utf7.c

Go to the documentation of this file.

 
#include "config.h"
#include <stdbool.h>
#include <string.h>
#include "private.h"
#include "mutt/lib.h"
#include "core/lib.h"

static const int Index64u[128] = {
  // clang-format off
  -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1,
  -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,-1,
  -1,-1,-1,-1, -1,-1,-1,-1, -1,-1,-1,62, 63,-1,-1,-1,
  52,53,54,55, 56,57,58,59, 60,61,-1,-1, -1,-1,-1,-1,
  -1, 0, 1, 2,  3, 4, 5, 6,  7, 8, 9,10, 11,12,13,14,
  15,16,17,18, 19,20,21,22, 23,24,25,-1, -1,-1,-1,-1,
  -1,26,27,28, 29,30,31,32, 33,34,35,36, 37,38,39,40,
  41,42,43,44, 45,46,47,48, 49,50,51,-1, -1,-1,-1,-1
  // clang-format on
};

static const char B64Chars[64] = {
  'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M',
  'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z',
  'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
  'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
  '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', ',',
};

static char *utf7_to_utf8(const char *u7, size_t u7len, char **u8, size_t *u8len)
{
  int b;
  int ch;
  int k;
 
  char *buf = MUTT_MEM_MALLOC(u7len + u7len / 8 + 1, char);
  char *p = buf;
  int pair1 = 0;
 
  for (; u7len; u7++, u7len--)
  {
    if (*u7 == '&')
    {
      u7++;
      u7len--;
 
      if (u7len && (*u7 == '-'))
      {
        *p++ = '&';
        continue;
      }
 
      ch = 0;
      k = 10;
      for (; u7len; u7++, u7len--)
      {
        if ((*u7 & 0x80) || ((b = Index64u[(int) *u7]) == -1))
          break;
        if (k > 0)
        {
          ch |= b << k;
          k -= 6;
        }
        else
        {
          ch |= b >> (-k);
          if (ch < 0x80)
          {
            if ((0x20 <= ch) && (ch < 0x7f))
            {
              /* Printable US-ASCII */
              goto bail;
            }
            *p++ = ch;
          }
          else if (ch < 0x800)
          {
            *p++ = 0xc0 | (ch >> 6);
            *p++ = 0x80 | (ch & 0x3f);
          }
          else
          {
            /* High surrogate pair */
            if ((ch & ~0x3ff) == 0xd800)
            {
              if (pair1)
                goto bail;
              pair1 = ch;
            }
            else
            {
              /* Low surrogate pair */
              if ((ch & ~0x3ff) == 0xdc00)
              {
                if (!pair1)
                  goto bail;
 
                ch = ((pair1 - 0xd800) << 10) + (ch - 0xdc00) + 0x10000;
                pair1 = 0;
              }
              if (pair1)
                goto bail;
 
              if (ch < 0x10000)
              {
                *p++ = 0xe0 | (ch >> 12);
                *p++ = 0x80 | ((ch >> 6) & 0x3f);
                *p++ = 0x80 | (ch & 0x3f);
              }
              else
              {
                *p++ = 0xf0 | (ch >> 18);
                *p++ = 0x80 | ((ch >> 12) & 0x3f);
                *p++ = 0x80 | ((ch >> 6) & 0x3f);
                *p++ = 0x80 | (ch & 0x3f);
              }
            }
          }
 
          ch = (b << (16 + k)) & 0xffff;
          k += 10;
        }
      }
      if (ch || (k < 6))
      {
        /* Non-zero or too many extra bits */
        goto bail;
      }
      if (!u7len || (*u7 != '-'))
      {
        /* BASE64 not properly terminated */
        goto bail;
      }
      if ((u7len > 2) && (u7[1] == '&') && (u7[2] != '-'))
      {
        /* Adjacent BASE64 sections */
        goto bail;
      }
    }
    else if ((*u7 < 0x20) || (*u7 >= 0x7f))
    {
      /* Not printable US-ASCII */
      goto bail;
    }
    else
    {
      *p++ = *u7;
    }
  }
  *p++ = '\0';
  if (u8len)
    *u8len = p - buf;
 
  MUTT_MEM_REALLOC(&buf, p - buf, char);
  if (u8)
    *u8 = buf;
  return buf;
 
bail:
  FREE(&buf);
  return NULL;
}

static char *utf8_to_utf7(const char *u8, size_t u8len, char **u7, size_t *u7len)
{
  int ch;
  int n;
  int b = 0;
  int k = 0;
  bool base64 = false;
 
