1
/**********************************************************************
2
iso8859_13.c - Oniguruma (regular expression library)
3
**********************************************************************/
5
* Copyright (c) 2002-2005 K.Kosako <sndgk393 AT ybb DOT ne DOT jp>
8
* Redistribution and use in source and binary forms, with or without
9
* modification, are permitted provided that the following conditions
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.
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
32
#define ENC_ISO_8859_13_TO_LOWER_CASE(c) EncISO_8859_13_ToLowerCaseTable[c]
33
#define ENC_IS_ISO_8859_13_CTYPE(code,ctype) \
34
((EncISO_8859_13_CtypeTable[code] & ctype) != 0)
36
static UChar EncISO_8859_13_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', '\241', '\242', '\243', '\244', '\245', '\246', '\247',
58
'\270', '\251', '\272', '\253', '\254', '\255', '\256', '\277',
59
'\260', '\261', '\262', '\263', '\264', '\265', '\266', '\267',
60
'\270', '\271', '\272', '\273', '\274', '\275', '\276', '\277',
61
'\340', '\341', '\342', '\343', '\344', '\345', '\346', '\347',
62
'\350', '\351', '\352', '\353', '\354', '\355', '\356', '\357',
63
'\360', '\361', '\362', '\363', '\364', '\365', '\366', '\327',
64
'\370', '\371', '\372', '\373', '\374', '\375', '\376', '\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'
71
static unsigned short EncISO_8859_13_CtypeTable[256] = {
72
0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
73
0x2008, 0x220c, 0x2209, 0x2208, 0x2208, 0x2208, 0x2008, 0x2008,
74
0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
75
0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
76
0x2284, 0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x21a0, 0x21a0, 0x21a0,
77
0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x21a0, 0x21a0, 0x21a0, 0x21a0,
78
0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0,
79
0x38b0, 0x38b0, 0x21a0, 0x21a0, 0x20a0, 0x20a0, 0x20a0, 0x21a0,
80
0x21a0, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x34a2,
81
0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
82
0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
83
0x34a2, 0x34a2, 0x34a2, 0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x31a0,
84
0x20a0, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x30e2,
85
0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
86
0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
87
0x30e2, 0x30e2, 0x30e2, 0x21a0, 0x20a0, 0x21a0, 0x20a0, 0x2008,
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, 0x01a0, 0x00a0, 0x00a0, 0x00a0, 0x01a0, 0x00a0, 0x00a0,
93
0x14a2, 0x00a0, 0x14a2, 0x01a0, 0x00a0, 0x01a0, 0x00a0, 0x14a2,
94
0x00a0, 0x00a0, 0x10a0, 0x10a0, 0x01a0, 0x10e2, 0x00a0, 0x01a0,
95
0x10e2, 0x10a0, 0x10e2, 0x01a0, 0x10a0, 0x10a0, 0x10a0, 0x10e2,
96
0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2,
97
0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2,
98
0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x00a0,
99
0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x10e2,
100
0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2,
101
0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2,
102
0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x00a0,
103
0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x01a0
107
iso_8859_13_mbc_to_normalize(OnigAmbigType flag,
108
const UChar** pp, const UChar* end, UChar* lower)
110
const UChar* p = *pp;
112
if (end > p + 1 && (flag & ONIGENC_AMBIGUOUS_MATCH_COMPOUND) != 0) {
113
if ((*p == 's' && *(p+1) == 's') ||
114
((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
115
(*p == 'S' && *(p+1) == 'S'))) {
122
if (((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
123
ONIGENC_IS_MBC_ASCII(p)) ||
124
((flag & ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE) != 0 &&
125
!ONIGENC_IS_MBC_ASCII(p))) {
126
*lower = ENC_ISO_8859_13_TO_LOWER_CASE(*p);
132
return 1; /* return byte length of converted char to lower */
136
iso_8859_13_is_mbc_ambiguous(OnigAmbigType flag,
137
const UChar** pp, const UChar* end)
139
const UChar* p = *pp;
141
if ((flag & ONIGENC_AMBIGUOUS_MATCH_COMPOUND) != 0) {
143
if ((*p == 's' && *(p+1) == 's') ||
144
((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
145
(*p == 'S' && *(p+1) == 'S'))) {
158
if (((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
159
ONIGENC_IS_MBC_ASCII(p)) ||
160
((flag & ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE) != 0 &&
161
!ONIGENC_IS_MBC_ASCII(p))) {
162
int v = (EncISO_8859_13_CtypeTable[*p] &
163
(ONIGENC_CTYPE_UPPER | ONIGENC_CTYPE_LOWER));
165
if ((v | ONIGENC_CTYPE_LOWER) != 0) {
166
/* 0xdf, 0xb5 are lower case letter, but can't convert. */
167
if (*p == 0xdf || *p == 0xb5)
173
return (v != 0 ? TRUE : FALSE);
179
iso_8859_13_is_code_ctype(OnigCodePoint code, unsigned int ctype)
182
return ENC_IS_ISO_8859_13_CTYPE(code, ctype);
188
iso_8859_13_get_all_pair_ambig_codes(OnigAmbigType flag,
189
OnigPairAmbigCodes** ccs)
191
static OnigPairAmbigCodes cc[] = {
257
if (flag == ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) {
258
*ccs = OnigAsciiPairAmbigCodes;
261
if (flag == ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE) {
263
return sizeof(cc) / sizeof(OnigPairAmbigCodes);
269
OnigEncodingType OnigEncodingISO_8859_13 = {
270
onigenc_single_byte_mbc_enc_len,
271
"ISO-8859-13", /* name */
272
1, /* max enc length */
273
1, /* min enc length */
274
(ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE |
275
ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE |
276
ONIGENC_AMBIGUOUS_MATCH_COMPOUND),
278
(OnigCodePoint )'\\' /* esc */
279
, (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anychar '.' */
280
, (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anytime '*' */
281
, (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* zero or one time '?' */
282
, (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* one or more time '+' */
283
, (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anychar anytime */
285
onigenc_is_mbc_newline_0x0a,
286
onigenc_single_byte_mbc_to_code,
287
onigenc_single_byte_code_to_mbclen,
288
onigenc_single_byte_code_to_mbc,
289
iso_8859_13_mbc_to_normalize,
290
iso_8859_13_is_mbc_ambiguous,
291
iso_8859_13_get_all_pair_ambig_codes,
292
onigenc_ess_tsett_get_all_comp_ambig_codes,
293
iso_8859_13_is_code_ctype,
294
onigenc_not_support_get_ctype_code_range,
295
onigenc_single_byte_left_adjust_char_head,
296
onigenc_always_true_is_allowed_reverse_match