Windows use UNICODE (wchar_t in utf16), this PR add a better narrow<->wide converter only for Windows as utf16 is useless on other OS. I added special WindowsError exception too to be able to get the string corresponding to GetLastError() (converted to utf8 of course). This will allow to have better exceptions on windows without too much trouble. I expect to have one for errno too
Windows
use UNICODE (wchar_t
inutf16
), this PR add a better narrow<->wide converter only for Windows as utf16 is useless on other OS. I added specialWindowsError
exception too to be able to get the string corresponding toGetLastError()
(converted to utf8 of course). This will allow to have better exceptions on windows without too much trouble. I expect to have one forerrno
too