使用C代码在Windows上使用波兰语字符读取文件名

Question

I must note that developing code for windows platform is a mystery for me at the moment.

I have an issue with going through a directory and reading up files, where polish characters are present. Following code snippets work on windows in english and polish version, but not in german. What's the reason? I've tried with POSIX opendir() / readdir() and with winapi FindFirstFile() / FindNextFile() functions. They gave same results.
Consider following program. It reads up files ( in POSIX and winapi way ) within test directory and prints them out.

#include <stdio.h>
#include <stdlib.h>
#include <windows.h>
#include <dirent.h>

int main(void)
{
    char* name;
    name = _getcwd(NULL, 0);

    char* tmp = realloc(name, strlen(name)*2);
    if (tmp != NULL) {
        name = tmp;
    }
    else {
        perror("realloc");
        free(name);
        exit(1);
    }

    strcat(name, "\\test");
    DIR* dir = opendir(name);
    struct dirent* s;
    while ((s = readdir(dir)) != NULL) {
        printf("%s\n", s->d_name);
    }
    closedir(dir);
    free(s);

    strcat(name, "\\*");
    HANDLE dir_w;
    WIN32_FIND_DATA s_w;
    if ((dir_w = FindFirstFile(name, &s_w)) != INVALID_HANDLE_VALUE) {
        while (FindNextFile(dir_w, &s_w) != 0) {
            printf("%s\n", s_w.cFileName);
        }
    }
    FindClose(dir_w);

    free(name);

    return 0;
}

ISSUE : on windows with german language polish characters are interpreted as ASCII characters. If, for example, file has name:

ąęćś.txt

then on english language windows it is read up correctly, but on german one is read as

aecs.txt

but such file does not exist. Should I go with wchar_t type and functions specific to it (FindNextFilew) ? I don't have german windows so it is hard for me to reproduce the error. I've been trying to change code page with command chcp 1250 but that didn't help. Setting locale to pl also did not help.

Compiler: gcc 5.0.0

Answer 1

The C standard library functions you call will be using ANSI encoded text which as I'm sure you know cannot handle international text.

So you'll want to use the native Unicode Windows APIs rather than the legacy ANSI versions. That means wchar_t based functions with W prefixes like FindFirstFileW . For a start you should enable the Unicode conditionals so that the FindFirstFile macro expands to FindFirstFileW .