Encoding of string returned by GetUserName()

Question

How do I get the encoding that is used for the string returned by GetUserName from the win32 API? I'm using pywin32 and it returns an 8-bit string. On my German XP, this string is obviously encoded using Latin-1, but this might not be the case for other Windows installations.

I could use GetUserNameW, but I would have to wrap that myself using ctypes, which I'd like to avoid for now if there is a simpler solution.

Answer 1

You can call GetACP to find the current ANSI codepage, which is what non-Unicode APIs use. You can also use MultiByteToWideChar, and pass zero as the codepage (CP_ACP is defined as zero in the Windows headers) to convert a codepage string to Unicode.

Answer 2

I realize this isn't answering your question directly, but I strongly recommend you go through the trouble of using the Unicode-clean GetUserNameW as you mentioned.

The non-wide commands work differently on different Windows editions (e.g. ME, although I admit that example is old!), so IMHO it's worth just getting it right.

Having done a lot of multi-lingual Windows development, although the wide API can add a layer of translation or wrapping (as you suggest!), it's worth it.

Answer 3

Okay, here's what I'm using right now:

>>> import win32api
>>> u = unicode(win32api.GetUserName(), "mbcs")
>>> type(u)
<type 'unicode'>

mbcs is a special standard encoding in Windows:

Windows only: Encode operand according to the ANSI codepage (CP_ACP)

Answer 4

From the API docs, GetUserNameA will return the name in ANSI and GetUserNameW returns the name in Unicode. You will have to use GetUserNameW.