1 /**********************************************************************
2 iso8859_5.c - Oniguruma (regular expression library)
3 **********************************************************************/
4 /*-
5 * Copyright (c) 2002-2008 K.Kosako <sndgk393 AT ybb DOT ne DOT jp>
6 * All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in the
15 * documentation and/or other materials provided with the distribution.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27 * SUCH DAMAGE.
28 */
29
30 #include "regenc.h"
31
32 #define ENC_ISO_8859_5_TO_LOWER_CASE(c) EncISO_8859_5_ToLowerCaseTable[c]
33 #define ENC_IS_ISO_8859_5_CTYPE(code,ctype) \
34 ((EncISO_8859_5_CtypeTable[code] & CTYPE_TO_BIT(ctype)) != 0)
35
36 static const UChar EncISO_8859_5_ToLowerCaseTable[256] = {
37 '\000', '\001', '\002', '\003', '\004', '\005', '\006', '\007',
38 '\010', '\011', '\012', '\013', '\014', '\015', '\016', '\017',
39 '\020', '\021', '\022', '\023', '\024', '\025', '\026', '\027',
40 '\030', '\031', '\032', '\033', '\034', '\035', '\036', '\037',
41 '\040', '\041', '\042', '\043', '\044', '\045', '\046', '\047',
42 '\050', '\051', '\052', '\053', '\054', '\055', '\056', '\057',
43 '\060', '\061', '\062', '\063', '\064', '\065', '\066', '\067',
44 '\070', '\071', '\072', '\073', '\074', '\075', '\076', '\077',
45 '\100', '\141', '\142', '\143', '\144', '\145', '\146', '\147',
46 '\150', '\151', '\152', '\153', '\154', '\155', '\156', '\157',
47 '\160', '\161', '\162', '\163', '\164', '\165', '\166', '\167',
48 '\170', '\171', '\172', '\133', '\134', '\135', '\136', '\137',
49 '\140', '\141', '\142', '\143', '\144', '\145', '\146', '\147',
50 '\150', '\151', '\152', '\153', '\154', '\155', '\156', '\157',
51 '\160', '\161', '\162', '\163', '\164', '\165', '\166', '\167',
52 '\170', '\171', '\172', '\173', '\174', '\175', '\176', '\177',
53 '\200', '\201', '\202', '\203', '\204', '\205', '\206', '\207',
54 '\210', '\211', '\212', '\213', '\214', '\215', '\216', '\217',
55 '\220', '\221', '\222', '\223', '\224', '\225', '\226', '\227',
56 '\230', '\231', '\232', '\233', '\234', '\235', '\236', '\237',
57 '\240', '\361', '\362', '\363', '\364', '\365', '\366', '\367',
58 '\370', '\371', '\372', '\373', '\374', '\255', '\376', '\377',
59 '\320', '\321', '\322', '\323', '\324', '\325', '\326', '\327',
60 '\330', '\331', '\332', '\333', '\334', '\335', '\336', '\337',
61 '\340', '\341', '\342', '\343', '\344', '\345', '\346', '\347',
62 '\350', '\351', '\352', '\353', '\354', '\355', '\356', '\357',
63 '\320', '\321', '\322', '\323', '\324', '\325', '\326', '\327',
64 '\330', '\331', '\332', '\333', '\334', '\335', '\336', '\337',
65 '\340', '\341', '\342', '\343', '\344', '\345', '\346', '\347',
66 '\350', '\351', '\352', '\353', '\354', '\355', '\356', '\357',
67 '\360', '\361', '\362', '\363', '\364', '\365', '\366', '\367',
68 '\370', '\371', '\372', '\373', '\374', '\375', '\376', '\377'
69 };
70
71 static const unsigned short EncISO_8859_5_CtypeTable[256] = {
72 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
73 0x4008, 0x420c, 0x4209, 0x4208, 0x4208, 0x4208, 0x4008, 0x4008,
74 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
75 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008, 0x4008,
