Annotation of embedaddon/libiconv/lib/utf32.h, revision 1.1.1.1

1.1       misho       1: /*
                      2:  * Copyright (C) 1999-2001, 2008 Free Software Foundation, Inc.
                      3:  * This file is part of the GNU LIBICONV Library.
                      4:  *
                      5:  * The GNU LIBICONV Library is free software; you can redistribute it
                      6:  * and/or modify it under the terms of the GNU Library General Public
                      7:  * License as published by the Free Software Foundation; either version 2
                      8:  * of the License, or (at your option) any later version.
                      9:  *
                     10:  * The GNU LIBICONV Library is distributed in the hope that it will be
                     11:  * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
                     12:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
                     13:  * Library General Public License for more details.
                     14:  *
                     15:  * You should have received a copy of the GNU Library General Public
                     16:  * License along with the GNU LIBICONV Library; see the file COPYING.LIB.
                     17:  * If not, write to the Free Software Foundation, Inc., 51 Franklin Street,
                     18:  * Fifth Floor, Boston, MA 02110-1301, USA.
                     19:  */
                     20: 
                     21: /*
                     22:  * UTF-32
                     23:  */
                     24: 
                     25: /* Specification: Unicode 3.1 Standard Annex #19 */
                     26: 
                     27: /* Here we accept FFFE0000/0000FEFF marks as endianness indicators
                     28:    everywhere in the stream, not just at the beginning. (This is contrary
                     29:    to what #19 D36c specifies, but it allows concatenation of byte
                     30:    sequences to work flawlessly, while disagreeing with #19 behaviour
                     31:    only for strings containing U+FEFF characters, which is quite rare.)
                     32:    The default is big-endian. */
                     33: /* The state is 0 if big-endian, 1 if little-endian. */
                     34: static int
                     35: utf32_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
                     36: {
                     37:   state_t state = conv->istate;
                     38:   int count = 0;
                     39:   for (; n >= 4;) {
                     40:     ucs4_t wc = (state
                     41:                   ? s[0] + (s[1] << 8) + (s[2] << 16) + (s[3] << 24)
                     42:                   : (s[0] << 24) + (s[1] << 16) + (s[2] << 8) + s[3]);
                     43:     count += 4;
                     44:     if (wc == 0x0000feff) {
                     45:     } else if (wc == 0xfffe0000u) {
                     46:       state ^= 1;
                     47:     } else {
                     48:       if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000)) {
                     49:         *pwc = wc;
                     50:         conv->istate = state;
                     51:         return count;
                     52:       } else {
                     53:         conv->istate = state;
                     54:         return RET_SHIFT_ILSEQ(count);
                     55:       }
                     56:     }
                     57:     s += 4; n -= 4;
                     58:   }
                     59:   conv->istate = state;
                     60:   return RET_TOOFEW(count);
                     61: }
                     62: 
                     63: /* We output UTF-32 in big-endian order, with byte-order mark. */
                     64: /* The state is 0 at the beginning, 1 after the BOM has been written. */
                     65: static int
                     66: utf32_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)
                     67: {
                     68:   if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000)) {
                     69:     int count = 0;
                     70:     if (!conv->ostate) {
                     71:       if (n >= 4) {
                     72:         r[0] = 0x00;
                     73:         r[1] = 0x00;
                     74:         r[2] = 0xFE;
                     75:         r[3] = 0xFF;
                     76:         r += 4; n -= 4; count += 4;
                     77:       } else
                     78:         return RET_TOOSMALL;
                     79:     }
                     80:     if (wc < 0x110000) {
                     81:       if (n >= 4) {
                     82:         r[0] = 0;
                     83:         r[1] = (unsigned char) (wc >> 16);
                     84:         r[2] = (unsigned char) (wc >> 8);
                     85:         r[3] = (unsigned char) wc;
                     86:         conv->ostate = 1;
                     87:         return count+4;
                     88:       } else
                     89:         return RET_TOOSMALL;
                     90:     }
                     91:   }
                     92:   return RET_ILUNI;
                     93: }

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>