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_cn.c
26 * by moriyoshi koizumi <moriyoshi@php.net> on 4 dec 2002.
27 *
28 */
29
30 #include "mbfilter.h"
31 #include "mbfilter_hz.h"
32
33 #include "unicode_table_cp936.h"
34 #include "unicode_table_gb2312.h"
35
36 static int mbfl_filt_conv_hz_wchar_flush(mbfl_convert_filter *filter);
37
38 const mbfl_encoding mbfl_encoding_hz = {
39 mbfl_no_encoding_hz,
40 "HZ",
41 "HZ-GB-2312",
42 NULL,
43 NULL,
44 MBFL_ENCTYPE_GL_UNSAFE,
45 &vtbl_hz_wchar,
46 &vtbl_wchar_hz,
47 NULL
48 };
49
50 const struct mbfl_convert_vtbl vtbl_hz_wchar = {
51 mbfl_no_encoding_hz,
52 mbfl_no_encoding_wchar,
53 mbfl_filt_conv_common_ctor,
54 NULL,
55 mbfl_filt_conv_hz_wchar,
56 mbfl_filt_conv_hz_wchar_flush,
57 NULL,
58 };
59
60 const struct mbfl_convert_vtbl vtbl_wchar_hz = {
61 mbfl_no_encoding_wchar,
62 mbfl_no_encoding_hz,
63 mbfl_filt_conv_common_ctor,
64 NULL,
65 mbfl_filt_conv_wchar_hz,
66 mbfl_filt_conv_any_hz_flush,
67 NULL,
68 };
69
70 #define CK(statement) do { if ((statement) < 0) return (-1); } while (0)
71
mbfl_filt_conv_hz_wchar(int c,mbfl_convert_filter * filter)72 int mbfl_filt_conv_hz_wchar(int c, mbfl_convert_filter *filter)
73 {
74 int c1, s, w;
75
76 switch (filter->status & 0xf) {
77 /* case 0x00: ASCII */
78 /* case 0x10: GB2312 */
79 case 0:
80 if (c == '~') {
81 filter->status += 2;
82 } else if (filter->status == 0x10 && ((c > 0x20 && c <= 0x29) || (c >= 0x30 && c <= 0x77))) {
83 /* DBCS first char */
84 filter->cache = c;
85 filter->status += 1;
86 } else if (filter->status == 0 && c >= 0 && c < 0x80) { /* latin, CTLs */
87 CK((*filter->output_function)(c, filter->data));
88 } else {
89 CK((*filter->output_function)(MBFL_BAD_INPUT, filter->data));
90 }
91 break;
92
93 /* case 0x11: GB2312 second char */
94 case 1:
95 filter->status &= ~0xf;
96 c1 = filter->cache;
97 if (c1 > 0x20 && c1 < 0x7F && c > 0x20 && c < 0x7F) {
98 s = (c1 - 1)*192 + c + 0x40; /* GB2312 */
99 if (s >= 0 && s < cp936_ucs_table_size) {
100 if (s == 0x1864) {
101 w = 0x30FB;
102 } else if (s == 0x186A) {
103 w = 0x2015;
104 } else if (s == 0x186C) {
105 w = 0x2225;
106 } else if ((s >= 0x1920 && s <= 0x192A) || s == 0x1963 || (s >= 0x1C60 && s <= 0x1C7F) || (s >= 0x1DBB && s <= 0x1DC4)) {
107 w = 0;
108 } else {
109 w = cp936_ucs_table[s];
110 }
111 } else {
112 w = 0;
113 }
114
115 if (w <= 0) {
116 w = MBFL_BAD_INPUT;
117 }
118
119 CK((*filter->output_function)(w, filter->data));
120 } else {
121 CK((*filter->output_function)(MBFL_BAD_INPUT, filter->data));
122 }
123 break;
124
125 /* '~' */
126 case 2:
127 if (c == '}' && filter->status == 0x12) {
128 filter->status = 0;
129 } else if (c == '{' && filter->status == 2) {
130 filter->status = 0x10;
131 } else if (c == '~' && filter->status == 2) {
132 CK((*filter->output_function)('~', filter->data));
133 } else if (c == '\n') {
134 /* "~\n" is a line continuation; no output is needed, nor should we shift modes */
135 filter->status -= 2;
136 } else {
137 /* Invalid character after ~ */
138 filter->status -= 2;
139 CK((*filter->output_function)(MBFL_BAD_INPUT, filter->data));
140 }
141 break;
142
143 default:
144 filter->status = 0;
145 break;
146 }
147
148 return 0;
149 }
150
mbfl_filt_conv_hz_wchar_flush(mbfl_convert_filter * filter)151 static int mbfl_filt_conv_hz_wchar_flush(mbfl_convert_filter *filter)
152 {
153 if (filter->status == 0x11) {
154 /* 2-byte character was truncated */
155 CK((*filter->output_function)(MBFL_BAD_INPUT, filter->data));
156 }
157
158 filter->status = 0;
