1 /*
2 * "streamable kanji code filter and converter"
3 * Copyright (c) 1998-2002 HappySize, Inc. All rights reserved.
4 *
5 * LICENSE NOTICES
6 *
7 * This file is part of "streamable kanji code filter and converter",
8 * which is distributed under the terms of GNU Lesser General Public
9 * License (version 2) as published by the Free Software Foundation.
10 *
11 * This software is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU Lesser General Public License for more details.
15 *
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with "streamable kanji code filter and converter";
18 * if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19 * Suite 330, Boston, MA 02111-1307 USA
20 *
21 * The author of this file:
22 *
23 */
24 /*
25 * The source code included in this files was separated from mbfilter_ja.c
26 * by moriyoshi koizumi <moriyoshi@php.net> on 4 dec 2002.
27 *
28 */
29
30 #ifdef HAVE_CONFIG_H
31 #include "config.h"
32 #endif
33
34 #include "mbfilter.h"
35 #include "mbfilter_sjis.h"
36
37 #include "unicode_table_cp932_ext.h"
38 #include "unicode_table_jis.h"
39
40 static int mbfl_filt_ident_sjis(int c, mbfl_identify_filter *filter);
41
42 static const unsigned char mblen_table_sjis[] = { /* 0x80-0x9f,0xE0-0xFF */
43 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
44 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
45 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
46 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
47 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
48 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
49 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
50 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
51 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
52 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
53 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
54 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
55 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
56 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
57 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
58 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2
59 };
60
61 static const char *mbfl_encoding_sjis_aliases[] = {"x-sjis", "SHIFT-JIS", NULL};
62
63 const mbfl_encoding mbfl_encoding_sjis = {
64 mbfl_no_encoding_sjis,
65 "SJIS",
66 "Shift_JIS",
67 (const char *(*)[])&mbfl_encoding_sjis_aliases,
68 mblen_table_sjis,
69 MBFL_ENCTYPE_MBCS
70 };
71
72 const struct mbfl_identify_vtbl vtbl_identify_sjis = {
73 mbfl_no_encoding_sjis,
74 mbfl_filt_ident_common_ctor,
75 mbfl_filt_ident_common_dtor,
76 mbfl_filt_ident_sjis
77 };
78
79 const struct mbfl_convert_vtbl vtbl_sjis_wchar = {
80 mbfl_no_encoding_sjis,
81 mbfl_no_encoding_wchar,
82 mbfl_filt_conv_common_ctor,
83 mbfl_filt_conv_common_dtor,
84 mbfl_filt_conv_sjis_wchar,
85 mbfl_filt_conv_common_flush
86 };
87
88 const struct mbfl_convert_vtbl vtbl_wchar_sjis = {
89 mbfl_no_encoding_wchar,
90 mbfl_no_encoding_sjis,
91 mbfl_filt_conv_common_ctor,
92 mbfl_filt_conv_common_dtor,
93 mbfl_filt_conv_wchar_sjis,
94 mbfl_filt_conv_common_flush
95 };
96
97 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
98
99 #define SJIS_ENCODE(c1,c2,s1,s2) \
100 do { \
101 s1 = c1; \
102 s1--; \
103 s1 >>= 1; \
104 if ((c1) < 0x5f) { \
105 s1 += 0x71; \
106 } else { \
107 s1 += 0xb1; \
108 } \
109 s2 = c2; \
110 if ((c1) & 1) { \
111 if ((c2) < 0x60) { \
112 s2--; \
113 } \
114 s2 += 0x20; \
115 } else { \
116 s2 += 0x7e; \
117 } \
118 } while (0)
119
120 #define SJIS_DECODE(c1,c2,s1,s2) \
121 do { \
122 s1 = c1; \
123 if (s1 < 0xa0) { \
124 s1 -= 0x81; \
125 } else { \
126 s1 -= 0xc1; \
127 } \
128 s1 <<= 1; \
129 s1 += 0x21; \
130 s2 = c2; \
131 if (s2 < 0x9f) { \
132 if (s2 < 0x7f) { \
133 s2++; \
134 } \
135 s2 -= 0x20; \
136 } else { \
137 s1++; \
138 s2 -= 0x7e; \
139 } \
140 } while (0)
141
142
143 /*
144 * SJIS => wchar
145 */
146 int
mbfl_filt_conv_sjis_wchar(int c,mbfl_convert_filter * filter)147 mbfl_filt_conv_sjis_wchar(int c, mbfl_convert_filter *filter)
148 {
149 int c1, s1, s2, w;
150
151 switch (filter->status) {
152 case 0:
153 if (c >= 0 && c < 0x80) { /* latin */