76 0x4284, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
77 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
78 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0, 0x78b0,
79 0x78b0, 0x78b0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x41a0,
80 0x41a0, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x7ca2, 0x74a2,
81 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2,
82 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2, 0x74a2,
83 0x74a2, 0x74a2, 0x74a2, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x51a0,
84 0x41a0, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x78e2, 0x70e2,
85 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2,
86 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2, 0x70e2,
87 0x70e2, 0x70e2, 0x70e2, 0x41a0, 0x41a0, 0x41a0, 0x41a0, 0x4008,
88 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
89 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
90 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
91 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
92 0x0284, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
93 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x01a0, 0x34a2, 0x34a2,
94 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
95 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
96 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
97 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
98 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
99 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
100 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
101 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
102 0x00a0, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
103 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x00a0, 0x30e2, 0x30e2
104 };
105
106 static int
mbc_case_fold(OnigCaseFoldType flag,const UChar ** pp,const UChar * end ARG_UNUSED,UChar * lower,OnigEncoding enc ARG_UNUSED)107 mbc_case_fold(OnigCaseFoldType flag,
108 const UChar** pp, const UChar* end ARG_UNUSED, UChar* lower,
109 OnigEncoding enc ARG_UNUSED)
110 {
111 const UChar* p = *pp;
112
113 *lower = ENC_ISO_8859_5_TO_LOWER_CASE(*p);
114 (*pp)++;
115 return 1;
116 }
117
118 #if 0
119 static int
120 is_mbc_ambiguous(OnigCaseFoldType flag, const UChar** pp, const UChar* end)
121 {
122 int v;
123 const UChar* p = *pp;
124
125 (*pp)++;
126 v = (EncISO_8859_5_CtypeTable[*p] & (BIT_CTYPE_UPPER | BIT_CTYPE_LOWER));
127 return (v != 0 ? TRUE : FALSE);
128 }
129 #endif
130
131 static int
is_code_ctype(OnigCodePoint code,unsigned int ctype,OnigEncoding enc ARG_UNUSED)132 is_code_ctype(OnigCodePoint code, unsigned int ctype, OnigEncoding enc ARG_UNUSED)
133 {
134 if (code < 256)
135 return ENC_IS_ISO_8859_5_CTYPE(code, ctype);
136 else
137 return FALSE;
138 }
139
140 static const OnigPairCaseFoldCodes CaseFoldMap[] = {
141 { 0xa1, 0xf1 },
142 { 0xa2, 0xf2 },
143 { 0xa3, 0xf3 },
144 { 0xa4, 0xf4 },
145 { 0xa5, 0xf5 },
146 { 0xa6, 0xf6 },
147 { 0xa7, 0xf7 },
148 { 0xa8, 0xf8 },
149 { 0xa9, 0xf9 },
150 { 0xaa, 0xfa },
151 { 0xab, 0xfb },
152 { 0xac, 0xfc },
153 { 0xae, 0xfe },
154 { 0xaf, 0xff },
155
156 { 0xb0, 0xd0 },
157 { 0xb1, 0xd1 },
158 { 0xb2, 0xd2 },
159 { 0xb3, 0xd3 },
160 { 0xb4, 0xd4 },
161 { 0xb5, 0xd5 },
162 { 0xb6, 0xd6 },
