Not the answer to your question, but if you don't need to perform extensive operations at the file system level (e.g. search, sort ...), there is a good cross-platform workaround for the problem described in this SO question : URLEncode() file names .
Hörensägen.txt
turns into
H%c3%b6rens%c3%a4gen.txt
which should be safe to use on any file system and capable of displaying any UTF-8 character.
I find it much preferable to try “natively” to deal with host OS capabilities, which is guaranteed to be complex and error prone (in addition to differences in the operating system, I am sure that different file system formats are FAT16, FAT32, NTFS, extFS versions 1/2/3 .... bring your own set of rules.)
Pekka 웃
source share