xref: /PHP-7.3/ext/mbstring/oniguruma/src/utf16_le.c (revision 1979c5d1)
1 /**********************************************************************
2   utf16_le.c -  Oniguruma (regular expression library)
3 **********************************************************************/
4 /*-
5  * Copyright (c) 2002-2019  K.Kosako
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 #include "regint.h"  /* for USE_CALLOUT */
30 
31 static int
init(void)32 init(void)
33 {
34 #ifdef USE_CALLOUT
35 
36     int id;
37     OnigEncoding enc;
38     char* name;
39     unsigned int args[4];
40     OnigValue    opts[4];
41 
42     enc = ONIG_ENCODING_UTF16_LE;
43 
44     name = "F\000A\000I\000L\000\000\000";            BC0_P(name, fail);
45     name = "M\000I\000S\000M\000A\000T\000C\000H\000\000\000"; BC0_P(name, mismatch);
46 
47     name = "M\000A\000X\000\000\000";
48     args[0] = ONIG_TYPE_TAG | ONIG_TYPE_LONG;
49     args[1] = ONIG_TYPE_CHAR;
50     opts[0].c = 'X';
51     BC_B_O(name, max, 2, args, 1, opts);
52 
53     name = "E\000R\000R\000O\000R\000\000\000";
54     args[0] = ONIG_TYPE_LONG; opts[0].l = ONIG_ABORT;
55     BC_P_O(name, error, 1, args, 1, opts);
56 
57     name = "C\000O\000U\000N\000T\000\000\000";
58     args[0] = ONIG_TYPE_CHAR; opts[0].c = '>';
59     BC_B_O(name, count, 1, args, 1, opts);
60 
61     name = "T\000O\000T\000A\000L\000_\000C\000O\000U\000N\000T\000\000\000";
62     args[0] = ONIG_TYPE_CHAR; opts[0].c = '>';
63     BC_B_O(name, total_count, 1, args, 1, opts);
64 
65     name = "C\000M\000P\000\000\000";
66     args[0] = ONIG_TYPE_TAG | ONIG_TYPE_LONG;
67     args[1] = ONIG_TYPE_STRING;
68     args[2] = ONIG_TYPE_TAG | ONIG_TYPE_LONG;
69     BC_P(name, cmp, 3, args);
70 
71 #endif /* USE_CALLOUT */
72 
73   return ONIG_NORMAL;
74 }
75 
76 static const int EncLen_UTF16[] = {
77   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
78   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
79   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
80   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
81   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
82   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
83   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
84   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
85   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
86   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
87   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
88   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
89   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
90   2, 2, 2, 2, 2, 2, 2, 2, 4, 4, 4, 4, 2, 2, 2, 2,
91   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
92   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2
93 };
94 
95 static int
utf16le_code_to_mbclen(OnigCodePoint code)96 utf16le_code_to_mbclen(OnigCodePoint code)
97 {
98   if (code > 0xffff) {
99     if (code > 0x10ffff)
100       return ONIGERR_INVALID_CODE_POINT_VALUE;
101     else
102       return 4;
103   }
104   else {
105     return 2;
106   }
107 }
108 
109 static int
utf16le_mbc_enc_len(const UChar * p)110 utf16le_mbc_enc_len(const UChar* p)
111 {
112   return EncLen_UTF16[*(p+1)];
113 }
114 
115 static int
is_valid_mbc_string(const UChar * p,const UChar * end)116 is_valid_mbc_string(const UChar* p, const UChar* end)
117 {
118   const UChar* end1 = end - 1;
119 
120   while (p < end1) {
121     int len = utf16le_mbc_enc_len(p);
122     if (len == 4) {
123       if (p + 3 < end && ! UTF16_IS_SURROGATE_SECOND(*(p + 3)))
124         return FALSE;
125     }
126     else
127       if (UTF16_IS_SURROGATE_SECOND(*(p + 1)))
128         return FALSE;
129 
130     p += len;
131   }
132 
133   if (p != end)
134     return FALSE;
135   else
136     return TRUE;
137 }
138 
139 static int
utf16le_is_mbc_newline(const UChar * p,const UChar * end)140 utf16le_is_mbc_newline(const UChar* p, const UChar* end)
141 {
142   if (p + 1 < end) {
143     if (*p == 0x0a && *(p+1) == 0x00)
144       return 1;