159
160 if (filter->flush_function) {
161 (*filter->flush_function)(filter->data);
162 }
163
164 return 0;
165 }
166
mbfl_filt_conv_wchar_hz(int c,mbfl_convert_filter * filter)167 int mbfl_filt_conv_wchar_hz(int c, mbfl_convert_filter *filter)
168 {
169 int s = 0;
170
171 if (c >= ucs_a1_cp936_table_min && c < ucs_a1_cp936_table_max) {
172 if (c == 0xB7 || c == 0x144 || c == 0x148 || c == 0x251 || c == 0x261 || c == 0x2CA || c == 0x2CB || c == 0x2D9) {
173 s = 0;
174 } else {
175 s = ucs_a1_cp936_table[c - ucs_a1_cp936_table_min];
176 }
177 } else if (c >= ucs_a2_cp936_table_min && c < ucs_a2_cp936_table_max) {
178 if (c == 0x2015) {
179 s = 0xA1AA;
180 } else if (c == 0x2010 || c == 0x2013 || c == 0x2014 || c == 0x2016 || c == 0x2025 || c == 0x2035 ||
181 c == 0x2105 || c == 0x2109 || c == 0x2121 || (c >= 0x2170 && c <= 0x2179) || (c >= 0x2196 && c <= 0x2199) ||
182 c == 0x2215 || c == 0x221F || c == 0x2223 || c == 0x2252 || c == 0x2266 || c == 0x2267 || c == 0x2295 ||
183 (c >= 0x2550 && c <= 0x2573) || c == 0x22BF || c == 0x2609 || (c >= 0x2581 && c <= 0x258F) ||
184 (c >= 0x2593 && c <= 0x2595) || c == 0x25BC || c == 0x25BD || (c >= 0x25E2 && c <= 0x25E5)) {
185 s = 0;
186 } else {
187 s = ucs_a2_cp936_table[c - ucs_a2_cp936_table_min];
188 }
189 } else if (c >= ucs_a3_cp936_table_min && c < ucs_a3_cp936_table_max) {
190 if (c == 0x30FB) {
191 s = 0xA1A4;
192 } else if (c == 0x3006 || c == 0x3007 || c == 0x3012 || c == 0x3231 || c == 0x32A3 || c >= 0x3300 ||
193 (c >= 0x3018 && c <= 0x3040) || (c >= 0x309B && c <= 0x309E) || (c >= 0x30FC && c <= 0x30FE)) {
194 s = 0;
195 } else {
196 s = ucs_a3_cp936_table[c - ucs_a3_cp936_table_min];
197 }
198 } else if (c >= ucs_i_gb2312_table_min && c < ucs_i_gb2312_table_max) {
199 s = ucs_i_gb2312_table[c - ucs_i_gb2312_table_min];
200 } else if (c >= ucs_hff_cp936_table_min && c < ucs_hff_cp936_table_max) {
201 if (c == 0xFF04) {
202 s = 0xA1E7;
203 } else if (c == 0xFF5E) {
204 s = 0xA1AB;
205 } else if (c >= 0xFF01 && c <= 0xFF5D) {
206 s = c - 0xFF01 + 0xA3A1;
207 } else if (c == 0xFFE0 || c == 0xFFE1 || c == 0xFFE3 || c == 0xFFE5) {
208 s = ucs_hff_s_cp936_table[c - 0xFFE0];
209 }
210 }
211
212 if (s & 0x8000) {
213 s -= 0x8080;
214 }
215
216 if (s <= 0) {
217 s = (c == 0) ? 0 : -1;
218 } else if ((s >= 0x80 && s < 0x2121) || s > 0x8080) {
219 s = -1;
220 }
221
222 if (s >= 0) {
223 if (s < 0x80) { /* ASCII */
224 if ((filter->status & 0xff00) != 0) {
225 CK((*filter->output_function)('~', filter->data));
226 CK((*filter->output_function)('}', filter->data));
227 }
228 filter->status = 0;
229 if (s == 0x7E) {
230 CK((*filter->output_function)('~', filter->data));
231 }
232 CK((*filter->output_function)(s, filter->data));
233 } else { /* GB 2312-80 */
234 if ((filter->status & 0xFF00) != 0x200) {
235 CK((*filter->output_function)('~', filter->data));
236 CK((*filter->output_function)('{', filter->data));
237 }
238 filter->status = 0x200;
239 CK((*filter->output_function)((s >> 8) & 0x7F, filter->data));
240 CK((*filter->output_function)(s & 0x7F, filter->data));
241 }
242 } else {
243 CK(mbfl_filt_conv_illegal_output(c, filter));
244 }
245
246 return 0;
247 }
248
mbfl_filt_conv_any_hz_flush(mbfl_convert_filter * filter)249 int mbfl_filt_conv_any_hz_flush(mbfl_convert_filter *filter)
250 {
251 /* back to latin */
252 if (filter->status & 0xFF00) {
253 CK((*filter->output_function)('~', filter->data));
254 CK((*filter->output_function)('}', filter->data));
255 }
256 filter->status = 0;
257 return 0;
258 }
259