Click here to Skip to main content
15,885,366 members
Articles / Programming Languages / C++

Convert Between std::string and std::wstring, UTF-8 and UTF-16

Rate me:
Please Sign up or sign in to vote.
3.60/5 (16 votes)
20 May 2007CPOL1 min read 330.8K   2.8K   63   43
How to convert safely STL strings between Unicode formats

Introduction

I needed to convert between UTF-8 coded std::string and UTF-16 coded std::wstring. I found some converting functions for native C strings, but these leave the memory handling to the caller. Not nice in modern times.

The best converter is probably the one from unicode.org. Here is a wrapper around this one which converts the STL strings.

Unlike other articles, this one has no other dependencies, does not introduce yet another string class, only converts the STL strings, and that's it. And it's better than the widely found...

C++
std::wstring widestring(sourcestring.begin(), sourcestring.end()); 

... which only works for ASCII text.

Source

The header goes like this:

C++
#ifndef UTFCONVERTER__H__
#define UTFCONVERTER__H__

namespace UtfConverter
{
    std::wstring FromUtf8(const std::string& utf8string);
    std::string ToUtf8(const std::wstring& widestring);
}

#endif

I guess this is simple and easy enough to use.

Here is the source code:

C++
#include "stdafx.h"
#include "UtfConverter.h"
#include "ConvertUTF.h"

namespace UtfConverter
{
    std::wstring FromUtf8(const std::string& utf8string)
    {
        size_t widesize = utf8string.length();
        if (sizeof(wchar_t) == 2)
        {
            wchar_t* widestringnative = new wchar_t[widesize+1];
            const UTF8* sourcestart = reinterpret_cast<const UTF8*>(utf8string.c_str());
            const UTF8* sourceend = sourcestart + widesize;
            UTF16* targetstart = reinterpret_cast<UTF16*>(widestringnative);
            UTF16* targetend = targetstart + widesize+1;
            ConversionResult res = ConvertUTF8toUTF16
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                delete [] widestringnative;
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            std::wstring resultstring(widestringnative);
            delete [] widestringnative;
            return resultstring;
        }
        else if (sizeof(wchar_t) == 4)
        {
            wchar_t* widestringnative = new wchar_t[widesize];
            const UTF8* sourcestart = reinterpret_cast<const UTF8*>(utf8string.c_str());
            const UTF8* sourceend = sourcestart + widesize;
            UTF32* targetstart = reinterpret_cast<UTF32*>(widestringnative);
            UTF32* targetend = targetstart + widesize;
            ConversionResult res = ConvertUTF8toUTF32
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                delete [] widestringnative;
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            std::wstring resultstring(widestringnative);
            delete [] widestringnative;
            return resultstring;
        }
        else
        {
            throw std::exception("La falla!");
        }
        return L"";
    }

    std::string ToUtf8(const std::wstring& widestring)
    {
        size_t widesize = widestring.length();

        if (sizeof(wchar_t) == 2)
        {
            size_t utf8size = 3 * widesize + 1;
            char* utf8stringnative = new char[utf8size];
            const UTF16* sourcestart = 
		reinterpret_cast<const UTF16*>(widestring.c_str());
            const UTF16* sourceend = sourcestart + widesize;
            UTF8* targetstart = reinterpret_cast<UTF8*>(utf8stringnative);
            UTF8* targetend = targetstart + utf8size;
            ConversionResult res = ConvertUTF16toUTF8
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                delete [] utf8stringnative;
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            std::string resultstring(utf8stringnative);
            delete [] utf8stringnative;
            return resultstring;
        }
        else if (sizeof(wchar_t) == 4)
        {
            size_t utf8size = 4 * widesize + 1;
            char* utf8stringnative = new char[utf8size];
            const UTF32* sourcestart = 
		reinterpret_cast<const UTF32*>(widestring.c_str());
            const UTF32* sourceend = sourcestart + widesize;
            UTF8* targetstart = reinterpret_cast<UTF8*>(utf8stringnative);
            UTF8* targetend = targetstart + utf8size;
            ConversionResult res = ConvertUTF32toUTF8
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                delete [] utf8stringnative;
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            std::string resultstring(utf8stringnative);
            delete [] utf8stringnative;
            return resultstring;
        }
        else
        {
            throw std::exception("La falla!");
        }
        return "";
    }
} 

How To Do It Better

Here's another version that avoids using new and delete, by writing directly into the string buffer. Does anyone know whether this is okay?