  /* In the worst case we convert 2 chars to 7 chars. For example:
   * "\x10&\x10&..." -> "&ABA-&-&ABA-&-...".  */
  char *buf = MUTT_MEM_MALLOC((u8len / 2) * 7 + 6, char);
  char *p = buf;
 
  while (u8len)
  {
    unsigned char c = *u8;
 
    if (c < 0x80)
    {
      ch = c;
      n = 0;
    }
    else if (c < 0xc2)
    {
      goto bail;
    }
    else if (c < 0xe0)
    {
      ch = c & 0x1f;
      n = 1;
    }
    else if (c < 0xf0)
    {
      ch = c & 0x0f;
      n = 2;
    }
    else if (c < 0xf8)
    {
      ch = c & 0x07;
      n = 3;
    }
    else if (c < 0xfc)
    {
      ch = c & 0x03;
      n = 4;
    }
    else if (c < 0xfe)
    {
      ch = c & 0x01;
      n = 5;
    }
    else
    {
      goto bail;
    }
 
    u8++;
    u8len--;
    if (n > u8len)
      goto bail;
    for (int i = 0; i < n; i++)
    {
      if ((u8[i] & 0xc0) != 0x80)
        goto bail;
      ch = (ch << 6) | (u8[i] & 0x3f);
    }
    if ((n > 1) && !(ch >> (n * 5 + 1)))
      goto bail;
    if (ch > 0x10ffff)
    {
      /* Outside Unicode, so it has no UTF-16 representation */
      goto bail;
    }
    u8 += n;
    u8len -= n;
 
    if ((ch < 0x20) || (ch >= 0x7f))
    {
      if (!base64)
      {
        *p++ = '&';
        base64 = true;
        b = 0;
        k = 10;
      }
 
      // For code points >= 0x10000 we need to use a UTF-16 surrogate pair
      if (ch & ~0xffff)
      {
        ch -= 0x10000;
        int pair1 = 0xd800 + (ch >> 10);
        int pair2 = 0xdc00 + (ch & 0x3ff);
 
        /* Output the high surrogate */
        *p++ = B64Chars[b | pair1 >> k];
        k -= 6;
        for (; k >= 0; k -= 6)
          *p++ = B64Chars[(pair1 >> k) & 0x3f];
        b = (pair1 << (-k)) & 0x3f;
        k += 16;
 
        /* The low surrogate will be output just below */
        ch = pair2;
      }
 
      *p++ = B64Chars[b | ch >> k];
      k -= 6;
      for (; k >= 0; k -= 6)
        *p++ = B64Chars[(ch >> k) & 0x3f];
      b = (ch << (-k)) & 0x3f;
      k += 16;
    }
    else
    {
      if (base64)
      {
        if (k > 10)
          *p++ = B64Chars[b];
        *p++ = '-';
        base64 = false;
      }
      *p++ = ch;
      if (ch == '&')
        *p++ = '-';
    }
  }
 
  if (base64)
  {
    if (k > 10)
      *p++ = B64Chars[b];
    *p++ = '-';
  }
 
  *p++ = '\0';
  if (u7len)
    *u7len = p - buf;
  MUTT_MEM_REALLOC(&buf, p - buf, char);
  if (u7)
    *u7 = buf;
  return buf;
 
bail:
  FREE(&buf);
  return NULL;
}

void imap_utf_encode(bool unicode, char **s)
{
  if (!s || !*s)
    return;
 
  const char *c_charset = cc_charset();
  if (!c_charset)
    return;
 
  if (unicode && mutt_ch_is_utf8(c_charset))
  {
    return;
  }
 
  if (mutt_ch_convert_string(s, c_charset, "utf-8", MUTT_ICONV_NONE) != 0)
  {
    FREE(s);
    return;
  }
 
  if (!unicode)
  {
    char *utf7 = utf8_to_utf7(*s, strlen(*s), NULL, 0);
    FREE(s);
    *s = utf7;
  }
}

void imap_utf_decode(bool unicode, char **s)
{
  if (!s || !*s)
    return;
 
  const char *c_charset = cc_charset();
  if (!c_charset)
    return;
 
  if (unicode && mutt_ch_is_utf8(c_charset))
  {
    return;
  }
 
  if (!unicode)
  {
    char *utf8 = utf7_to_utf8(*s, strlen(*s), 0, 0);
    FREE(s);
    *s = utf8;
  }
 
  if (mutt_ch_convert_string(s, "utf-8", c_charset, MUTT_ICONV_NONE) != 0)
  {
    FREE(s);
  }
}