1 /*
2 * Copyright (C) 2005 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 * CP936
23 */
24
25 /*
26 * The IANA has CP936 as an alias of GBK. But GBK is an official Chinese
27 * specification, whereas CP936 is de-facto maintained by Microsoft. And,
28 * of course, Microsoft modified CP936 since 1999.
29 *
30 * The differences from GBK are:
31 *
32 * 1. A single character:
33 *
34 * code CP936.TXT
35 * 0x80 0x20AC # EURO SIGN
36 *
37 * Some variants of CP936 (in JDK, Windows-2000, ICU) also add:
38 *
39 * 2. Private area mappings:
40 *
41 * code Unicode
42 * 0x{A1..A2}{40..7E,80..A0} U+E4C6..U+E585
43 * 0x{AA..AF,F8..FE}{A1..FE} U+E000..U+E4C5
44 *
45 * We add them too because, although there are backward compatibility problems
46 * when a character from a private area is moved to an official Unicode code
47 * point, they are useful for some people in practice.
48 */
49
50 static int
cp936_mbtowc(conv_t conv,ucs4_t * pwc,const unsigned char * s,int n)51 cp936_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
52 {
53 /* Try GBK first. */
54 {
55 int ret = ces_gbk_mbtowc(conv,pwc,s,n);
56 if (ret != RET_ILSEQ)
57 return ret;
58 }
59 /* Then handle the additional mappings. */
60 {
61 unsigned char c = *s;
62 if (c == 0x80) {
63 *pwc = 0x20ac;
64 return 1;
65 }
66 /* User-defined characters */
67 if (c >= 0xa1 && c <= 0xa2) {
68 if (n < 2)
69 return RET_TOOFEW(0);
70 {
71 unsigned char c2 = s[1];
72 if ((c2 >= 0x40 && c2 < 0x7f) || (c2 >= 0x80 && c2 < 0xa1)) {
73 *pwc = 0xe4c6 + 96 * (c - 0xa1) + (c2 - (c2 >= 0x80 ? 0x41 : 0x40));
74 return 2;
75 }
76 }
77 } else if ((c >= 0xaa && c < 0xb0) || (c >= 0xf8 && c < 0xff)) {
78 if (n < 2)
79 return RET_TOOFEW(0);
80 {
81 unsigned char c2 = s[1];
82 if (c2 >= 0xa1 && c2 < 0xff) {
83 *pwc = 0xe000 + 94 * (c - (c >= 0xf8 ? 0xf2 : 0xaa)) + (c2 - 0xa1);
84 return 2;
85 }
86 }
87 }
88 }
89 return RET_ILSEQ;
90 }
91
92 static int
cp936_wctomb(conv_t conv,unsigned char * r,ucs4_t wc,int n)93 cp936_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n)
94 {
95 /* Try GBK first. */
96 {
97 int ret = ces_gbk_wctomb(conv,r,wc,n);
98 if (ret != RET_ILUNI)
99 return ret;
100 }
101 /* Then handle the additional mappings. */
102 if (wc >= 0xe000 && wc < 0xe586) {
103 /* User-defined characters */
104 if (n < 2)
105 return RET_TOOFEW(0);
106 if (wc < 0xe4c6) {
107 unsigned int i = wc - 0xe000;
108 unsigned int c1 = i / 94;
109 unsigned int c2 = i % 94;
110 r[0] = c1 + (c1 < 6 ? 0xaa : 0xf2);
111 r[1] = c2 + 0xa1;
112 return 2;
113 } else {
114 unsigned int i = wc - 0xe4c6;
115 unsigned int c1 = i / 96;
116 unsigned int c2 = i % 96;
117 r[0] = c1 + 0xa1;
118 r[1] = c2 + (c2 < 0x3f ? 0x40 : 0x41);
119 return 2;
120 }
121 } else if (wc == 0x20ac) {
122 r[0] = 0x80;
123 return 1;
124 }
125 return RET_ILUNI;
126 }
127