1 /**********************************************************************
2   euc_jp.c -  Oniguruma (regular expression library)
3 **********************************************************************/
4 /*-
5  * Copyright (c) 2002-2018  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 "regint.h"
31 
32 #define eucjp_islead(c)    ((UChar )((c) - 0xa1) > 0xfe - 0xa1)
33 
34 static const int EncLen_EUCJP[] = {
35   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
36   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
37   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
38   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
39   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
40   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
41   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
42   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
43   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 3,
44   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
45   1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
46   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
47   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
48   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
49   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
50   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1
51 };
52 
53 static int
mbc_enc_len(const UChar * p)54 mbc_enc_len(const UChar* p)
55 {
56   return EncLen_EUCJP[*p];
57 }
58 
59 static int
is_valid_mbc_string(const UChar * p,const UChar * end)60 is_valid_mbc_string(const UChar* p, const UChar* end)
61 {
62   while (p < end) {
63     if (*p < 0x80) {
64       p++;
65     }
66     else if (*p > 0xa0) {
67       if (*p == 0xff) return FALSE;
68       p++;
69       if (p >= end) return FALSE;
70       if (*p < 0xa1 || *p == 0xff) return FALSE;
71       p++;
72     }
73     else if (*p == 0x8e) {
74       p++;
75       if (p >= end) return FALSE;
76       if (*p < 0xa1 || *p > 0xdf) return FALSE;
77       p++;
78     }
79     else if (*p == 0x8f) {
80       p++;
81       if (p >= end) return FALSE;
82       if (*p < 0xa1 || *p == 0xff) return FALSE;
83       p++;
84       if (p >= end) return FALSE;
85       if (*p < 0xa1 || *p == 0xff) return FALSE;
86       p++;
87     }
88     else
89       return FALSE;
90   }
91 
92   return TRUE;
93 }
94 
95 static OnigCodePoint
mbc_to_code(const UChar * p,const UChar * end)96 mbc_to_code(const UChar* p, const UChar* end)
97 {
98   int c, i, len;
99   OnigCodePoint n;
100 
101   len = enclen(ONIG_ENCODING_EUC_JP, p);
102   n = (OnigCodePoint )*p++;
103   if (len == 1) return n;
104 
105   for (i = 1; i < len; i++) {
106     if (p >= end) break;
107     c = *p++;
108     n <<= 8;  n += c;
109   }
110   return n;
111 }
112 
113 static int
code_to_mbclen(OnigCodePoint code)114 code_to_mbclen(OnigCodePoint code)
115 {
116   if (ONIGENC_IS_CODE_ASCII(code)) return 1;
117   else if ((code & 0xff0000) != 0) return 3;
118   else if ((code &   0xff00) != 0) return 2;
119   else
120     return ONIGERR_INVALID_CODE_POINT_VALUE;
121 }
122 
123 #if 0
124 static int
125 code_to_mbc_first(OnigCodePoint code)
126 {
127   int first;
128 
129   if ((code & 0xff0000) != 0) {
130     first = (code >> 16) & 0xff;
131   }
132   else if ((code & 0xff00) != 0) {
133     first = (code >> 8) & 0xff;
134   }
135   else {
136     return (int )code;
137   }
138   return first;
139 }
140 #endif
141 
142 static int
code_to_mbc(OnigCodePoint code,UChar * buf)143 code_to_mbc(OnigCodePoint code, UChar *buf)
144 {
145   UChar *p = buf;
146 
147   if ((code & 0xff0000) != 0) *p++ = (UChar )(((code >> 16) & 0xff));
148   if ((code &   0xff00) != 0) *p++ = (UChar )(((code >>  8) & 0xff));
149   *p++ = (UChar )(code & 0xff);
150 
151 #if 1
152   if (enclen(ONIG_ENCODING_EUC_JP, buf) != (p - buf))
153     return ONIGERR_INVALID_CODE_POINT_VALUE;
154 #endif
155   return (int )(p - buf);
156 }
157 
158 static int
mbc_case_fold(OnigCaseFoldType flag ARG_UNUSED,const UChar ** pp,const UChar * end ARG_UNUSED,UChar * lower)159 mbc_case_fold(OnigCaseFoldType flag ARG_UNUSED,
160 	      const UChar** pp, const UChar* end ARG_UNUSED, UChar* lower)
161 {
162   int len;
163   const UChar* p = *pp;
164 
165   if (ONIGENC_IS_MBC_ASCII(p)) {
166     *lower = ONIGENC_ASCII_CODE_TO_LOWER_CASE(*p);
167     (*pp)++;
168     return 1;
169   }
170   else {
171     int i;
172 
173     len = enclen(ONIG_ENCODING_EUC_JP, p);
174     for (i = 0; i < len; i++) {
175       *lower++ = *p++;
176     }
177     (*pp) += len;
178     return len; /* return byte length of converted char to lower */
179   }
180 }
181 
182 static UChar*
left_adjust_char_head(const UChar * start,const UChar * s)183 left_adjust_char_head(const UChar* start, const UChar* s)
184 {
185   /* In this encoding
186      mb-trail bytes doesn't mix with single bytes.
