C# Strings are (likely) not always UTF-8 compatible

[CryZe]

The string argument page says the following:

Native strings are automatically marshalled to C-compatible strings.

But it turns out that on Japanese Windows the strings are marshalled as Shift-JIS and not UTF-8. I'm still in the middle of figuring this bug out, but it seems like this is wrong.

[iinuwa]

The answer to this is that if the MarshalAs attribute isn't specified for a string, the string is encoded according to the system's ANSI code page. On Linux, this is always UTF-8, but on Windows, it can vary, since it uses this function.