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_iso2022_jp_ms.c
26 * by Rui Hirokawa <hirokawa@php.net> on 25 July 2011.
27 *
28 */
29
30 #ifdef HAVE_CONFIG_H
31 #include "config.h"
32 #endif
33
34 #include "mbfilter.h"
35 #include "mbfilter_iso2022jp_mobile.h"
36 #include "mbfilter_sjis_mobile.h"
37
38 #include "unicode_table_cp932_ext.h"
39 #include "unicode_table_jis.h"
40 #include "cp932_table.h"
41
42 extern int mbfl_filt_conv_any_jis_flush(mbfl_convert_filter *filter);
43 extern int mbfl_filt_ident_2022jpms(int c, mbfl_identify_filter *filter);
44
45 static const char *mbfl_encoding_2022jp_kddi_aliases[] = {"ISO-2022-JP-KDDI", NULL};
46
47 const mbfl_encoding mbfl_encoding_2022jp_kddi = {
48 mbfl_no_encoding_2022jp_kddi,
49 "ISO-2022-JP-MOBILE#KDDI",
50 "ISO-2022-JP",
51 &mbfl_encoding_2022jp_kddi_aliases,
52 NULL,
53 MBFL_ENCTYPE_MBCS | MBFL_ENCTYPE_SHFTCODE | MBFL_ENCTYPE_GL_UNSAFE,
54 &vtbl_2022jp_kddi_wchar,
55 &vtbl_wchar_2022jp_kddi
56 };
57
58 const struct mbfl_identify_vtbl vtbl_identify_2022jp_kddi = {
59 mbfl_no_encoding_2022jp_kddi,
60 mbfl_filt_ident_common_ctor,
61 mbfl_filt_ident_common_dtor,
62 mbfl_filt_ident_2022jpms
63 };
64
65 const struct mbfl_convert_vtbl vtbl_2022jp_kddi_wchar = {
66 mbfl_no_encoding_2022jp_kddi,
67 mbfl_no_encoding_wchar,
68 mbfl_filt_conv_common_ctor,
69 mbfl_filt_conv_common_dtor,
70 mbfl_filt_conv_2022jp_mobile_wchar,
71 mbfl_filt_conv_common_flush
72 };
73
74 const struct mbfl_convert_vtbl vtbl_wchar_2022jp_kddi = {
75 mbfl_no_encoding_wchar,
76 mbfl_no_encoding_2022jp_kddi,
77 mbfl_filt_conv_common_ctor,
78 mbfl_filt_conv_common_dtor,
79 mbfl_filt_conv_wchar_2022jp_mobile,
80 mbfl_filt_conv_any_jis_flush
81 };
82
83 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
84
85 #define sjistoidx(c1, c2) \
86 (((c1) > 0x9f) \
87 ? (((c1) - 0xc1) * 188 + (c2) - (((c2) > 0x7e) ? 0x41 : 0x40)) \
88 : (((c1) - 0x81) * 188 + (c2) - (((c2) > 0x7e) ? 0x41 : 0x40)))
89 #define idxtojis1(c) (((c) / 94) + 0x21)
90 #define idxtojis2(c) (((c) % 94) + 0x21)
91
92 #define SJIS_ENCODE(c1,c2,s1,s2) \
93 do { \
94 s1 = c1; \
95 s1--; \
96 s1 >>= 1; \
97 if ((c1) < 0x5f) { \
98 s1 += 0x71; \
99 } else { \
100 s1 += 0xb1; \
101 } \
102 s2 = c2; \
103 if ((c1) & 1) { \
104 if ((c2) < 0x60) { \
105 s2--; \
106 } \
107 s2 += 0x20; \
108 } else { \
109 s2 += 0x7e; \
110 } \
111 } while (0)
112
113 #define SJIS_DECODE(c1,c2,s1,s2) \
114 do { \
115 s1 = c1; \
116 if (s1 < 0xa0) { \
117 s1 -= 0x81; \
118 } else { \
119 s1 -= 0xc1; \
120 } \
121 s1 <<= 1; \
122 s1 += 0x21; \
123 s2 = c2; \
124 if (s2 < 0x9f) { \
125 if (s2 < 0x7f) { \
126 s2++; \
127 } \
128 s2 -= 0x20; \
129 } else { \
130 s1++; \
131 s2 -= 0x7e; \
132 } \
133 } while (0)
134
135 #define CODE2JIS(c1,c2,s1,s2) \
136 c1 = (s1)/94+0x21; \
137 c2 = (s1)-94*((c1)-0x21)+0x21; \
138 s1 = ((c1) << 8) | (c2); \
139 s2 = 1
140
141 /*
