MutiByteToWideChar can return incorrect Unicode characters using code page 50225 (ISO-2022-KR).

Consider the following scenario. You develop an application that uses Microsoft Windows GDI+ API functions. You use the application in Windows Server 2008, in Windows Vista, in Windows Server 2003, or in Windows XP. In this scenario, when you use the application to display or to print Japanese full-width (zenkaku) characters, the Unicode hyphen character (U+2010) is not drawn.

For example, you use a hyphen (-) to separate some digits as follows:

1-2-3A

However, when the characters are displayed or printed by the application that uses GDI+ API functions, you see the following string:

123A

Note: This problem does not occur when the application uses Graphics Device Interface (GDI) API functions.

On a Windows Vista-based or a Windows Server 2008-based computer, you try to establish a connection to in order to access a local network. If the user name of the user account contains Unicode characters, the connection attempt fails the Extensible Authentication Protocol (EAP) authentication. Therefore, you cannot access the local network.

Applications that are built with a new dependency on the Visual C++ RunTime Library version 7.0, the Visual C++ RunTime Library version 8.0, or the Visual C++ RunTime Library version 9.0 (MSVCR80.dll or MSVCR90.dll) may not correctly observe regional settings. Or, these applications may not start when they are configured to use the default code page of a Unicode-only locale.

To view a list of locales that are supported in Windows, visit the following Microsoft Web site:
http://msdn2.microsoft.com/en-us/library/ms776260.aspx
The locales that are listed as Unicode only in the “ANSI code page” column do not have a corresponding ANSI code page. Some software that uses the CRT for managing locale-specific will not accept these locales. This currently applies to the following locales.

Identifier	Locale	Locale name	Script tag
0x045e 	Amharic (Ethiopia) 	am-ET	Ethi
0x042b 	Armenian (Armenia) 	hy-AM	Armn
0x044d 	Assamese (India) 	as-IN	Beng
0x0445 	Bengali (India) 	bn-IN	Beng
0x0465 	Divehi (Maldives) 	dv-MV	Thaa
0x0437 	Georgian (Georgia) 	ka-GE	Geor
0x0447 	Gujarati (India) 	gu-IN	Gujr
0x0439 	Hindi (India) 	hi-IN	Deva
0x045d 	Inuktitut (Canada) 	iu-Cans-CA	Cans
0x044b 	Kannada (India) 	kn-IN	Knda
0x0453 	Khmer (Cambodia) 	kh-KH	Khmr
0x0457 	Konkani (India) 	kok-IN	Deva
0x0454 	Lao (Lao PDR) 	lo-LA	Laoo
0x044c 	Malayalam (India) 	ml-IN	Mlym
0x044e 	Marathi (India) 	mr-IN	Deva
0x0850 	Mongolian (PRC) 	mn-Mong-CN	Mong
0x0461 	Nepali (Nepal) 	ne-NP	Deva
0x0448 	Oriya (India) 	or-IN	Orya
0x0446 	Punjabi (India) 	pa-IN	Guru
0x044f 	Sanskrit (India) 	sa-IN	Deva
0x045b 	Sinhala (Sri Lanka) 	si-LK	Sinh
0x045a 	Syriac (Syria) 	syr-SY	Syrc
0x0449 	Tamil (India) 	ta-IN	Taml
0x044a 	Telugu (India) 	te-IN	Telu
0x0851 	Tibetan (Bhutan) 	bo-BT	Tibt
0x0451 	Tibetan (PRC) 	bo-CN	Tibt
0x0478 	Yi (PRC) 	ii-CN	Yiii

This may cause the program to fail on startup, or behave unpredictably in user interfaces where localization is relevant.