/* toutf8.c	Convert strings from system locale into UTF-8.
 * Copyright (C) 2002, 2003  Simon Josefsson
 *
 * This file is part of GNU Libidn.
 *
 * GNU Libidn is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * GNU Libidn is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with GNU Libidn; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 *
 */

#if HAVE_CONFIG_H
# include "config.h"
#endif

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "idna/stringprep.h"

#ifdef _LIBC
# define HAVE_ICONV 1
# define LOCALE_WORKS 1
# define ICONV_CONST
#endif

#if defined(HAVE_ERRNO_H) || defined(_LIBC)
# include <errno.h>
#endif

#ifdef HAVE_ICONV
# include <iconv.h>

# if LOCALE_WORKS
#  include <langinfo.h>
#  include <locale.h>
# endif

static const char *
stringprep_locale_charset_slow (void)
{
  const char *charset = getenv ("CHARSET");	/* flawfinder: ignore */

  if (charset && *charset)
    return charset;

# ifdef LOCALE_WORKS
  {
    char *p;

    p = setlocale (LC_CTYPE, NULL);
    setlocale (LC_CTYPE, "");

    charset = nl_langinfo (CODESET);

    setlocale (LC_CTYPE, p);

    if (charset && *charset)
      return charset;
  }
# endif

  return "ASCII";
}

static const char *stringprep_locale_charset_cache = NULL;

/**
 * stringprep_locale_charset:
 *
 * Find out system locale charset.
 *
 * Note that this function return what it believe the SYSTEM is using
 * as a locale, not what locale the program is currently in (modified,
 * e.g., by a setlocale(LC_CTYPE, "ISO-8859-1")).  The reason is that
 * data read from argv[], stdin etc comes from the system, and is more
 * likely to be encoded using the system locale than the program
 * locale.
 *
 * You can set the environment variable CHARSET to override the value
 * returned.  Note that this function caches the result, so you will
 * have to modify CHARSET before calling (even indirectly) any
 * stringprep functions, e.g., by setting it when invoking the
 * application.
 *
 * Return value: Return the character set used by the system locale.
 *   It will never return NULL, but use "ASCII" as a fallback.
 **/
const char *
stringprep_locale_charset (void)
{
  if (!stringprep_locale_charset_cache)
    stringprep_locale_charset_cache = stringprep_locale_charset_slow ();

  return stringprep_locale_charset_cache;
}

/**
 * stringprep_convert:
 * @str: input zero-terminated string.
 * @to_codeset: name of destination character set.
 * @from_codeset: name of origin character set, as used by @str.
 *
 * Convert the string from one character set to another using the
 * system's iconv() function.
 *
 * Return value: Returns newly allocated zero-terminated string which
 *   is @str transcoded into to_codeset.
 **/
char *
stringprep_convert (const char *str,
		    const char *to_codeset, const char *from_codeset,
                    int best_effort)
{
  iconv_t cd;
  char *dest;
  char *outp;
  char *p, *startp;
  size_t inbytes_remaining;
  size_t outbytes_remaining;
  size_t err;
  size_t outbuf_size;
  int have_error = 0;
  int from_utf8;
  int len;

  if (strcmp (to_codeset, from_codeset) == 0)
    {
      char *p;
      p = malloc (strlen (str) + 1);
      if (!p)
	return NULL;
      strcpy (p, str);
      return p;
    }

  from_utf8 = (strcmp (from_codeset, "UTF-8") == 0);