142 * ISO-2022-JP-Mobile => wchar
143 */
144 int
mbfl_filt_conv_2022jp_mobile_wchar(int c,mbfl_convert_filter * filter)145 mbfl_filt_conv_2022jp_mobile_wchar(int c, mbfl_convert_filter *filter)
146 {
147 int c1, s, w, snd = 0;
148
149 retry:
150 switch (filter->status & 0xf) {
151 /* case 0x00: ASCII */
152 /* case 0x10: X 0201 latin */
153 /* case 0x20: X 0201 kana */
154 /* case 0x80: X 0208 */
155 case 0:
156 if (c == 0x1b) {
157 filter->status += 2;
158 } else if (filter->status == 0x20 && c > 0x20 && c < 0x60) { /* kana */
159 CK((*filter->output_function)(0xff40 + c, filter->data));
160 } else if (filter->status == 0x80 && c > 0x20 && c < 0x80) { /* kanji first char */
161 filter->cache = c;
162 filter->status += 1;
163 } else if (c >= 0 && c < 0x80) { /* latin, CTLs */
164 CK((*filter->output_function)(c, filter->data));
165 } else if (c > 0xa0 && c < 0xe0) { /* GR kana */
166 CK((*filter->output_function)(0xfec0 + c, filter->data));
167 } else {
168 w = c & MBFL_WCSGROUP_MASK;
169 w |= MBFL_WCSGROUP_THROUGH;
170 CK((*filter->output_function)(w, filter->data));
171 }
172 break;
173
174 /* case 0x81: X 0208 second char */
175 case 1:
176 w = 0;
177 filter->status &= ~0xf;
178 c1 = filter->cache;
179 if (c > 0x20 && c < 0x7f) {
180 s = (c1 - 0x21)*94 + c - 0x21;
181
182 if (s <= 137) {
183 if (s == 31) {
184 w = 0xff3c; /* FULLWIDTH REVERSE SOLIDUS */
185 } else if (s == 32) {
186 w = 0xff5e; /* FULLWIDTH TILDE */
187 } else if (s == 33) {
188 w = 0x2225; /* PARALLEL TO */
189 } else if (s == 60) {
190 w = 0xff0d; /* FULLWIDTH HYPHEN-MINUS */
191 } else if (s == 80) {
192 w = 0xffe0; /* FULLWIDTH CENT SIGN */
193 } else if (s == 81) {
194 w = 0xffe1; /* FULLWIDTH POUND SIGN */
195 } else if (s == 137) {
196 w = 0xffe2; /* FULLWIDTH NOT SIGN */
197 }
198 }
199
200 if (w == 0) {
201 if (s >= cp932ext1_ucs_table_min && s < cp932ext1_ucs_table_max) { /* vendor ext1 (13ku) */
202 w = cp932ext1_ucs_table[s - cp932ext1_ucs_table_min];
203 } else if (s >= 0 && s < jisx0208_ucs_table_size) {
204 w = jisx0208_ucs_table[s];
205 } else {
206 w = 0;
207 }
208 }
209
210 if (s >= (84*94) && s < 91*94) {
211 s += 22*94;
212 if (filter->from->no_encoding == mbfl_no_encoding_2022jp_kddi) {
213 w = mbfilter_sjis_emoji_kddi2unicode(s, &snd);
214 }
215 if (w > 0 && snd > 0) {
216 CK((*filter->output_function)(snd, filter->data));
217 }
218 }
219
220 if (w <= 0) {
221 w = (c1 << 8) | c;
222 w &= MBFL_WCSPLANE_MASK;
223 w |= MBFL_WCSPLANE_JIS0208;
224 }
225 CK((*filter->output_function)(w, filter->data));
226 } else if (c == 0x1b) {
227 filter->status += 2;
228 } else if ((c >= 0 && c < 0x21) || c == 0x7f) { /* CTLs */
229 CK((*filter->output_function)(c, filter->data));
230 } else {
231 w = (c1 << 8) | c;
232 w &= MBFL_WCSGROUP_MASK;
233 w |= MBFL_WCSGROUP_THROUGH;
234 CK((*filter->output_function)(w, filter->data));
235 }
236 break;
