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.c
26 * by moriyoshi koizumi <moriyoshi@php.net> on 4 dec 2002.
27 *
28 */
29
30 #include "mbfilter.h"
31 #include "mbfilter_utf16.h"
32
33 static const char *mbfl_encoding_utf16_aliases[] = {"utf16", NULL};
34
35 const mbfl_encoding mbfl_encoding_utf16 = {
36 mbfl_no_encoding_utf16,
37 "UTF-16",
38 "UTF-16",
39 (const char *(*)[])&mbfl_encoding_utf16_aliases,
40 NULL,
41 MBFL_ENCTYPE_MWC2BE,
42 &vtbl_utf16_wchar,
43 &vtbl_wchar_utf16
44 };
45
46 const mbfl_encoding mbfl_encoding_utf16be = {
47 mbfl_no_encoding_utf16be,
48 "UTF-16BE",
49 "UTF-16BE",
50 NULL,
51 NULL,
52 MBFL_ENCTYPE_MWC2BE,
53 &vtbl_utf16be_wchar,
54 &vtbl_wchar_utf16be
55 };
56
57 const mbfl_encoding mbfl_encoding_utf16le = {
58 mbfl_no_encoding_utf16le,
59 "UTF-16LE",
60 "UTF-16LE",
61 NULL,
62 NULL,
63 MBFL_ENCTYPE_MWC2LE,
64 &vtbl_utf16le_wchar,
65 &vtbl_wchar_utf16le
66 };
67
68 const struct mbfl_convert_vtbl vtbl_utf16_wchar = {
69 mbfl_no_encoding_utf16,
70 mbfl_no_encoding_wchar,
71 mbfl_filt_conv_common_ctor,
72 NULL,
73 mbfl_filt_conv_utf16_wchar,
74 mbfl_filt_conv_common_flush,
75 NULL,
76 };
77
78 const struct mbfl_convert_vtbl vtbl_wchar_utf16 = {
79 mbfl_no_encoding_wchar,
80 mbfl_no_encoding_utf16,
81 mbfl_filt_conv_common_ctor,
82 NULL,
83 mbfl_filt_conv_wchar_utf16be,
84 mbfl_filt_conv_common_flush,
85 NULL,
86 };
87
88 const struct mbfl_convert_vtbl vtbl_utf16be_wchar = {
89 mbfl_no_encoding_utf16be,
90 mbfl_no_encoding_wchar,
91 mbfl_filt_conv_common_ctor,
92 NULL,
93 mbfl_filt_conv_utf16be_wchar,
94 mbfl_filt_conv_common_flush,
95 NULL,
96 };
97
98 const struct mbfl_convert_vtbl vtbl_wchar_utf16be = {
99 mbfl_no_encoding_wchar,
100 mbfl_no_encoding_utf16be,
101 mbfl_filt_conv_common_ctor,
102 NULL,
103 mbfl_filt_conv_wchar_utf16be,
104 mbfl_filt_conv_common_flush,
105 NULL,
106 };
107
108 const struct mbfl_convert_vtbl vtbl_utf16le_wchar = {
109 mbfl_no_encoding_utf16le,
110 mbfl_no_encoding_wchar,
111 mbfl_filt_conv_common_ctor,
112 NULL,
113 mbfl_filt_conv_utf16le_wchar,
114 mbfl_filt_conv_common_flush,
115 NULL,
116 };
117
118 const struct mbfl_convert_vtbl vtbl_wchar_utf16le = {
119 mbfl_no_encoding_wchar,
120 mbfl_no_encoding_utf16le,
121 mbfl_filt_conv_common_ctor,
122 NULL,
123 mbfl_filt_conv_wchar_utf16le,
124 mbfl_filt_conv_common_flush,
125 NULL,
126 };
127
128 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
129
130 /*
131 * UTF-16 => wchar
132 */
mbfl_filt_conv_utf16_wchar(int c,mbfl_convert_filter * filter)133 int mbfl_filt_conv_utf16_wchar(int c, mbfl_convert_filter *filter)
134 {
135 int n, endian;
136
137 endian = filter->status & 0xff00;
138 switch (filter->status & 0x0f) {
139 case 0:
140 if (endian) {
141 n = c & 0xff;
142 } else {
143 n = (c & 0xff) << 8;
144 }
145 filter->cache |= n;
146 filter->status++;
147 break;
148 default:
149 if (endian) {
150 n = (c & 0xff) << 8;
151 } else {
152 n = c & 0xff;
153 }
154 n |= filter->cache & 0xffff;
155 filter->status &= ~0x0f;
156 if (n >= 0xd800 && n < 0xdc00) {
157 filter->cache = ((n & 0x3ff) << 16) + 0x400000;
158 } else if (n >= 0xdc00 && n < 0xe000) {
159 n &= 0x3ff;
160 n |= (filter->cache & 0xfff0000) >> 6;
161 filter->cache = 0;
162 if (n >= MBFL_WCSPLANE_SUPMIN && n < MBFL_WCSPLANE_SUPMAX) {
163 CK((*filter->output_function)(n, filter->data));
164 } else { /* illegal character */
165 n &= MBFL_WCSGROUP_MASK;
166 n |= MBFL_WCSGROUP_THROUGH;
167 CK((*filter->output_function)(n, filter->data));
168 }
169 } else {
170 int is_first = filter->status & 0x10;
171 filter->cache = 0;
172 filter->status |= 0x10;
173 if (!is_first) {
174 if (n == 0xfffe) {
175 if (endian) {
176 filter->status &= ~0x100; /* big-endian */
177 } else {
178 filter->status |= 0x100; /* little-endian */
179 }
180 break;
181 } else if (n == 0xfeff) {
182 break;
183 }
184 }
185 CK((*filter->output_function)(n, filter->data));
