https://gcc.gnu.org/bugzilla/show_bug.cgi?id=41041
Samuel Thibault <samuel.thiba...@ens-lyon.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution|--- |WONTFIX
--- Comment #9 from Samuel Thibault <samuel.thiba...@ens-lyon.org> ---
It seems it indeed is by default a UTF encoding rather than a UCS encoding:
$ LANG= gcc -fshort-wchar test.c -o test
$ LANG= gcc -fshort-wchar test.c -o test -fwide-exec-charset=UTF-16LE
$ LANG= gcc -fshort-wchar test.c -o test -fwide-exec-charset=UCS-2LE
test.c: In function `main':
test.c:7:27: error: converting to execution character set: Invalid or
incomplete multibyte or wide character
7 | wchar_t s[] = L"𝄞";
| ^
Now there is indeed the question of the BOM. Ideally the text could mention all
of UTF-32LE, UTF-32BE, UTF-16LE, UTF-16BE, but not sure it's really worth it.