187   */
188   const UChar *p;
189   int len;
190 
191   if (s <= start) return (UChar* )s;
192   p = s;
193 
194   while (!eucjp_islead(*p) && p > start) p--;
195   len = enclen(ONIG_ENCODING_EUC_JP, p);
196   if (p + len > s) return (UChar* )p;
197   p += len;
198   return (UChar* )(p + ((s - p) & ~1));
199 }
200 
201 static int
is_allowed_reverse_match(const UChar * s,const UChar * end ARG_UNUSED)202 is_allowed_reverse_match(const UChar* s, const UChar* end ARG_UNUSED)
203 {
204   const UChar c = *s;
205   if (c <= 0x7e || c == 0x8e || c == 0x8f)
206     return TRUE;
207   else
208     return FALSE;
209 }
210 
211 
212 static const OnigCodePoint CR_Hiragana[] = {
213   1,
214   0xa4a1, 0xa4f3
215 }; /* CR_Hiragana */
216 
217 static const OnigCodePoint CR_Katakana[] = {
218   3,
219   0xa5a1, 0xa5f6,
220   0xaaa6, 0xaaaf,
221   0xaab1, 0xaadd
222 }; /* CR_Katakana */
223 
224 static const OnigCodePoint* PropertyList[] = {
225   CR_Hiragana,
226   CR_Katakana
227 };
228 
229 static int
property_name_to_ctype(OnigEncoding enc,UChar * p,UChar * end)230 property_name_to_ctype(OnigEncoding enc, UChar* p, UChar* end)
231 {
232   struct PropertyNameCtype* pc;
233   int len = (int )(end - p);
234   char q[32];
235 
236   if (len < sizeof(q) - 1) {
237     xmemcpy(q, p, (size_t )len);
238     q[len] = '\0';
239     pc = onigenc_euc_jp_lookup_property_name(q, len);
240     if (pc != 0)
241       return pc->ctype;
242   }
243 
244   return ONIGERR_INVALID_CHAR_PROPERTY_NAME;
245 }
246 
247 static int
is_code_ctype(OnigCodePoint code,unsigned int ctype)248 is_code_ctype(OnigCodePoint code, unsigned int ctype)
249 {
250   if (ctype <= ONIGENC_MAX_STD_CTYPE) {
251     if (code < 128)
252       return ONIGENC_IS_ASCII_CODE_CTYPE(code, ctype);
253     else {
254       if (CTYPE_IS_WORD_GRAPH_PRINT(ctype)) {
255         return (code_to_mbclen(code) > 1 ? TRUE : FALSE);
256       }
257     }
258   }
259   else {
260     ctype -= (ONIGENC_MAX_STD_CTYPE + 1);
261     if (ctype >= (unsigned int )(sizeof(PropertyList)/sizeof(PropertyList[0])))
262       return ONIGERR_TYPE_BUG;
263 
264     return onig_is_in_code_range((UChar* )PropertyList[ctype], code);
265   }
266 
267   return FALSE;
268 }
269 
270 static int
get_ctype_code_range(OnigCtype ctype,OnigCodePoint * sb_out,const OnigCodePoint * ranges[])271 get_ctype_code_range(OnigCtype ctype, OnigCodePoint* sb_out,
272 		     const OnigCodePoint* ranges[])
273 {
274   if (ctype <= ONIGENC_MAX_STD_CTYPE) {
275     return ONIG_NO_SUPPORT_CONFIG;
276   }
277   else {
278     *sb_out = 0x80;
279 
280     ctype -= (ONIGENC_MAX_STD_CTYPE + 1);
281     if (ctype >= (OnigCtype )sizeof(PropertyList)/sizeof(PropertyList[0]))
282       return ONIGERR_TYPE_BUG;
283 
284     *ranges = PropertyList[ctype];
285     return 0;
286   }
287 }
288 
289 
290 OnigEncodingType OnigEncodingEUC_JP = {
291   mbc_enc_len,
292   "EUC-JP",   /* name */
293   3,          /* max enc length */
294   1,          /* min enc length */
295   onigenc_is_mbc_newline_0x0a,
296   mbc_to_code,
297   code_to_mbclen,
298   code_to_mbc,
299   mbc_case_fold,
300   onigenc_ascii_apply_all_case_fold,
301   onigenc_ascii_get_case_fold_codes_by_str,
302   property_name_to_ctype,
303   is_code_ctype,
304   get_ctype_code_range,
305   left_adjust_char_head,
306   is_allowed_reverse_match,
307   NULL, /* init */
308   NULL, /* is_initialized */
309   is_valid_mbc_string,
310   ENC_FLAG_ASCII_COMPATIBLE|ENC_FLAG_SKIP_OFFSET_1_OR_0,
311   0, 0
312 };
313