Creating a new file with multibyte japanese string in C

I am trying to create a new file with name "履歴書.txt" using C language's fopen(). I am using Windows 10 English version.

File gets created but file name is always garbled.

Code:

---

setlocale(LC_ALL, "ja_JP.utf8");
FILE *outputfile;
outputfile = fopen("C:/sources/履歴書.txt", "w, ccs=UTF-8");
if (outputfile == NULL) {
                printf("cannot open\n");
                exit(1);
        }
fprintf(outputfile, "this is test.\n");
fclose(outputfile);

---

How can I create file with exact name "履歴書.txt" in C language?

Quote:

---

Originally Posted by Mithlesh10

I am trying to create a new file with name "履歴書.txt" using C language's fopen(). I am using Windows 10 English version.

File gets created but file name is always garbled.

Code:

---

setlocale(LC_ALL, "ja_JP.utf8");
FILE *outputfile;
outputfile = fopen("C:/sources/履歴書.txt", "w, ccs=UTF-8");
if (outputfile == NULL) {
        printf("cannot open\n");
        exit(1);
    }
fprintf(outputfile, "this is test.\n");
fclose(outputfile);

---

How can I create file with exact name "履歴書.txt" in C language?

---

C's fopen takes a char *. This is a sequence of 8-bit bytes designed to hold Latin characters. You might be able to pass it UTF-8, which encodes ASCII as 8 bits and Japanese as multibytes, but it probably won't work.
So you will have to use a function with a name like _wfopen(), which takes wide characters (16 bit).

Quote:

---

Originally Posted by Malcolm McLean

C's fopen takes a char *. This is a sequence of 8-bit bytes designed to hold Latin characters. You might be able to pass it UTF-8, which encodes ASCII as 8 bits and Japanese as multibytes, but it probably won't work.
So you will have to use a function with a name like _wfopen(), which takes wide characters (16 bit).

---

Thanks for your reply.

I think I should have mentioned it while asking question, but to keep it simple I didn't.

FYI, _wfopen() also doesn't work. I have already tried like this :

Code:

---

_wfopen(L"C:/sources/履歴書.txt", L"w,ccs=UTF-8");

---

Quote:

---

Originally Posted by Malcolm McLean

C's fopen takes a char *. This is a sequence of 8-bit bytes designed to hold Latin characters.

---

ISO 9899 accepts limited sets of multibyte charsets like UNICODE, as defined in Annex D of the standard. Including using multibyte charset with identifiers (thou some compilers don't respect this - clang and GCC 10 does! But not GCC 9 or below and maybe not Visual Studio's C++).

Quote:

---

You might be able to pass it UTF-8, which encodes ASCII as 8 bits and Japanese as multibytes, but it probably won't work.

---

It works just fine:

Attachment 16218

Mithiesh10 problem is because Windows has 2 charsets: One for single byte chars (Windows 1252?), other for multibyte chars (UTF-16 or WideChar), it doesn't recognize UTF-8. This is limitatio of Windows, not C.

In that aspect you are right: For Windows you must use wchar_t and w* functions to use UNICODE. But again, it is not a limitation of C.

Quote:

---

Originally Posted by Mithlesh10

Thanks for your reply.

I think I should have mentioned it while asking question, but to keep it simple I didn't.

FYI, _wfopen() also doesn't work. I have already tried like this :

Code:

---

_wfopen(L"C:/sources/履歴書.txt", L"w,ccs=UTF-8");

---

---

Here worked fine. The problem, it seems, is the Windows terminal don't recognize UNICODE:

Attachment 16219

Didn't get your attachment. Please reattach or share code.

Sorry... it is not a code, but an image of file explorer, showing the name correctly.
I'am getting ???.txt on terminal, but on file explorer is correct.

The test code is simply:

Code:

---

if ( ! ( f = _wfopen( L"履歴書.txt", L"w,ccs=UTF-8" ) ) )
{ perror( "_wfopen" ); exit(1); }
fclose( f );

---

Thanks for the code. I also have exact same code and it runs fine. But file created has garbled name. To be precise it becomes - å±¥æ..´æ›¸.txt

Which compiler are you using?

If you are willing for a somewhat hacky solution, try this:

Code:

---

#include <stdio.h>


int main(int argc, char *argv[]) {
  // UTF-8 coding for unicode filename
  char filename[] = {0xe5, 0xb1, 0xa5, 0xe6, 0xad, 0xb4, 0xe6, 0x9b, 0xb8, '.', 't', 'x', 't', 0 };
  FILE *f = fopen(filename,"wb");
  if(f) fclose(f);
}

---

Quote:

---

Originally Posted by hamster_nz

If you are willing for a somewhat hacky solution, try this:

---

Unfortunately, this doesn't work on Windows. Here's the same code, using fopen() (test.exe) and _wfopen() (test2.exe). Notice that in file explorer the created flie is shown with the "correct" name when using _wfopen(), but in the terminal is ???.txt. And, when using fopen() the individual bytes of UTF-8 chars ara used as if they were in Windows-1252 codepage (the '.txt' portion isn't there because file explorer hides the extension, by default):

Attachment 16220

PS: On Windows you need encode your source code in UTF16LE if using multibyte charsets.