163 { 0xb7, 0xd7 },
164 { 0xb8, 0xd8 },
165 { 0xb9, 0xd9 },
166 { 0xba, 0xda },
167 { 0xbb, 0xdb },
168 { 0xbc, 0xdc },
169 { 0xbd, 0xdd },
170 { 0xbe, 0xde },
171 { 0xbf, 0xdf },
172
173 { 0xc0, 0xe0 },
174 { 0xc1, 0xe1 },
175 { 0xc2, 0xe2 },
176 { 0xc3, 0xe3 },
177 { 0xc4, 0xe4 },
178 { 0xc5, 0xe5 },
179 { 0xc6, 0xe6 },
180 { 0xc7, 0xe7 },
181 { 0xc8, 0xe8 },
182 { 0xc9, 0xe9 },
183 { 0xca, 0xea },
184 { 0xcb, 0xeb },
185 { 0xcc, 0xec },
186 { 0xcd, 0xed },
187 { 0xce, 0xee },
188 { 0xcf, 0xef }
189 };
190
191 static int
apply_all_case_fold(OnigCaseFoldType flag,OnigApplyAllCaseFoldFunc f,void * arg,OnigEncoding enc ARG_UNUSED)192 apply_all_case_fold(OnigCaseFoldType flag,
193 OnigApplyAllCaseFoldFunc f, void* arg,
194 OnigEncoding enc ARG_UNUSED)
195 {
196 return onigenc_apply_all_case_fold_with_map(
197 numberof(CaseFoldMap), CaseFoldMap, 0,
198 flag, f, arg);
199 }
200
201 static int
get_case_fold_codes_by_str(OnigCaseFoldType flag,const OnigUChar * p,const OnigUChar * end,OnigCaseFoldCodeItem items[],OnigEncoding enc ARG_UNUSED)202 get_case_fold_codes_by_str(OnigCaseFoldType flag,
203 const OnigUChar* p, const OnigUChar* end,
204 OnigCaseFoldCodeItem items[],
205 OnigEncoding enc ARG_UNUSED)
206 {
207 return onigenc_get_case_fold_codes_by_str_with_map(
208 numberof(CaseFoldMap), CaseFoldMap, 0,
209 flag, p, end, items);
210 }
211
212 static int
case_map(OnigCaseFoldType * flagP,const OnigUChar ** pp,const OnigUChar * end,OnigUChar * to,OnigUChar * to_end,const struct OnigEncodingTypeST * enc)213 case_map(OnigCaseFoldType* flagP, const OnigUChar** pp,
214 const OnigUChar* end, OnigUChar* to, OnigUChar* to_end,
215 const struct OnigEncodingTypeST* enc)
216 {
217 OnigCodePoint code;
218 OnigUChar *to_start = to;
219 OnigCaseFoldType flags = *flagP;
220
221 while (*pp < end && to < to_end) {
222 code = *(*pp)++;
223 if ((EncISO_8859_5_CtypeTable[code] & BIT_CTYPE_UPPER)
224 && (flags & (ONIGENC_CASE_DOWNCASE | ONIGENC_CASE_FOLD))) {
225 flags |= ONIGENC_CASE_MODIFIED;
226 code = ENC_ISO_8859_5_TO_LOWER_CASE(code);
227 }
228 else if ((EncISO_8859_5_CtypeTable[code]&BIT_CTYPE_LOWER)
229 && (flags & ONIGENC_CASE_UPCASE)) {
230 flags |= ONIGENC_CASE_MODIFIED;
231 if (0xF1 <= code && code <= 0xFF)
232 code -= 0x50;
233 else
234 code -= 0x20;
235 }
236 *to++ = code;
237 if (flags & ONIGENC_CASE_TITLECASE) /* switch from titlecase to lowercase for capitalize */
238 flags ^= (ONIGENC_CASE_UPCASE | ONIGENC_CASE_DOWNCASE | ONIGENC_CASE_TITLECASE);
239 }
240 *flagP = flags;
241 return (int )(to - to_start);
242 }
243
244 OnigEncodingDefine(iso_8859_5, ISO_8859_5) = {
245 onigenc_single_byte_mbc_enc_len,
246 "ISO-8859-5", /* name */
247 1, /* max enc length */
248 1, /* min enc length */
249 onigenc_is_mbc_newline_0x0a,
250 onigenc_single_byte_mbc_to_code,
251 onigenc_single_byte_code_to_mbclen,
252 onigenc_single_byte_code_to_mbc,
253 mbc_case_fold,
254 apply_all_case_fold,
255 get_case_fold_codes_by_str,
256 onigenc_minimum_property_name_to_ctype,
257 is_code_ctype,
258 onigenc_not_support_get_ctype_code_range,
259 onigenc_single_byte_left_adjust_char_head,
260 onigenc_always_true_is_allowed_reverse_match,
261 case_map,
262 0,
263 ONIGENC_FLAG_NONE,
264 };
265 ENC_ALIAS("ISO8859-5", "ISO-8859-5")
266