154 CK((*filter->output_function)(c, filter->data));
155 } else if (c > 0xa0 && c < 0xe0) { /* kana */
156 CK((*filter->output_function)(0xfec0 + c, filter->data));
157 } else if (c > 0x80 && c < 0xfd && c != 0xa0) { /* kanji first char */
158 filter->status = 1;
159 filter->cache = c;
160 } else {
161 w = c & MBFL_WCSGROUP_MASK;
162 w |= MBFL_WCSGROUP_THROUGH;
163 CK((*filter->output_function)(w, filter->data));
164 }
165 break;
166
167 case 1: /* kanji second char */
168 filter->status = 0;
169 c1 = filter->cache;
170 if (c >= 0x40 && c <= 0xfc && c != 0x7f) {
171 SJIS_DECODE(c1, c, s1, s2);
172 w = (s1 - 0x21)*94 + s2 - 0x21;
173 if (w >= 0 && w < jisx0208_ucs_table_size) {
174 w = jisx0208_ucs_table[w];
175 } else {
176 w = 0;
177 }
178 if (w <= 0) {
179 if (s1 < 0x7f && s2 < 0x7f) {
180 w = (s1 << 8) | s2;
181 w &= MBFL_WCSPLANE_MASK;
182 w |= MBFL_WCSPLANE_JIS0208;
183 } else {
184 w = (c1 << 8) | c;
185 w &= MBFL_WCSGROUP_MASK;
186 w |= MBFL_WCSGROUP_THROUGH;
187 }
188 }
189 CK((*filter->output_function)(w, filter->data));
190 } else if ((c >= 0 && c < 0x21) || c == 0x7f) { /* CTLs */
191 CK((*filter->output_function)(c, filter->data));
192 } else {
193 w = (c1 << 8) | c;
194 w &= MBFL_WCSGROUP_MASK;
195 w |= MBFL_WCSGROUP_THROUGH;
196 CK((*filter->output_function)(w, filter->data));
197 }
198 break;
199
200 default:
201 filter->status = 0;
202 break;
203 }
204
205 return c;
206 }
207
208 /*
209 * wchar => SJIS
210 */
211 int
mbfl_filt_conv_wchar_sjis(int c,mbfl_convert_filter * filter)212 mbfl_filt_conv_wchar_sjis(int c, mbfl_convert_filter *filter)
213 {
214 int c1, c2, s1, s2;
215
216 s1 = 0;
217 if (c >= ucs_a1_jis_table_min && c < ucs_a1_jis_table_max) {
218 s1 = ucs_a1_jis_table[c - ucs_a1_jis_table_min];
219 } else if (c >= ucs_a2_jis_table_min && c < ucs_a2_jis_table_max) {
220 s1 = ucs_a2_jis_table[c - ucs_a2_jis_table_min];
221 } else if (c >= ucs_i_jis_table_min && c < ucs_i_jis_table_max) {
222 s1 = ucs_i_jis_table[c - ucs_i_jis_table_min];
223 } else if (c >= ucs_r_jis_table_min && c < ucs_r_jis_table_max) {
224 s1 = ucs_r_jis_table[c - ucs_r_jis_table_min];
225 }
226 if (s1 <= 0) {
227 c1 = c & ~MBFL_WCSPLANE_MASK;
228 if (c1 == MBFL_WCSPLANE_JIS0208) {
229 s1 = c & MBFL_WCSPLANE_MASK;
230 } else if (c == 0xa5) { /* YEN SIGN */
231 s1 = 0x216f; /* FULLWIDTH YEN SIGN */
232 } else if (c == 0x203e) { /* OVER LINE */
233 s1 = 0x2131; /* FULLWIDTH MACRON */
234 } else if (c == 0xff3c) { /* FULLWIDTH REVERSE SOLIDUS */
235 s1 = 0x2140;
236 } else if (c == 0xff5e) { /* FULLWIDTH TILDE */
237 s1 = 0x2141;
238 } else if (c == 0x2225) { /* PARALLEL TO */
239 s1 = 0x2142;
240 } else if (c == 0xff0d) { /* FULLWIDTH HYPHEN-MINUS */
241 s1 = 0x215d;
242 } else if (c == 0xffe0) { /* FULLWIDTH CENT SIGN */
243 s1 = 0x2171;
244 } else if (c == 0xffe1) { /* FULLWIDTH POUND SIGN */
245 s1 = 0x2172;
246 } else if (c == 0xffe2) { /* FULLWIDTH NOT SIGN */
247 s1 = 0x224c;
248 }
249 if (c == 0) {
250 s1 = 0;
251 } else if (s1 <= 0) {
252 s1 = -1;
253 }
254 } else if (s1 >= 0x8080) {
255 s1 = -1;
256 }
257 if (s1 >= 0) {
258 if (s1 < 0x100) { /* latin or kana */
259 CK((*filter->output_function)(s1, filter->data));
260 } else { /* kanji */
261 c1 = (s1 >> 8) & 0xff;
262 c2 = s1 & 0xff;
263 SJIS_ENCODE(c1, c2, s1, s2);
264 CK((*filter->output_function)(s1, filter->data));
265 CK((*filter->output_function)(s2, filter->data));
266 }
267 } else {
268 if (filter->illegal_mode != MBFL_OUTPUTFILTER_ILLEGAL_MODE_NONE) {
269 CK(mbfl_filt_conv_illegal_output(c, filter));
270 }
271 }
272
273 return c;
274 }
275
mbfl_filt_ident_sjis(int c,mbfl_identify_filter * filter)276 static int mbfl_filt_ident_sjis(int c, mbfl_identify_filter *filter)
277 {
278 if (filter->status) { /* kanji second char */
279 if (c < 0x40 || c > 0xfc || c == 0x7f) { /* bad */
280 filter->flag = 1;
281 }
282 filter->status = 0;
283 } else if (c >= 0 && c < 0x80) { /* latin ok */
284 ;
285 } else if (c > 0xa0 && c < 0xe0) { /* kana ok */
286 ;
287 } else if (c > 0x80 && c < 0xf0 && c != 0xa0) { /* kanji first char */
288 filter->status = 1;
289 } else { /* bad */
290 filter->flag = 1;
291 }
292
293 return c;
294 }
295
296