237
238 /* ESC */
239 /* case 0x02: */
240 /* case 0x12: */
241 /* case 0x22: */
242 /* case 0x82: */
243 case 2:
244 if (c == 0x24) { /* '$' */
245 filter->status++;
246 } else if (c == 0x28) { /* '(' */
247 filter->status += 3;
248 } else {
249 filter->status &= ~0xf;
250 CK((*filter->output_function)(0x1b, filter->data));
251 goto retry;
252 }
253 break;
254
255 /* ESC $ */
256 /* case 0x03: */
257 /* case 0x13: */
258 /* case 0x23: */
259 /* case 0x83: */
260 case 3:
261 if (c == 0x40 || c == 0x42) { /* '@' or 'B' */
262 filter->status = 0x80;
263 } else if (c == 0x28) { /* '(' */
264 filter->status++;
265 } else {
266 filter->status &= ~0xf;
267 CK((*filter->output_function)(0x1b, filter->data));
268 CK((*filter->output_function)(0x24, filter->data));
269 goto retry;
270 }
271 break;
272
273 /* ESC $ ( */
274 /* case 0x04: */
275 /* case 0x14: */
276 /* case 0x24: */
277 /* case 0x84: */
278 case 4:
279 if (c == 0x40 || c == 0x42) { /* '@' or 'B' */
280 filter->status = 0x80;
281 } else {
282 filter->status &= ~0xf;
283 CK((*filter->output_function)(0x1b, filter->data));
284 CK((*filter->output_function)(0x24, filter->data));
285 CK((*filter->output_function)(0x28, filter->data));
286 goto retry;
287 }
288 break;
289
290 /* ESC ( */
291 /* case 0x05: */
292 /* case 0x15: */
293 /* case 0x25: */
294 /* case 0x85: */
295 case 5:
296 if (c == 0x42) { /* 'B' */
297 filter->status = 0;
298 } else if (c == 0x4a) { /* 'J' */
299 filter->status = 0;
300 } else if (c == 0x49) { /* 'I' */
301 filter->status = 0x20;
302 } else {
303 filter->status &= ~0xf;
304 CK((*filter->output_function)(0x1b, filter->data));
305 CK((*filter->output_function)(0x28, filter->data));
306 goto retry;
307 }
308 break;
309
310 default:
311 filter->status = 0;
312 break;
313 }
314
315 return c;
316 }
317
318 /*
319 * wchar => ISO-2022-JP-Mobile
320 */
321 int
mbfl_filt_conv_wchar_2022jp_mobile(int c,mbfl_convert_filter * filter)322 mbfl_filt_conv_wchar_2022jp_mobile(int c, mbfl_convert_filter *filter)
323 {
324 int c1, c2, s1, s2;
325
326 s1 = 0;
327 s2 = 0;
328 if (c >= ucs_a1_jis_table_min && c < ucs_a1_jis_table_max) {
329 s1 = ucs_a1_jis_table[c - ucs_a1_jis_table_min];
330 } else if (c >= ucs_a2_jis_table_min && c < ucs_a2_jis_table_max) {
331 s1 = ucs_a2_jis_table[c - ucs_a2_jis_table_min];
332 } else if (c >= ucs_i_jis_table_min && c < ucs_i_jis_table_max) {
333 s1 = ucs_i_jis_table[c - ucs_i_jis_table_min];
334 } else if (c >= ucs_r_jis_table_min && c < ucs_r_jis_table_max) {
335 s1 = ucs_r_jis_table[c - ucs_r_jis_table_min];
336 } else if (c >= 0xe000 && c < (0xe000 + 20*94)) { /* user (95ku - 114ku) */
337 s1 = c - 0xe000;
338 c1 = s1/94 + 0x7f;
339 c2 = s1%94 + 0x21;
340 s1 = (c1 << 8) | c2;
341 }
342 if (s1 <= 0) {
343 c1 = c & ~MBFL_WCSPLANE_MASK;
344 if (c1 == MBFL_WCSPLANE_WINCP932) {
345 s1 = c & MBFL_WCSPLANE_MASK;
346 s2 = 1;