#ifdef ICONV_TRANSLIT
  if (best_effort)
    {
      char to_c[strlen (to_codeset) + 10];
      strcpy (to_c, to_codeset);
      strcat (to_c, "//TRANSLIT");
      cd = iconv_open (to_c, from_codeset);
    }
  else
    cd = iconv_open (to_codeset, from_codeset);
#else
  cd = iconv_open (to_codeset, from_codeset);
#endif

  if (cd == (iconv_t) - 1)
    return NULL;

  p = (char *) malloc (strlen (str) + 1);
  if (p == NULL)
    return NULL;
  strcpy (p, str);
  len = strlen (p);
  startp = p;
  inbytes_remaining = len;
  outbuf_size = len + 1;	/* + 1 for nul in case len == 1 */

  outbytes_remaining = outbuf_size - 1;	/* -1 for nul */
  outp = dest = malloc (outbuf_size);

again:

  err = iconv (cd, (ICONV_CONST char **) &p, &inbytes_remaining,
	       &outp, &outbytes_remaining);

  if (err == (size_t) - 1)
    {
      switch (errno)
	{
	case EINVAL:
	  /* Incomplete text, do not report an error */
	  break;

	case E2BIG:
	  {
	    size_t used = outp - dest;

	    outbuf_size *= 2;
	    dest = realloc (dest, outbuf_size);

	    outp = dest + used;
	    outbytes_remaining = outbuf_size - used - 1;	/* -1 for nul */

	    goto again;
	  }
	  break;

	case EILSEQ:
          if (!best_effort || outbytes_remaining == 0)
            {
              have_error = 1;
              break;
            }
          else
            {
              int char_len;
              if (!from_utf8)
                char_len = 1; // not from UTF-8, one '?' will do
              else
                {
                  if      ((*p & 0x80) == 0)
                    char_len = 1;
                  else if ((*p & 0x40) == 0)
                    char_len = 1; // error: not allowed to begin a sequence
                  else if ((*p & 0x20) == 0)
                    char_len = 2;
                  else if ((*p & 0x10) == 0)
                    char_len = 3;
                  else if ((*p & 0x08) == 0)
                    char_len = 4;
                  else if ((*p & 0x04) == 0)
                    char_len = 5;
                  else if ((*p & 0x02) == 0)
                    char_len = 6;
                  else
                    char_len = 1; // error: 0xFE/0xFF not used by UTF-8
                }
              if (char_len > inbytes_remaining)
                char_len = inbytes_remaining;
              p += char_len;
              inbytes_remaining -= char_len;
              *outp++ = '?';
              --outbytes_remaining;
              if (inbytes_remaining > 0)
                goto again;
            }
          break;

	default:
	  have_error = 1;
	  break;
	}
    }

  *outp = '\0';

  if ((p - startp) != len)
    have_error = 1;


  free (startp);

  iconv_close (cd);

  if (have_error)
    {
      free (dest);
      dest = NULL;
    }

  return dest;
}

#else /* HAVE_ICONV */

const char *
stringprep_locale_charset ()
{
  return "ASCII";
}

char *
stringprep_convert (const char *str,
		    const char *to_codeset, const char *from_codeset,
                    int best_effort)
{
  char *p;
  fprintf (stderr, "libidn: warning: libiconv not installed, cannot "
	   "convert data from %s to %s\n", from_codeset, to_codeset);
  p = malloc (strlen (str) + 1);
  if (!p)
    return NULL;
  strcpy (p, str);
  return p;
}

#endif /* HAVE_ICONV */

/**
 * stringprep_locale_to_utf8:
 * @str: input zero terminated string.
 *
 * Convert string encoded in the locale's character set into UTF-8 by
 * using stringprep_convert().
 *
 * Return value: Returns newly allocated zero-terminated string which
 *   is @str transcoded into UTF-8.
 **/
char *
stringprep_locale_to_utf8 (const char *str)
{
  return stringprep_convert (str, "UTF-8", stringprep_locale_charset (), 0);
}

/**
 * stringprep_utf8_to_locale:
 * @str: input zero terminated string.
 *
 * Convert string encoded in UTF-8 into the locale's character set by
 * using stringprep_convert().
 *
 * Return value: Returns newly allocated zero-terminated string which
 *   is @str transcoded into the locale's character set.
 **/
char *
stringprep_utf8_to_locale (const char *str)
{
  return stringprep_convert (str, stringprep_locale_charset (), "UTF-8", 0);
}