ShuriZma
/
dolphin
mirror of https://github.com/dolphin-emu/dolphin.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use a multibyte-neutral codepage when calling *printf on Windows. 

More explanation in code comments.

Fixes issue 4046.
This commit is contained in:
Pierre Bourdon 2013-08-09 00:17:29 +02:00
parent cce809ac90
commit 9ea01aa7a8
2 changed files with 35 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Languages/po/zh_TW.po
View File

@ -5503,7 +5503,7 @@ msgstr "您必須輸入一個有效的設定檔名稱。"
#: Source/Core/DolphinWX/Src/ConfigMain.cpp:917
msgid "You must restart Dolphin in order for the change to take effect."
msgstr "You must restart Dolphin in order for the change to take effect."
msgstr "您必須重新啟動 Dolphin 使更改生效。"
#: Source/Core/Core/Src/DSP/DSPCore.cpp:109
msgid ""

37
Source/Core/Common/Src/StringUtil.cpp
View File

@ -34,7 +34,39 @@ bool AsciiToHex(const char* _szValue, u32& result)
bool CharArrayFromFormatV(char* out, int outsize, const char* format, va_list args)
{
int writtenCount = vsnprintf(out, outsize, format, args);
int writtenCount;
#ifdef _WIN32
// You would think *printf are simple, right? Iterate on each character,
// if it's a format specifier handle it properly, etc.
//
// Nooooo. Not according to the C standard.
//
// According to the C99 standard (7.19.6.1 "The fprintf function")
// The format shall be a multibyte character sequence
//
// Because some character encodings might have '%' signs in the middle of
// a multibyte sequence (SJIS for example only specifies that the first
// byte of a 2 byte sequence is "high", the second byte can be anything),
// printf functions have to decode the multibyte sequences and try their
// best to not screw up.
//
// Unfortunately, on Windows, the locale for most languages is not UTF-8
// as we would need. Notably, for zh_TW, Windows chooses EUC-CN as the
// locale, and completely fails when trying to decode UTF-8 as EUC-CN.
//
// On the other hand, the fix is simple: because we use UTF-8, no such
// multibyte handling is required as we can simply assume that no '%' char
// will be present in the middle of a multibyte sequence.
//
// This is why we lookup an ANSI (cp1252) locale here and use _vsnprintf_l.
static locale_t c_locale = NULL;
if (!c_locale)
c_locale = _create_locale(LC_ALL, ".1252");
writtenCount = _vsnprintf_l(out, outsize, format, c_locale, args);
#else
writtenCount = vsnprintf(out, outsize, format, args);
#endif
if (writtenCount > 0 && writtenCount < outsize)
{
@ -58,10 +90,9 @@ std::string StringFromFormat(const char* format, ...)
va_start(args, format);
required = _vscprintf(format, args);
buf = new char[required + 1];
vsnprintf(buf, required, format, args);
CharArrayFromFormatV(buf, required + 1, format, args);
va_end(args);
buf[required] = '\0';
std::string temp = buf;
delete[] buf;
#else