C调用c#的托管字符串传送UTF8字符集方法

最新推荐文章于 2025-07-23 12:22:43 发布

原创最新推荐文章于 2025-07-23 12:22:43 发布 · 2.3k 阅读

1 ·

CC 4.0 BY-SA版权

工具--效率同时被 2 个专栏收录

129 篇文章

订阅专栏

C/C++

74 篇文章

订阅专栏

本文介绍了一种在C语言中实现UTF8编码与Unicode编码相互转换的方法。通过具体的代码示例，展示了如何处理不同长度的UTF8字符，并将其转换为Unicode编码，反之亦然。此外，还指出了在处理ASCII码时可能出现的问题并提供了修正方案。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

//将非托管的ANSI字符串转换成托管字符串
String^ UnmanagedStringA2ManagedString(char *pIn)
{
String^ strOut =System::Runtime::InteropServices::Marshal::PtrToStringAnsi(static_cast<IntPtr>(pIn));
return strOut;
}

这种方式传送的是Ansi编码的字符串，如果传入UTF8格式的字符串，在调用后会出现乱码

系统提供的还有PtrToStringUni等方法，PtrToStringUni接收unicode编码的字符串

但是没有提供处理UTF8字符集的方法，

这个时候，需要把UTF8转换成unicode编码

下面是一个网上的例子，有个地方有问题，处理ascii码会出现乱码，需要修改

/************************************************************************/
/* Author: NadOo (nadoo@21cn.com)
    Blog:
            http://nadoo.cnblogs.com
    Reference:
            http://blog.youkuaiyun.com/lovekatherine/archive/2007/08/30/1765903.aspx
    Testing Tool:
            http://www.hanzify.org/?Go=Show::List&ID=9627
*/
/************************************************************************/
#include <stdio.h>
#include <string.h>
#include <malloc.h>
#include <memory.h>

#ifdef WIN32
#define uint8_t unsigned __int8
#define uint16_t unsigned __int16
#define uint32_t unsigned __int32
#define uint64_t unsigned __int64

#define int8_t __int8
#define int16_t __int16
#define int32_t __int32
#endif

int unicode_to_utf8(uint16_t *in, int insize, uint8_t **out)
{
    int i = 0;
    int outsize = 0;
    int charscount = 0;
    uint8_t *result = NULL;
    uint8_t *tmp = NULL;

    charscount = insize / sizeof(uint16_t);
    result = (uint8_t *)malloc(charscount * 3 + 1);
    memset(result, 0, charscount * 3 + 1);
    tmp = result;

    for (i = 0; i < charscount; i++)
    {
        uint16_t unicode = in[i];

        if (unicode >= 0x0000 && unicode <= 0x007f)
        {
            *tmp = (uint8_t)unicode;
            tmp += 1;
            outsize += 1;
        }
        else if (unicode >= 0x0080 && unicode <= 0x07ff)
        {
            *tmp = 0xc0 | (unicode >> 6);
            tmp += 1;
            *tmp = 0x80 | (unicode & (0xff >> 2));
            tmp += 1;
            outsize += 2;
        }
        else if (unicode >= 0x0800 && unicode <= 0xffff)
        {
            *tmp = 0xe0 | (unicode >> 12);
            tmp += 1;
            *tmp = 0x80 | (unicode >> 6 & 0x00ff);
            tmp += 1;
            *tmp = 0x80 | (unicode & (0xff >> 2));
            tmp += 1;
            outsize += 3;
        }

    }

    *tmp = '\0';
    *out = result;
    return 0;
}

int utf8_to_unicode(uint8_t *in, uint16_t **out, int *outsize)
{
    uint8_t *p = in;
    uint16_t *result = NULL;
    int resultsize = 0;
    uint8_t *tmp = NULL;

    result = (uint16_t *)malloc(strlen(in) * 2 + 2); /* should be enough */
    memset(result, 0, strlen(in) * 2 + 2);
    tmp = (uint8_t *)result;

    while(*p)
    {
        if (*p >= 0x00 && *p <= 0x7f)
        {
            *tmp = *p;
            tmp++;
            *tmp = '\0';

//必须这样修改 ==========================================================================================

tmp++;

//=====================================================================================================
            resultsize += 2;
        }
        else if ((*p & (0xff << 5))== 0xc0)
        {
            uint16_t t = 0;
            uint8_t t1 = 0;
            uint8_t t2 = 0;

            t1 = *p & (0xff >> 3);
            p++;
            t2 = *p & (0xff >> 2);

            *tmp = t2 | ((t1 & (0xff >> 6)) << 6);//t1 >> 2;
            tmp++;

            *tmp = t1 >> 2;//t2 | ((t1 & (0xff >> 6)) << 6);
            tmp++;

            resultsize += 2;
        }
        else if ((*p & (0xff << 4))== 0xe0)
        {
            uint16_t t = 0;
            uint8_t t1 = 0;
            uint8_t t2 = 0;
            uint8_t t3 = 0;

            t1 = *p & (0xff >> 3);
            p++;
            t2 = *p & (0xff >> 2);
            p++;
            t3 = *p & (0xff >> 2);

            //Little Endian
            *tmp = ((t2 & (0xff >> 6)) << 6) | t3;//(t1 << 4) | (t2 >> 2);
            tmp++;

            *tmp = (t1 << 4) | (t2 >> 2);//((t2 & (0xff >> 6)) << 6) | t3;
            tmp++;
            resultsize += 2;
        }

        p++;
    }

    *tmp = '\0';
    tmp++;
    *tmp = '\0';
    resultsize += 2;

    *out = result;
    *outsize = resultsize;
    return 0;
}

void dump_utf8(uint8_t *utf8)
{
    uint8_t *p = utf8;

    while(*p)
    {
        printf("%02X", *p);
        p++;
    }
    putchar('\n');
}

void dump_unicode(uint16_t *utf16, int size)
{
    uint8_t *p = (uint8_t *)utf16;
    int i = 0;

    for (i = 0; i < size; i++)
    {
        printf("%02X", *p);
        p++;
    }
    putchar('\n');
}

int main()
{
    uint16_t unicode[] = L"中文";
    uint8_t *utf8 = NULL;

    int unisize = 0;
    uint16_t *uni = NULL;

    printf("original unicode: \n");
    dump_unicode(unicode, sizeof(unicode));

    printf("converted to utf8: \n");
    unicode_to_utf8(unicode, sizeof(unicode), &utf8);
    dump_utf8(utf8);

    printf("converted to unicode: \n");
    utf8_to_unicode(utf8, &uni, &unisize);
    dump_unicode(uni, unisize);

    free(utf8);
    free(uni);

    return 0;
}