++ = ch; + } + } else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */ + result = sourceIllegal; + *target++ = UNI_REPLACEMENT_CHAR; + } + } + *sourceStart = source; + *targetStart = target; + return result; +} + +/* --------------------------------------------------------------------- + + Note A. + The fall-through switches in UTF-8 reading code save a + temp variable, some decrements & conditionals. The switches + are equivalent to the following loop: + { + int tmpBytesToRead = extraBytesToRead+1; + do { + ch += *source++; + --tmpBytesToRead; + if (tmpBytesToRead) ch <<= 6; + } while (tmpBytesToRead > 0); + } + In UTF-8 writing code, the switches on "bytesToWrite" are + similarly unrolled loops. + + --------------------------------------------------------------------- */ diff --git a/desmume/src/utils/ConvertUTF.h b/desmume/src/utils/ConvertUTF.h new file mode 100644 index 000000000..9f57bd438 --- /dev/null +++ b/desmume/src/utils/ConvertUTF.h @@ -0,0 +1,162 @@ +/* + * Copyright 2001-2004 Unicode, Inc. + * + * Disclaimer + * + * This source code is provided as is by Unicode, Inc. No claims are + * made as to fitness for any particular purpose. No warranties of any + * kind are expressed or implied. The recipient agrees to determine + * applicability of information provided. If this file has been + * purchased on magnetic or optical media from Unicode, Inc., the + * sole remedy for any claim will be exchange of defective media + * within 90 days of receipt. + * + * Limitations on Rights to Redistribute This Code + * + * Unicode, Inc. hereby grants the right to freely use the information + * supplied in this file in the creation of products supporting the + * Unicode Standard, and to make copies of this file in any form + * for internal or external distribution as long as this notice + * remains attached. + */ + +/* --------------------------------------------------------------------- + + Conversions between UTF32, UTF-16, and UTF-8. Header file. + + Several funtions are included here, forming a complete set of + conversions between the three formats. UTF-7 is not included + here, but is handled in a separate source file. + + Each of these routines takes pointers to input buffers and output + buffers. The input buffers are const. + + Each routine converts the text between *sourceStart and sourceEnd, + putting the result into the buffer between *targetStart and + targetEnd. Note: the end pointers are *after* the last item: e.g. + *(sourceEnd - 1) is the last item. + + The return result indicates whether the conversion was successful, + and if not, whether the problem was in the source or target buffers. + (Only the first encountered problem is indicated.) + + After the conversion, *sourceStart and *targetStart are both + updated to point to the end of last text successfully converted in + the respective buffers. + + Input parameters: + sourceStart - pointer to a pointer to the source buffer. + The contents of this are modified on return so that + it points at the next thing to be converted. + targetStart - similarly, pointer to pointer to the target buffer. + sourceEnd, targetEnd - respectively pointers to the ends of the + two buffers, for overflow checking only. + + These conversion functions take a ConversionFlags argument. When this + flag is set to strict, both irregular sequences and isolated surrogates + will cause an error. When the flag is set to lenient, both irregular + sequences and isolated surrogates are converted. + + Whether the flag is strict or lenient, all illegal sequences will cause + an error return. This includes sequences such as: , , + or in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code + must check for illegal sequences. + + When the flag is set to lenient, characters over 0x10FFFF are converted + to the replacement character; otherwise (when the flag is set to strict) + they constitute an error. + + Output parameters: + The value "sourceIllegal" is returned from some routines if the input + sequence is malformed. When "sourceIllegal" is returned, the source + value will point to the illegal value that caused the problem. E.g., + in UTF-8 when a sequence is malformed, it points to the start of the + malformed sequence. + + Author: Mark E. Davis, 1994. + Rev History: Rick McGowan, fixes & updates May 2001. + Fixes & updates, Sept 2001. + +------------------------------------------------------------------------ */ + +/* --------------------------------------------------------------------- + The following 4 definitions are compiler-specific. + The C standard does not guarantee that wchar_t has at least + 16 bits, so wchar_t is no less portable than unsigned short! + All should be unsigned values to avoid sign extension during + bit mask & shift operations. +------------------------------------------------------------------------ */ + +#ifndef __CONVERTUTF_H__ +#define __CONVERTUTF_H__ + +#if defined(_MSC_VER) + typedef unsigned __int8 uint8_t; + typedef unsigned __int16 uint16_t; + typedef unsigned __int32 uint32_t; +#else + #include +#endif + +typedef uint32_t UTF32; +typedef uint16_t UTF16; +typedef uint8_t UTF8; +typedef uint8_t Boolean; /* 0 or 1 */ + +/* Some fundamental constants */ +#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD +#define UNI_MAX_BMP (UTF32)0x0000FFFF +#define UNI_MAX_UTF16 (UTF32)0x0010FFFF +#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF +#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF + +typedef enum { + conversionOK, /* conversion successful */ + sourceExhausted, /* partial