1 : /**********************************************************************
2 : iso8859_1.c - Oniguruma (regular expression library)
3 : **********************************************************************/
4 : /*-
5 : * Copyright (c) 2002-2005 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_IS_ISO_8859_1_CTYPE(code,ctype) \
33 : ((EncISO_8859_1_CtypeTable[code] & ctype) != 0)
34 :
35 : static const unsigned short EncISO_8859_1_CtypeTable[256] = {
36 : 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
37 : 0x2008, 0x220c, 0x2209, 0x2208, 0x2208, 0x2208, 0x2008, 0x2008,
38 : 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
39 : 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008, 0x2008,
40 : 0x2284, 0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x21a0, 0x21a0, 0x21a0,
41 : 0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x21a0, 0x21a0, 0x21a0, 0x21a0,
42 : 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0, 0x38b0,
43 : 0x38b0, 0x38b0, 0x21a0, 0x21a0, 0x20a0, 0x20a0, 0x20a0, 0x21a0,
44 : 0x21a0, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x3ca2, 0x34a2,
45 : 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
46 : 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2, 0x34a2,
47 : 0x34a2, 0x34a2, 0x34a2, 0x21a0, 0x21a0, 0x21a0, 0x20a0, 0x31a0,
48 : 0x20a0, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x38e2, 0x30e2,
49 : 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
50 : 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2, 0x30e2,
51 : 0x30e2, 0x30e2, 0x30e2, 0x21a0, 0x20a0, 0x21a0, 0x20a0, 0x2008,
52 : 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
53 : 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
54 : 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
55 : 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008, 0x0008,
56 : 0x0284, 0x01a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0, 0x00a0,
57 : 0x00a0, 0x00a0, 0x10e2, 0x01a0, 0x00a0, 0x01a0, 0x00a0, 0x00a0,
58 : 0x00a0, 0x00a0, 0x10a0, 0x10a0, 0x00a0, 0x10e2, 0x00a0, 0x01a0,
59 : 0x00a0, 0x10a0, 0x10e2, 0x01a0, 0x10a0, 0x10a0, 0x10a0, 0x01a0,
60 : 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2,
61 : 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2,
62 : 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x00a0,
63 : 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x14a2, 0x10e2,
64 : 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2,
65 : 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2,
66 : 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x00a0,
67 : 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2, 0x10e2
68 : };
69 :
70 : static int
71 : iso_8859_1_mbc_to_normalize(OnigAmbigType flag, const UChar** pp, const UChar* end, UChar* lower)
72 6 : {
73 6 : const UChar* p = *pp;
74 :
75 6 : if (end > p + 1 && (flag & ONIGENC_AMBIGUOUS_MATCH_COMPOUND) != 0) {
76 0 : if ((*p == 's' && *(p+1) == 's') ||
77 : ((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
78 : (*p == 'S' && *(p+1) == 'S'))) {
79 0 : *lower = 0xdf;
80 0 : (*pp) += 2;
81 0 : return 1;
82 : }
83 : }
84 :
85 12 : if (((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
86 : ONIGENC_IS_MBC_ASCII(p)) ||
87 : ((flag & ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE) != 0 &&
88 : !ONIGENC_IS_MBC_ASCII(p))) {
89 6 : *lower = ONIGENC_ISO_8859_1_TO_LOWER_CASE(*p);
90 : }
91 : else {
92 0 : *lower = *p;
93 : }
94 6 : (*pp)++;
95 6 : return 1; /* return byte length of converted char to lower */
96 : }
97 :
98 : static int
99 : iso_8859_1_is_mbc_ambiguous(OnigAmbigType flag,
100 : const UChar** pp, const UChar* end)
101 2 : {
102 2 : const UChar* p = *pp;
103 :
104 2 : if ((flag & ONIGENC_AMBIGUOUS_MATCH_COMPOUND) != 0) {
105 0 : if (end > p + 1) {
106 0 : if ((*p == 's' && *(p+1) == 's') ||
107 : ((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
108 : (*p == 'S' && *(p+1) == 'S'))) {
109 0 : (*pp) += 2;
110 0 : return TRUE;
111 : }
112 : }
113 :
114 0 : if (*p == 0xdf) {
115 0 : (*pp)++;
116 0 : return TRUE;
117 : }
118 : }
119 :
120 2 : (*pp)++;
121 2 : if (((flag & ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE) != 0 &&
122 : ONIGENC_IS_MBC_ASCII(p)) ||
123 : ((flag & ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE) != 0 &&
124 : !ONIGENC_IS_MBC_ASCII(p))) {
125 : int v = (EncISO_8859_1_CtypeTable[*p] &
126 2 : (ONIGENC_CTYPE_UPPER | ONIGENC_CTYPE_LOWER));
127 :
128 : if ((v | ONIGENC_CTYPE_LOWER) != 0) {
129 : /* 0xdf, 0xaa, 0xb5, 0xba are lower case letter, but can't convert. */
130 2 : if (*p == 0xdf || (*p >= 0xaa && *p <= 0xba))
131 0 : return FALSE;
132 : else
133 2 : return TRUE;
134 : }
135 :
136 : return (v != 0 ? TRUE : FALSE);
137 : }
138 0 : return FALSE;
139 : }
140 :
141 : static int
142 : iso_8859_1_is_code_ctype(OnigCodePoint code, unsigned int ctype)
143 256 : {
144 256 : if (code < 256)
145 256 : return ENC_IS_ISO_8859_1_CTYPE(code, ctype);
146 : else
147 0 : return FALSE;
148 : }
149 :
150 : OnigEncodingType OnigEncodingISO_8859_1 = {
151 : onigenc_single_byte_mbc_enc_len,
152 : "ISO-8859-1", /* name */
153 : 1, /* max enc length */
154 : 1, /* min enc length */
155 : (ONIGENC_AMBIGUOUS_MATCH_ASCII_CASE |
156 : ONIGENC_AMBIGUOUS_MATCH_NONASCII_CASE |
157 : ONIGENC_AMBIGUOUS_MATCH_COMPOUND),
158 : {
159 : (OnigCodePoint )'\\' /* esc */
160 : , (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anychar '.' */
161 : , (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anytime '*' */
162 : , (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* zero or one time '?' */
163 : , (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* one or more time '+' */
164 : , (OnigCodePoint )ONIG_INEFFECTIVE_META_CHAR /* anychar anytime */
165 : },
166 : onigenc_is_mbc_newline_0x0a,
167 : onigenc_single_byte_mbc_to_code,
168 : onigenc_single_byte_code_to_mbclen,
169 : onigenc_single_byte_code_to_mbc,
170 : iso_8859_1_mbc_to_normalize,
171 : iso_8859_1_is_mbc_ambiguous,
172 : onigenc_iso_8859_1_get_all_pair_ambig_codes,
173 : onigenc_ess_tsett_get_all_comp_ambig_codes,
174 : iso_8859_1_is_code_ctype,
175 : onigenc_not_support_get_ctype_code_range,
176 : onigenc_single_byte_left_adjust_char_head,
177 : onigenc_always_true_is_allowed_reverse_match
178 : };
|
