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 20 dec 2002.
27 *
28 */
29
30 #ifdef HAVE_CONFIG_H
31 #include "config.h"
32 #endif
33
34 #include "mbfilter.h"
35 #include "mbfilter_utf32.h"
36
37 static const char *mbfl_encoding_utf32_aliases[] = {"utf32", NULL};
38
39 const mbfl_encoding mbfl_encoding_utf32 = {
40 mbfl_no_encoding_utf32,
41 "UTF-32",
42 "UTF-32",
43 (const char *(*)[])&mbfl_encoding_utf32_aliases,
44 NULL,
45 MBFL_ENCTYPE_WCS4BE
46 };
47
48 const mbfl_encoding mbfl_encoding_utf32be = {
49 mbfl_no_encoding_utf32be,
50 "UTF-32BE",
51 "UTF-32BE",
52 NULL,
53 NULL,
54 MBFL_ENCTYPE_WCS4BE
55 };
56
57 const mbfl_encoding mbfl_encoding_utf32le = {
58 mbfl_no_encoding_utf32le,
59 "UTF-32LE",
60 "UTF-32LE",
61 NULL,
62 NULL,
63 MBFL_ENCTYPE_WCS4LE
64 };
65
66 const struct mbfl_convert_vtbl vtbl_utf32_wchar = {
67 mbfl_no_encoding_utf32,
68 mbfl_no_encoding_wchar,
69 mbfl_filt_conv_common_ctor,
70 mbfl_filt_conv_common_dtor,
71 mbfl_filt_conv_utf32_wchar,
72 mbfl_filt_conv_common_flush
73 };
74
75 const struct mbfl_convert_vtbl vtbl_wchar_utf32 = {
76 mbfl_no_encoding_wchar,
77 mbfl_no_encoding_utf32,
78 mbfl_filt_conv_common_ctor,
79 mbfl_filt_conv_common_dtor,
80 mbfl_filt_conv_wchar_utf32be,
81 mbfl_filt_conv_common_flush
82 };
83
84 const struct mbfl_convert_vtbl vtbl_utf32be_wchar = {
85 mbfl_no_encoding_utf32be,
86 mbfl_no_encoding_wchar,
87 mbfl_filt_conv_common_ctor,
88 mbfl_filt_conv_common_dtor,
89 mbfl_filt_conv_utf32be_wchar,
90 mbfl_filt_conv_common_flush
91 };
92
93 const struct mbfl_convert_vtbl vtbl_wchar_utf32be = {
94 mbfl_no_encoding_wchar,
95 mbfl_no_encoding_utf32be,
96 mbfl_filt_conv_common_ctor,
97 mbfl_filt_conv_common_dtor,
98 mbfl_filt_conv_wchar_utf32be,
99 mbfl_filt_conv_common_flush
100 };
101
102 const struct mbfl_convert_vtbl vtbl_utf32le_wchar = {
103 mbfl_no_encoding_utf32le,
104 mbfl_no_encoding_wchar,
105 mbfl_filt_conv_common_ctor,
106 mbfl_filt_conv_common_dtor,
107 mbfl_filt_conv_utf32le_wchar,
108 mbfl_filt_conv_common_flush
109 };
110
111 const struct mbfl_convert_vtbl vtbl_wchar_utf32le = {
112 mbfl_no_encoding_wchar,
113 mbfl_no_encoding_utf32le,
114 mbfl_filt_conv_common_ctor,
115 mbfl_filt_conv_common_dtor,
116 mbfl_filt_conv_wchar_utf32le,
117 mbfl_filt_conv_common_flush
118 };
119
120 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
121
122 /*
123 * UTF-32 => wchar
124 */
mbfl_filt_conv_utf32_wchar(int c,mbfl_convert_filter * filter)125 int mbfl_filt_conv_utf32_wchar(int c, mbfl_convert_filter *filter)
126 {
127 int n, endian;
128
129 endian = filter->status & 0xff00;
130 switch (filter->status & 0xff) {
131 case 0:
132 if (endian) {
133 n = c & 0xff;
134 } else {
135 n = (c & 0xff) << 24;
136 }
137 filter->cache = n;
138 filter->status++;
139 break;
140 case 1:
141 if (endian) {
142 n = (c & 0xff) << 8;
143 } else {
144 n = (c & 0xff) << 16;
145 }
146 filter->cache |= n;
147 filter->status++;
148 break;
149 case 2:
150 if (endian) {
151 n = (c & 0xff) << 16;
152 } else {
153 n = (c & 0xff) << 8;
154 }
155 filter->cache |= n;
156 filter->status++;
157 break;
158 default:
159 if (endian) {
160 n = (c & 0xff) << 24;
161 } else {
162 n = c & 0xff;
163 }
164 n |= filter->cache;
165 if ((n & 0xffff) == 0 && ((n >> 16) & 0xffff) == 0xfffe) {
166 if (endian) {
167 filter->status = 0; /* big-endian */
168 } else {
169 filter->status = 0x100; /* little-endian */
170 }
171 CK((*filter->output_function)(0xfeff, filter->data));