character in source, but hit end */ + targetExhausted, /* insuff. room in target for conversion */ + sourceIllegal /* source sequence is illegal/malformed */ +} ConversionResult; + +typedef enum { + strictConversion = 0, + lenientConversion +} ConversionFlags; + +/* This is for C++ and does no harm in C */ +#ifdef __cplusplus +extern "C" { +#endif + +ConversionResult ConvertUTF8toUTF16 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF16toUTF8 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF8toUTF32 ( + const UTF8** sourceStart, const UTF8* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF32toUTF8 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF16toUTF32 ( + const UTF16** sourceStart, const UTF16* sourceEnd, + UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags); + +ConversionResult ConvertUTF32toUTF16 ( + const UTF32** sourceStart, const UTF32* sourceEnd, + UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags); + +Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd); + +#ifdef __cplusplus +} +#endif + +#endif // __CONVERTUTF_H__ + +/* --------------------------------------------------------------------- */ diff --git a/desmume/src/utils/xstring.cpp b/desmume/src/utils/xstring.cpp index 291824221..eaabbcf99 100644 --- a/desmume/src/utils/xstring.cpp +++ b/desmume/src/utils/xstring.cpp @@ -2,7 +2,7 @@ //subsequently modified for desmume /* - Copyright (C) 2008-2016 DeSmuME team + Copyright (C) 2008-2009 DeSmuME team This file is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -18,14 +18,8 @@ along with the this software. If not, see . */ -//WARNING: utf8/utf16 conversion is presently UNTESTED -//requires libretro-common's HAVE_UTF8 defined -//this is all very sad. why do we need this at all? - -#include - -#include "encodings/utf.h" #include "xstring.h" +#include //a vc-style substring operation (very kind and lenient) std::string strsub(const std::string& str, int pos, int len) { @@ -289,31 +283,111 @@ std::string mass_replace(const std::string &source, const std::string &victim, c return answer; } +//http://www.codeproject.com/KB/string/UtfConverter.aspx +#include "ConvertUTF.h" +namespace UtfConverter +{ + static std::wstring FromUtf8(const std::string& utf8string) + { + size_t widesize = utf8string.length(); + if (sizeof(wchar_t) == 2) + { + wchar_t* widestringnative = new wchar_t[widesize+1]; + const UTF8* sourcestart = reinterpret_cast(utf8string.c_str()); + const UTF8* sourceend = sourcestart + widesize; + UTF16* targetstart = reinterpret_cast(widestringnative); + UTF16* targetend = targetstart + widesize+1; + ConversionResult res = ConvertUTF8toUTF16(&sourcestart, sourceend, &targetstart, targetend, strictConversion); + if (res != conversionOK) + { + delete [] widestringnative; + throw std::exception(); + } + *targetstart = 0; + std::wstring resultstring(widestringnative); + delete [] widestringnative; + return resultstring; + } + else if (sizeof(wchar_t) == 4) + { + wchar_t* widestringnative = new wchar_t[widesize+1]; + const UTF8* sourcestart = reinterpret_cast(utf8string.c_str()); + const UTF8* sourceend = sourcestart + widesize; + UTF32* targetstart = reinterpret_cast(widestringnative); + UTF32* targetend = targetstart + widesize+1; + ConversionResult res = ConvertUTF8toUTF32(&sourcestart, sourceend, &targetstart, targetend, strictConversion); + if (res != conversionOK) + { + delete [] widestringnative; + throw std::exception(); + } + *targetstart = 0; + std::wstring resultstring(widestringnative); + delete [] widestringnative; + return resultstring; + } + else + { + throw std::exception(); + } + return L""; + } + + static std::string ToUtf8(const std::wstring& widestring) + { + size_t widesize = widestring.length(); + + if (sizeof(wchar_t) == 2) + { + size_t utf8size = 3 * widesize + 1; + char* utf8stringnative = new char[utf8size]; + const UTF16* sourcestart = reinterpret_cast(widestring.c_str()); + const UTF16* sourceend = sourcestart + widesize; + UTF8* targetstart = reinterpret_cast(utf8stringnative); + UTF8* targetend = targetstart + utf8size; + ConversionResult res = ConvertUTF16toUTF8(&sourcestart, sourceend, &targetstart, targetend, strictConversion); + if (res != conversionOK) + { + delete [] utf8stringnative; + throw std::exception(); + } + *targetstart = 0; + std::string resultstring(utf8stringnative); + delete [] utf8stringnative; + return resultstring; + } + else if (sizeof(wchar_t) == 4) + { + size_t utf8size = 4 * widesize + 1; + char* utf8stringnative = new char[utf8size]; + const UTF32* sourcestart = reinterpret_cast(widestring.c_str()); + const UTF32* sourceend = sourcestart + widesize; + UTF8* targetstart = reinterpret_cast(utf8stringnative); + UTF8* targetend = targetstart + utf8size; + ConversionResult res = ConvertUTF32toUTF8(&sourcestart, sourceend, &targetstart, targetend, strictConversion); + if (res != conversionOK) + { + delete [] utf8stringnative; + throw std::exception(); + } + *targetstart = 0; + std::string resultstring(utf8stringnative); + delete [] utf8stringnative; + return resultstring; + } + else + { + throw std::exception(); + } + return ""; + } +} + //convert a std::string to std::wstring std::wstring mbstowcs(std::string str) { try { - - //BAD! - - //i wish we had 8-to-16 converting - //WAIT! better yet, 8-to-wchar converting (so we dont have `seriously` and `naive` below - size_t len = utf8len(str.c_str()); - u32 *buffer32 = new u32[len+1]; - utf8_conv_utf32(buffer32, len, str.c_str(), str.size()); - u16 *buffer16 = new u16[len+1]; - //NAIVE! - for(int i=0;i