186 }
187 break;
188 }
189
190 return c;
191 }
192
193 /*
194 * UTF-16BE => wchar
195 */
mbfl_filt_conv_utf16be_wchar(int c,mbfl_convert_filter * filter)196 int mbfl_filt_conv_utf16be_wchar(int c, mbfl_convert_filter *filter)
197 {
198 int n;
199
200 switch (filter->status) {
201 case 0:
202 filter->status = 1;
203 n = (c & 0xff) << 8;
204 filter->cache |= n;
205 break;
206 default:
207 filter->status = 0;
208 n = (filter->cache & 0xff00) | (c & 0xff);
209 if (n >= 0xd800 && n < 0xdc00) {
210 filter->cache = ((n & 0x3ff) << 16) + 0x400000;
211 } else if (n >= 0xdc00 && n < 0xe000) {
212 n &= 0x3ff;
213 n |= (filter->cache & 0xfff0000) >> 6;
214 filter->cache = 0;
215 if (n >= MBFL_WCSPLANE_SUPMIN && n < MBFL_WCSPLANE_SUPMAX) {
216 CK((*filter->output_function)(n, filter->data));
217 } else { /* illegal character */
218 n &= MBFL_WCSGROUP_MASK;
219 n |= MBFL_WCSGROUP_THROUGH;
220 CK((*filter->output_function)(n, filter->data));
221 }
222 } else {
223 filter->cache = 0;
224 CK((*filter->output_function)(n, filter->data));
225 }
226 break;
227 }
228
229 return c;
230 }
231
232 /*
233 * wchar => UTF-16BE
234 */
mbfl_filt_conv_wchar_utf16be(int c,mbfl_convert_filter * filter)235 int mbfl_filt_conv_wchar_utf16be(int c, mbfl_convert_filter *filter)
236 {
237 int n;
238
239 if (c >= 0 && c < MBFL_WCSPLANE_UCS2MAX) {
240 CK((*filter->output_function)((c >> 8) & 0xff, filter->data));
241 CK((*filter->output_function)(c & 0xff, filter->data));
242 } else if (c >= MBFL_WCSPLANE_SUPMIN && c < MBFL_WCSPLANE_SUPMAX) {
243 n = ((c >> 10) - 0x40) | 0xd800;
244 CK((*filter->output_function)((n >> 8) & 0xff, filter->data));
245 CK((*filter->output_function)(n & 0xff, filter->data));
246 n = (c & 0x3ff) | 0xdc00;
247 CK((*filter->output_function)((n >> 8) & 0xff, filter->data));
248 CK((*filter->output_function)(n & 0xff, filter->data));
249 } else {
250 CK(mbfl_filt_conv_illegal_output(c, filter));
251 }
252
253 return c;
254 }
255
256 /*
257 * UTF-16LE => wchar
258 */
mbfl_filt_conv_utf16le_wchar(int c,mbfl_convert_filter * filter)259 int mbfl_filt_conv_utf16le_wchar(int c, mbfl_convert_filter *filter)
260 {
261 int n;
262
263 switch (filter->status) {
264 case 0:
265 filter->status = 1;
266 n = c & 0xff;
267 filter->cache |= n;
268 break;
269 default:
270 filter->status = 0;
271 n = (filter->cache & 0xff) | ((c & 0xff) << 8);
272 if (n >= 0xd800 && n < 0xdc00) {
273 filter->cache = ((n & 0x3ff) << 16) + 0x400000;
274 } else if (n >= 0xdc00 && n < 0xe000) {
275 n &= 0x3ff;
276 n |= (filter->cache & 0xfff0000) >> 6;
277 filter->cache = 0;
278 if (n >= MBFL_WCSPLANE_SUPMIN && n < MBFL_WCSPLANE_SUPMAX) {
279 CK((*filter->output_function)(n, filter->data));
280 } else { /* illegal character */
281 n &= MBFL_WCSGROUP_MASK;
282 n |= MBFL_WCSGROUP_THROUGH;
283 CK((*filter->output_function)(n, filter->data));
284 }
285 } else {
286 filter->cache = 0;
287 CK((*filter->output_function)(n, filter->data));
288 }
289 break;
290 }
291
292 return c;
293 }
294
295 /*
296 * wchar => UTF-16LE
297 */
mbfl_filt_conv_wchar_utf16le(int c,mbfl_convert_filter * filter)298 int mbfl_filt_conv_wchar_utf16le(int c, mbfl_convert_filter *filter)
299 {
300 int n;
301
302 if (c >= 0 && c < MBFL_WCSPLANE_UCS2MAX) {
303 CK((*filter->output_function)(c & 0xff, filter->data));
304 CK((*filter->output_function)((c >> 8) & 0xff, filter->data));
305 } else if (c >= MBFL_WCSPLANE_SUPMIN && c < MBFL_WCSPLANE_SUPMAX) {
306 n = ((c >> 10) - 0x40) | 0xd800;
307 CK((*filter->output_function)(n & 0xff, filter->data));
308 CK((*filter->output_function)((n >> 8) & 0xff, filter->data));
309 n = (c & 0x3ff) | 0xdc00;
310 CK((*filter->output_function)(n & 0xff, filter->data));
311 CK((*filter->output_function)((n >> 8) & 0xff, filter->data));
312 } else {
313 CK(mbfl_filt_conv_illegal_output(c, filter));
314 }
315
316 return c;
317 }
318
