2010년 1월 19일 화요일

UTF-8 to ANSI Char* 변환


#include "stdio.h"
#include "stdlib.h"
#include "conio.h"
#include "windows.h"
static wchar_t *make_unicode_string(const unsigned char *utf8)
{
int size = 0, index = 0, out_index = 0;
wchar_t *out;
unsigned char c;
/* first calculate the size of the target string */
c = utf8[index++];
while(c) {
if((c & 0x80) == 0) {
index += 0;
} else if((c & 0xe0) == 0xe0) {
index += 2;
} else {
index += 1;
}
size += 1;
c = utf8[index++];
}  

out = (wchar_t*)malloc((size + 1) * sizeof(wchar_t));
if (out == NULL)
return NULL;
index = 0;

c = utf8[index++];
while(c)
{
if((c & 0x80) == 0) {
out[out_index++] = c;
} else if((c & 0xe0) == 0xe0) {
out[out_index] = (c & 0x1F) << 12;
            c = utf8[index++];
            out[out_index] |= (c & 0x3F) << 6;
            c = utf8[index++];
            out[out_index++] |= (c & 0x3F);
        } else {
            out[out_index] = (c & 0x3F) << 6;
            c = utf8[index++];
            out[out_index++] |= (c & 0x3F);
        }
        c = utf8[index++];
    }
    out[out_index] = 0;

    return out;
}


int utf8_decode(const char *from, char **to)
{
    wchar_t *unicode;
    int chars, err;
    /* On NT-based windows systems, we could use MultiByteToWideChar(CP_UTF8), but
     * MS doesn't actually have a consistent API across win32.
     */
    unicode = make_unicode_string((const unsigned char*)from);
    if(unicode == NULL)
    {
        fprintf(stderr, "Out of memory processing string from UTF8 to UNICODE16\n");
        return -1;
    }

    chars = WideCharToMultiByte(GetConsoleCP(), WC_COMPOSITECHECK, unicode,
            -1, NULL, 0, NULL, NULL);

    if(chars == 0)
    {
        fprintf(stderr, "Unicode translation error %d\n", GetLastError());
        free(unicode);
        return -1;
    }

    *to = (char*)calloc(chars + 1, sizeof(unsigned char));
    if(*to == NULL)
    {
        fprintf(stderr, "Out of memory processing string to local charset\n");
        free(unicode);
        return -1;
    }

    err = WideCharToMultiByte(GetConsoleCP(), WC_COMPOSITECHECK, unicode,
            -1, *to, chars, NULL, NULL);
    if(err != chars)
    {
        fprintf(stderr, "Unicode translation error %d\n", GetLastError());
        free(unicode);
        free(*to);
        *to = NULL;
        return -1;
    }
    free(unicode);
    return 0;
}

char* UTF8ToANSI(char *pszCode)
{
    BSTR    bstrWide;
    char*   pszAnsi;
    int     nLength;

    // Get nLength of the Wide Char buffer
    nLength = MultiByteToWideChar(CP_UTF8, 0, pszCode, lstrlen(pszCode) + 1,
                                  NULL, NULL);

    bstrWide = SysAllocStringLen(NULL, nLength);

   // Change UTF-8 to Unicode (UTF-16)
    MultiByteToWideChar(CP_UTF8, 0, pszCode, lstrlen(pszCode) + 1, bstrWide,
                       nLength);

    // Get nLength of the multi byte buffer
    nLength = WideCharToMultiByte(CP_ACP, 0, bstrWide, -1, NULL, 0, NULL, NULL);

    pszAnsi = new char[nLength];

    // Change from unicode to mult byte
    WideCharToMultiByte(CP_ACP, 0, bstrWide, -1, pszAnsi, nLength, NULL, NULL);
    SysFreeString(bstrWide);

    return pszAnsi;
}

main()
{
    FILE *fp;
    char str[80];
    char *str2;
    fp=fopen("data.txt","r");
    fscanf(fp,"%[^\n]",str);
//    utf8_decode(str,&str2);
    str2=UTF8ToANSI(str);
    printf("%s\n",str2);
    getch();
}


Link with oleaut32.lib.

[출처] UTF8 --> char* 변환|작성자 난뽀다

댓글 없음:

댓글 쓰기