in linux & c/c++: how do I convert an ascii string to utf8 & vice versa?

davidh_uk · 02-04-2005, 02:37 PM

I'm using the libxml gnome libraries to try and convert ascii strings to utf8 and vice versa but I cant get it to work. Here's the test code;

#include <assert.h>
#include <getopt.h>
#include <glib-2.0/glib.h> //for all that cool stuff..
#include <exception>
#include <iostream>
#include <getopt.h>

using namespace std;

wstring makewidestring(string narrowstr)
{
int n;
wstring _widestr;
n = strlen(narrowstr.c_str()) * sizeof(wchar_t);
; //make string wide
if (mbstowcs((wchar_t *)_widestr.c_str(), narrowstr.c_str(), n)<=0)
throw "Regex::Regex failed to make match wide target string ";
wcout <<"stub1"<<_widestr <<":"<<endl;
return wstring(_widestr);
};
//--
int main(int argc, char **argv)
{
string nstr;
wstring wstr;

nstr = "narrow string";
wstr = makewidestring(nstr);
cout <<"narrow string:'" << nstr <<"'<-" << endl;
wcout <<"wide string:'" << wstr <<"'<-" << endl;
}

Can anyone hit me with the clue stick? is mbstowcs the function to use? hows it done?

TIA

jlliagre · 02-06-2005, 01:39 AM

You'd better off with iconv to convert from character sets.

Here is some sample code, (edited while under ISO-8859-1 encoding):

Code:

#include <iconv.h>

main()
{
  iconv_t it;
  size_t il=16;
  char *ibuf=calloc(il, 1);
  size_t ol=64;
  char *obuf=malloc(64);
  char *toCode="UTF8";
  char *fromCode="ISO-8859-1";

  strncpy(ibuf, "çà va où ?", il);
  it=iconv_open(toCode, fromCode);
  if(it!=(iconv_t)-1)
  {
    size_t c;
    char *source=ibuf;
    char *result=obuf;
    il=strlen(ibuf);
    if(iconv(it, (const char **)&ibuf, &il, &obuf, &ol)!=-1)
    {
      printf("%s -> %s\n", source, result);
    }
    else
    {
      perror("iconv");
    }
    iconv_close(it);
  }
  else
  {
    perror("iconv_open");
  }
}

Its output:

Code:

çà va où ? -> Ã§Ã  va oÃ¹ ?

davidh_uk · 02-06-2005, 05:55 PM

jlliagre
That was a real eye opener - I hadn't appreciated how to use iconv, and its advantages over mbswhatever. Based on your post I Googled & found this documentation:
http://www.lrz-muenchen.de/services/...c_6.html#SEC99
This is what I'll use instead.
Thanks