347 } else if (c1 == MBFL_WCSPLANE_JIS0208) {
348 s1 = c & MBFL_WCSPLANE_MASK;
349 } else if (c1 == MBFL_WCSPLANE_JIS0212) {
350 s1 = c & MBFL_WCSPLANE_MASK;
351 s1 |= 0x8080;
352 } else if (c == 0xa5) { /* YEN SIGN */
353 s1 = 0x216f; /* FULLWIDTH YEN SIGN */
354 } else if (c == 0x203e) { /* OVER LINE */
355 s1 = 0x2131; /* FULLWIDTH MACRON */
356 } else if (c == 0xff3c) { /* FULLWIDTH REVERSE SOLIDUS */
357 s1 = 0x2140;
358 } else if (c == 0xff5e) { /* FULLWIDTH TILDE */
359 s1 = 0x2141;
360 } else if (c == 0x2225) { /* PARALLEL TO */
361 s1 = 0x2142;
362 } else if (c == 0xff0d) { /* FULLWIDTH HYPHEN-MINUS */
363 s1 = 0x215d;
364 } else if (c == 0xffe0) { /* FULLWIDTH CENT SIGN */
365 s1 = 0x2171;
366 } else if (c == 0xffe1) { /* FULLWIDTH POUND SIGN */
367 s1 = 0x2172;
368 } else if (c == 0xffe2) { /* FULLWIDTH NOT SIGN */
369 s1 = 0x224c;
370 }
371 }
372
373 if ((s1 <= 0) || (s1 >= 0xa1a1 && s2 == 0)) { /* not found or X 0212 */
374 s1 = -1;
375 c1 = 0;
376 c2 = cp932ext1_ucs_table_max - cp932ext1_ucs_table_min;
377 while (c1 < c2) { /* CP932 vendor ext1 (13ku) */
378 if (c == cp932ext1_ucs_table[c1]) {
379 s1 = ((c1/94 + 0x2d) << 8) + (c1%94 + 0x21);
380 break;
381 }
382 c1++;
383 }
384 if (c == 0) {
385 s1 = 0;
386 } else if (s1 <= 0) {
387 s1 = -1;
388 }
389 }
390
391 if (filter->to->no_encoding == mbfl_no_encoding_2022jp_kddi &&
392 mbfilter_unicode2sjis_emoji_kddi(c, &s1, filter) > 0) {
393 CODE2JIS(c1,c2,s1,s2);
394 s1 -= 0x1600;
395 }
396
397 if (filter->status == 1 && filter->cache > 0) {
398 return c;
399 }
400
401 if (s1 >= 0) {
402 if (s1 < 0x80) { /* latin */
403 if ((filter->status & 0xff00) != 0) {
404 CK((*filter->output_function)(0x1b, filter->data)); /* ESC */
405 CK((*filter->output_function)(0x28, filter->data)); /* '(' */
406 CK((*filter->output_function)(0x42, filter->data)); /* 'B' */
407 }
408 CK((*filter->output_function)(s1, filter->data));
409 filter->status = 0;
410 } else if (s1 > 0xa0 && s1 < 0xe0) { /* kana */
411 if ((filter->status & 0xff00) != 0x100) {
412 CK((*filter->output_function)(0x1b, filter->data)); /* ESC */
413 CK((*filter->output_function)(0x28, filter->data)); /* '(' */
414 CK((*filter->output_function)(0x49, filter->data)); /* 'I' */
415 }
416 filter->status = 0x100;
417 CK((*filter->output_function)(s1 & 0x7f, filter->data));
418 } else if (s1 < 0x7e7f) { /* X 0208 */
419 if ((filter->status & 0xff00) != 0x200) {
420 CK((*filter->output_function)(0x1b, filter->data)); /* ESC */
421 CK((*filter->output_function)(0x24, filter->data)); /* '$' */
422 CK((*filter->output_function)(0x42, filter->data)); /* 'B' */
423 }
424 filter->status = 0x200;
425 CK((*filter->output_function)((s1 >> 8) & 0xff, filter->data));
426 CK((*filter->output_function)(s1 & 0x7f, filter->data));
427 }
428 } else {
429 CK(mbfl_filt_conv_illegal_output(c, filter));
430 }
431
432 return c;
433 }
434