172 } else {
173 filter->status &= ~0xff;
174 if (n < MBFL_WCSPLANE_UTF32MAX && (n < 0xd800 || n > 0xdfff)) {
175 CK((*filter->output_function)(n, filter->data));
176 } else {
177 n = (n & MBFL_WCSGROUP_MASK) | MBFL_WCSGROUP_THROUGH;
178 CK((*filter->output_function)(n, filter->data));
179 }
180 }
181 break;
182 }
183
184 return c;
185 }
186
187 /*
188 * UTF-32BE => wchar
189 */
mbfl_filt_conv_utf32be_wchar(int c,mbfl_convert_filter * filter)190 int mbfl_filt_conv_utf32be_wchar(int c, mbfl_convert_filter *filter)
191 {
192 int n;
193
194 if (filter->status == 0) {
195 filter->status = 1;
196 n = (c & 0xff) << 24;
197 filter->cache = n;
198 } else if (filter->status == 1) {
199 filter->status = 2;
200 n = (c & 0xff) << 16;
201 filter->cache |= n;
202 } else if (filter->status == 2) {
203 filter->status = 3;
204 n = (c & 0xff) << 8;
205 filter->cache |= n;
206 } else {
207 filter->status = 0;
208 n = (c & 0xff) | filter->cache;
209 if (n < MBFL_WCSPLANE_UTF32MAX && (n < 0xd800 || n > 0xdfff)) {
210 CK((*filter->output_function)(n, filter->data));
211 } else {
212 n = (n & MBFL_WCSGROUP_MASK) | MBFL_WCSGROUP_THROUGH;
213 CK((*filter->output_function)(n, filter->data));
214 }
215 }
216 return c;
217 }
218
219 /*
220 * wchar => UTF-32BE
221 */
mbfl_filt_conv_wchar_utf32be(int c,mbfl_convert_filter * filter)222 int mbfl_filt_conv_wchar_utf32be(int c, mbfl_convert_filter *filter)
223 {
224 if (c >= 0 && c < MBFL_WCSPLANE_UTF32MAX) {
225 CK((*filter->output_function)((c >> 24) & 0xff, filter->data));
226 CK((*filter->output_function)((c >> 16) & 0xff, filter->data));
227 CK((*filter->output_function)((c >> 8) & 0xff, filter->data));
228 CK((*filter->output_function)(c & 0xff, filter->data));
229 } else {
230 if (filter->illegal_mode != MBFL_OUTPUTFILTER_ILLEGAL_MODE_NONE) {
231 CK(mbfl_filt_conv_illegal_output(c, filter));
232 }
233 }
234
235 return c;
236 }
237
238 /*
239 * UTF-32LE => wchar
240 */
mbfl_filt_conv_utf32le_wchar(int c,mbfl_convert_filter * filter)241 int mbfl_filt_conv_utf32le_wchar(int c, mbfl_convert_filter *filter)
242 {
243 int n;
244
245 if (filter->status == 0) {
246 filter->status = 1;
247 n = (c & 0xff);
248 filter->cache = n;
249 } else if (filter->status == 1) {
250 filter->status = 2;
251 n = (c & 0xff) << 8;
252 filter->cache |= n;
253 } else if (filter->status == 2) {
254 filter->status = 3;
255 n = (c & 0xff) << 16;
256 filter->cache |= n;
257 } else {
258 filter->status = 0;
259 n = ((c & 0xff) << 24) | filter->cache;
260 if (n < MBFL_WCSPLANE_UTF32MAX && (n < 0xd800 || n > 0xdfff)) {
261 CK((*filter->output_function)(n, filter->data));
262 } else {
263 n = (n & MBFL_WCSGROUP_MASK) | MBFL_WCSGROUP_THROUGH;
264 CK((*filter->output_function)(n, filter->data));
265 }
266 }
267 return c;
268 }
269
270 /*
271 * wchar => UTF-32LE
272 */
mbfl_filt_conv_wchar_utf32le(int c,mbfl_convert_filter * filter)273 int mbfl_filt_conv_wchar_utf32le(int c, mbfl_convert_filter *filter)
274 {
275 if (c >= 0 && c < MBFL_WCSPLANE_UTF32MAX) {
276 CK((*filter->output_function)(c & 0xff, filter->data));
277 CK((*filter->output_function)((c >> 8) & 0xff, filter->data));
278 CK((*filter->output_function)((c >> 16) & 0xff, filter->data));
279 CK((*filter->output_function)((c >> 24) & 0xff, filter->data));
280 } else {
281 if (filter->illegal_mode != MBFL_OUTPUTFILTER_ILLEGAL_MODE_NONE) {
282 CK(mbfl_filt_conv_illegal_output(c, filter));
283 }
284 }
285
286 return c;
287 }
288