145 #ifdef USE_UNICODE_ALL_LINE_TERMINATORS
146     if ((
147 #ifndef USE_CRNL_AS_LINE_TERMINATOR
148          *p == 0x0d ||
149 #endif
150          *p == 0x85) && *(p+1) == 0x00)
151       return 1;
152 
153     if (*(p+1) == 0x20 && (*p == 0x29 || *p == 0x28))
154       return 1;
155 #endif
156   }
157   return 0;
158 }
159 
160 static OnigCodePoint
utf16le_mbc_to_code(const UChar * p,const UChar * end ARG_UNUSED)161 utf16le_mbc_to_code(const UChar* p, const UChar* end ARG_UNUSED)
162 {
163   OnigCodePoint code;
164   UChar c0 = *p;
165   UChar c1 = *(p+1);
166 
167   if (UTF16_IS_SURROGATE_FIRST(c1)) {
168     code = ((((c1 - 0xd8) << 2) + ((c0  & 0xc0) >> 6) + 1) << 16)
169          + ((((c0 & 0x3f) << 2) + (p[3] - 0xdc)) << 8)
170          + p[2];
171   }
172   else {
173     code = c1 * 256 + p[0];
174   }
175   return code;
176 }
177 
178 static int
utf16le_code_to_mbc(OnigCodePoint code,UChar * buf)179 utf16le_code_to_mbc(OnigCodePoint code, UChar *buf)
180 {
181   UChar* p = buf;
182 
183   if (code > 0xffff) {
184     unsigned int plane, high;
185 
186     plane = (code >> 16) - 1;
187     high = (code & 0xff00) >> 8;
188 
189     *p++ = ((plane & 0x03) << 6) + (high >> 2);
190     *p++ = (plane >> 2) + 0xd8;
191     *p++ = (UChar )(code & 0xff);
192     *p   = (high & 0x03) + 0xdc;
193     return 4;
194   }
195   else {
196     *p++ = (UChar )(code & 0xff);
197     *p++ = (UChar )((code & 0xff00) >> 8);
198     return 2;
199   }
200 }
201 
202 static int
utf16le_mbc_case_fold(OnigCaseFoldType flag,const UChar ** pp,const UChar * end,UChar * fold)203 utf16le_mbc_case_fold(OnigCaseFoldType flag,
204                       const UChar** pp, const UChar* end, UChar* fold)
205 {
206   const UChar* p = *pp;
207 
208   if (ONIGENC_IS_ASCII_CODE(*p) && *(p+1) == 0) {
209 #ifdef USE_UNICODE_CASE_FOLD_TURKISH_AZERI
210     if ((flag & ONIGENC_CASE_FOLD_TURKISH_AZERI) != 0) {
211       if (*p == 0x49) {
212         *fold++ = 0x31;
213         *fold   = 0x01;
214         (*pp) += 2;
215         return 2;
216       }
217     }
218 #endif
219 
220     *fold++ = ONIGENC_ASCII_CODE_TO_LOWER_CASE(*p);
221     *fold   = 0;
222     *pp += 2;
223     return 2;
224   }
225   else
226     return onigenc_unicode_mbc_case_fold(ONIG_ENCODING_UTF16_LE, flag, pp, end,
227                                          fold);
228 }
229 
230 static UChar*
utf16le_left_adjust_char_head(const UChar * start,const UChar * s)231 utf16le_left_adjust_char_head(const UChar* start, const UChar* s)
232 {
233   if (s <= start) return (UChar* )s;
234 
235   if ((s - start) % 2 == 1) {
236     s--;
237   }
238 
239   if (UTF16_IS_SURROGATE_SECOND(*(s+1)) && s > start + 1 &&
240       UTF16_IS_SURROGATE_FIRST(*(s-1)))
241     s -= 2;
242 
243   return (UChar* )s;
244 }
245 
246 static int
utf16le_get_case_fold_codes_by_str(OnigCaseFoldType flag,const OnigUChar * p,const OnigUChar * end,OnigCaseFoldCodeItem items[])247 utf16le_get_case_fold_codes_by_str(OnigCaseFoldType flag,
248     const OnigUChar* p, const OnigUChar* end, OnigCaseFoldCodeItem items[])
249 {
250   return onigenc_unicode_get_case_fold_codes_by_str(ONIG_ENCODING_UTF16_LE,
251                                                     flag, p, end, items);
252 }
253 
254 OnigEncodingType OnigEncodingUTF16_LE = {
255   utf16le_mbc_enc_len,
256   "UTF-16LE",   /* name */
257   4,            /* max enc length */
258   2,            /* min enc length */
259   utf16le_is_mbc_newline,
260   utf16le_mbc_to_code,
261   utf16le_code_to_mbclen,
262   utf16le_code_to_mbc,
263   utf16le_mbc_case_fold,
264   onigenc_unicode_apply_all_case_fold,
265   utf16le_get_case_fold_codes_by_str,
266   onigenc_unicode_property_name_to_ctype,
267   onigenc_unicode_is_code_ctype,
268   onigenc_utf16_32_get_ctype_code_range,
269   utf16le_left_adjust_char_head,
270   onigenc_always_false_is_allowed_reverse_match,
271   init,
272   0, /* is_initialized */
273   is_valid_mbc_string,
274   ENC_FLAG_UNICODE|ENC_FLAG_SKIP_OFFSET_1,
275   0, 0
276 };
277