C++
#include "stdafx.h"
#include "UtfConverter.h"
#include "ConvertUTF.h"

namespace UtfConverter
{
    std::wstring FromUtf8(const std::string& utf8string)
    {
        size_t widesize = utf8string.length();
        if (sizeof(wchar_t) == 2)
        {
            std::wstring resultstring;
            resultstring.resize(widesize+1, L'\0');
            const UTF8* sourcestart = reinterpret_cast<const UTF8*>(utf8string.c_str());
            const UTF8* sourceend = sourcestart + widesize;
            UTF16* targetstart = reinterpret_cast<UTF16*>(&resultstring[0]);
            UTF16* targetend = targetstart + widesize;
            ConversionResult res = ConvertUTF8toUTF16
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            return resultstring;
        }
        else if (sizeof(wchar_t) == 4)
        {
            std::wstring resultstring;
            resultstring.resize(widesize+1, L'\0');
            const UTF8* sourcestart = reinterpret_cast<const UTF8*>(utf8string.c_str());
            const UTF8* sourceend = sourcestart + widesize;
            UTF32* targetstart = reinterpret_cast<UTF32*>(&resultstring[0]);
            UTF32* targetend = targetstart + widesize;
            ConversionResult res = ConvertUTF8toUTF32
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            return resultstring;
        }
        else
        {
            throw std::exception("La falla!");
        }
        return L"";
    }

    std::string ToUtf8(const std::wstring& widestring)
    {
        size_t widesize = widestring.length();

        if (sizeof(wchar_t) == 2)
        {
            size_t utf8size = 3 * widesize + 1;
            std::string resultstring;
            resultstring.resize(utf8size, '\0');
            const UTF16* sourcestart = 
		reinterpret_cast<const UTF16*>(widestring.c_str());
            const UTF16* sourceend = sourcestart + widesize;
            UTF8* targetstart = reinterpret_cast<UTF8*>(&resultstring[0]);
            UTF8* targetend = targetstart + utf8size;
            ConversionResult res = ConvertUTF16toUTF8
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            return resultstring;
        }
        else if (sizeof(wchar_t) == 4)
        {
            size_t utf8size = 4 * widesize + 1;
            std::string resultstring;
            resultstring.resize(utf8size, '\0');
            const UTF32* sourcestart = 
		reinterpret_cast<const UTF32*>(widestring.c_str());
            const UTF32* sourceend = sourcestart + widesize;
            UTF8* targetstart = reinterpret_cast<UTF8*>(&resultstring[0]);
            UTF8* targetend = targetstart + utf8size;
            ConversionResult res = ConvertUTF32toUTF8
		(&sourcestart, sourceend, &targetstart, targetend, strictConversion);
            if (res != conversionOK)
            {
                throw std::exception("La falla!");
            }
            *targetstart = 0;
            return resultstring;
        }
        else
        {
            throw std::exception("La falla!");
        }
        return "";
    }
}

How to Use It

Just add it to your project. Download the Unicode converter from here and add that to the project, too. It should just work.

Of course, you can throw whatever exceptions you like upon failure.

I must admit I tried it only for 2-byte wchar_t.

Comments are welcome.

License

This article, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)


Written By
rh_
Germany Germany
This member has not yet provided a Biography. Assume it's interesting and varied, and probably something to do with programming.

Comments and Discussions

 
QuestionBug in the code? Pin
_ema_10-May-07 13:51
_ema_10-May-07 13:51 
AnswerRe: Bug in the code? Pin
rh_20-May-07 20:32
rh_20-May-07 20:32 
GeneralSimplest way Pin
tracker200213-Feb-07 4:06
tracker200213-Feb-07 4:06 
GeneralRe: Simplest way Pin
rh_13-Feb-07 20:04
rh_13-Feb-07 20:04 
GeneralUTF-8 and Multibyte are not the same Pin
Ted Dunlop12-Feb-07 5:59
Ted Dunlop12-Feb-07 5:59 
GeneralRe: UTF-8 and Multibyte are not the same Pin
rh_13-Feb-07 0:21
rh_13-Feb-07 0:21 
GeneralMultiByteToWideChar and WideCharToMultiByte... Pin
Johann Gerell11-Feb-07 19:27
Johann Gerell11-Feb-07 19:27 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
rh_11-Feb-07 22:37
rh_11-Feb-07 22:37 
Well, yes and no. MultiByteToWideChar and WideCharToMultiByte also need a buffer with a fixed size which must be provided before.
Certainly you can replace ConvertUTF16toUTF8 by WideCharToMultiByte etc. in my code, such that you don't need to download the stuff from unicode.org.

NewsRe: MultiByteToWideChar and WideCharToMultiByte... Pin
Johann Gerell11-Feb-07 22:54
Johann Gerell11-Feb-07 22:54 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
rh_12-Feb-07 1:32
rh_12-Feb-07 1:32 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
Johann Gerell12-Feb-07 1:49
Johann Gerell12-Feb-07 1:49 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
rh_12-Feb-07 2:34
rh_12-Feb-07 2:34 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
Bjornar19-Feb-07 3:44
Bjornar19-Feb-07 3:44 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
konmel7-Nov-07 6:53
konmel7-Nov-07 6:53 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
Dmytro Skrypnyk7-Aug-08 11:25
Dmytro Skrypnyk7-Aug-08 11:25 
GeneralRe: MultiByteToWideChar and WideCharToMultiByte... Pin
johnny longxzy26-Jun-09 11:33
johnny longxzy26-Jun-09 11:33 
AnswerRe: MultiByteToWideChar and WideCharToMultiByte... Pin
Johann Gerell26-Jun-09 12:34
Johann Gerell26-Jun-09 12:34 

General General    News News    Suggestion Suggestion    Question Question    Bug Bug    Answer Answer    Joke Joke    Praise Praise    Rant Rant    Admin Admin   

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.