1 /* Character set conversion.
2 Copyright (C) 1999-2001, 2007, 2009-2021 Free Software Foundation, Inc.
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation; either version 3, or (at your option)
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License along
15 with this program; if not, see <https://www.gnu.org/licenses/>. */
33 /* UTF-{16,32}{BE,LE} converters taken from GNU libiconv 1.11. */
35 /* Return code if invalid. (xxx_mbtowc) */
37 /* Return code if no bytes were read. (xxx_mbtowc) */
38 # define RET_TOOFEW -2
40 /* Return code if invalid. (xxx_wctomb) */
42 /* Return code if output buffer is too small. (xxx_wctomb, xxx_reset) */
43 # define RET_TOOSMALL -2
49 /* Specification: RFC 2781 */
52 utf16be_mbtowc (ucs4_t *pwc, const unsigned char *s, size_t n)
56 ucs4_t wc = (s[0] << 8) + s[1];
57 if (wc >= 0xd800 && wc < 0xdc00)
61 ucs4_t wc2 = (s[2] << 8) + s[3];
62 if (!(wc2 >= 0xdc00 && wc2 < 0xe000))
64 *pwc = 0x10000 + ((wc - 0xd800) << 10) + (wc2 - 0xdc00);
68 else if (wc >= 0xdc00 && wc < 0xe000)
82 utf16be_wctomb (unsigned char *r, ucs4_t wc, size_t n)
84 if (!(wc >= 0xd800 && wc < 0xe000))
90 r[0] = (unsigned char) (wc >> 8);
91 r[1] = (unsigned char) wc;
97 else if (wc < 0x110000)
101 ucs4_t wc1 = 0xd800 + ((wc - 0x10000) >> 10);
102 ucs4_t wc2 = 0xdc00 + ((wc - 0x10000) & 0x3ff);
103 r[0] = (unsigned char) (wc1 >> 8);
104 r[1] = (unsigned char) wc1;
105 r[2] = (unsigned char) (wc2 >> 8);
106 r[3] = (unsigned char) wc2;
120 /* Specification: RFC 2781 */
123 utf16le_mbtowc (ucs4_t *pwc, const unsigned char *s, size_t n)
127 ucs4_t wc = s[0] + (s[1] << 8);
128 if (wc >= 0xd800 && wc < 0xdc00)
132 ucs4_t wc2 = s[2] + (s[3] << 8);
133 if (!(wc2 >= 0xdc00 && wc2 < 0xe000))
135 *pwc = 0x10000 + ((wc - 0xd800) << 10) + (wc2 - 0xdc00);
139 else if (wc >= 0xdc00 && wc < 0xe000)
153 utf16le_wctomb (unsigned char *r, ucs4_t wc, size_t n)
155 if (!(wc >= 0xd800 && wc < 0xe000))
161 r[0] = (unsigned char) wc;
162 r[1] = (unsigned char) (wc >> 8);
168 else if (wc < 0x110000)
172 ucs4_t wc1 = 0xd800 + ((wc - 0x10000) >> 10);
173 ucs4_t wc2 = 0xdc00 + ((wc - 0x10000) & 0x3ff);
174 r[0] = (unsigned char) wc1;
175 r[1] = (unsigned char) (wc1 >> 8);
176 r[2] = (unsigned char) wc2;
177 r[3] = (unsigned char) (wc2 >> 8);
191 /* Specification: Unicode 3.1 Standard Annex #19 */
194 utf32be_mbtowc (ucs4_t *pwc, const unsigned char *s, size_t n)
198 ucs4_t wc = (s[0] << 24) + (s[1] << 16) + (s[2] << 8) + s[3];
199 if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000))
211 utf32be_wctomb (unsigned char *r, ucs4_t wc, size_t n)
213 if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000))
218 r[1] = (unsigned char) (wc >> 16);
219 r[2] = (unsigned char) (wc >> 8);
220 r[3] = (unsigned char) wc;
233 /* Specification: Unicode 3.1 Standard Annex #19 */
236 utf32le_mbtowc (ucs4_t *pwc, const unsigned char *s, size_t n)
240 ucs4_t wc = s[0] + (s[1] << 8) + (s[2] << 16) + (s[3] << 24);
241 if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000))
253 utf32le_wctomb (unsigned char *r, ucs4_t wc, size_t n)
255 if (wc < 0x110000 && !(wc >= 0xd800 && wc < 0xe000))
259 r[0] = (unsigned char) wc;
260 r[1] = (unsigned char) (wc >> 8);
261 r[2] = (unsigned char) (wc >> 16);
274 rpl_iconv (iconv_t cd,
275 ICONV_CONST char **inbuf, size_t *inbytesleft,
276 char **outbuf, size_t *outbytesleft)
279 #if REPLACE_ICONV_UTF
280 switch ((uintptr_t) cd)
283 int (*xxx_wctomb) (unsigned char *, ucs4_t, size_t);
285 case (uintptr_t) _ICONV_UTF8_UTF16BE:
286 xxx_wctomb = utf16be_wctomb;
288 case (uintptr_t) _ICONV_UTF8_UTF16LE:
289 xxx_wctomb = utf16le_wctomb;
291 case (uintptr_t) _ICONV_UTF8_UTF32BE:
292 xxx_wctomb = utf32be_wctomb;
294 case (uintptr_t) _ICONV_UTF8_UTF32LE:
295 xxx_wctomb = utf32le_wctomb;
299 if (inbuf == NULL || *inbuf == NULL)
302 ICONV_CONST char *inptr = *inbuf;
303 size_t inleft = *inbytesleft;
304 char *outptr = *outbuf;
305 size_t outleft = *outbytesleft;
310 int m = u8_mbtoucr (&uc, (const uint8_t *) inptr, inleft);
329 int n = xxx_wctomb ((uint8_t *) outptr, uc, outleft);
338 if (n == RET_TOOSMALL)
356 *inbytesleft = inleft;
358 *outbytesleft = outleft;
364 int (*xxx_mbtowc) (ucs4_t *, const unsigned char *, size_t);
366 case (uintptr_t) _ICONV_UTF16BE_UTF8:
367 xxx_mbtowc = utf16be_mbtowc;
369 case (uintptr_t) _ICONV_UTF16LE_UTF8:
370 xxx_mbtowc = utf16le_mbtowc;
372 case (uintptr_t) _ICONV_UTF32BE_UTF8:
373 xxx_mbtowc = utf32be_mbtowc;
375 case (uintptr_t) _ICONV_UTF32LE_UTF8:
376 xxx_mbtowc = utf32le_mbtowc;
380 if (inbuf == NULL || *inbuf == NULL)
383 ICONV_CONST char *inptr = *inbuf;
384 size_t inleft = *inbytesleft;
385 char *outptr = *outbuf;
386 size_t outleft = *outbytesleft;
391 int m = xxx_mbtowc (&uc, (const uint8_t *) inptr, inleft);
410 int n = u8_uctomb ((uint8_t *) outptr, uc, outleft);
437 *inbytesleft = inleft;
439 *outbytesleft = outleft;
445 return iconv (cd, inbuf, inbytesleft, outbuf, outbytesleft);