1 /*************************************************
2 * Perl-Compatible Regular Expressions *
3 *************************************************/
4
5 /* PCRE is a library of functions to support regular expressions whose syntax
6 and semantics are as close as possible to those of the Perl 5 language.
7
8 Written by Philip Hazel
9 Original API code Copyright (c) 1997-2012 University of Cambridge
10 New API code Copyright (c) 2016-2023 University of Cambridge
11
12 -----------------------------------------------------------------------------
13 Redistribution and use in source and binary forms, with or without
14 modification, are permitted provided that the following conditions are met:
15
16 * Redistributions of source code must retain the above copyright notice,
17 this list of conditions and the following disclaimer.
18
19 * Redistributions in binary form must reproduce the above copyright
20 notice, this list of conditions and the following disclaimer in the
21 documentation and/or other materials provided with the distribution.
22
23 * Neither the name of the University of Cambridge nor the names of its
24 contributors may be used to endorse or promote products derived from
25 this software without specific prior written permission.
26
27 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
28 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
29 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
30 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
31 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
32 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
33 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
34 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
35 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
37 POSSIBILITY OF SUCH DAMAGE.
38 -----------------------------------------------------------------------------
39 */
40
41 /* This module contains an internal function that is used to match an extended
42 class. It is used by pcre2_auto_possessify() and by both pcre2_match() and
43 pcre2_def_match(). */
44
45
46 #ifdef HAVE_CONFIG_H
47 #include "config.h"
48 #endif
49
50
51 #include "pcre2_internal.h"
52
53 /*************************************************
54 * Match character against an XCLASS *
55 *************************************************/
56
57 /* This function is called to match a character against an extended class that
58 might contain codepoints above 255 and/or Unicode properties.
59
60 Arguments:
61 c the character
62 data points to the flag code unit of the XCLASS data
63 utf TRUE if in UTF mode
64
65 Returns: TRUE if character matches, else FALSE
66 */
67
68 BOOL
PRIV(xclass)69 PRIV(xclass)(uint32_t c, PCRE2_SPTR data, BOOL utf)
70 {
71 PCRE2_UCHAR t;
72 BOOL negated = (*data & XCL_NOT) != 0;
73
74 #if PCRE2_CODE_UNIT_WIDTH == 8
75 /* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */
76 utf = TRUE;
77 #endif
78
79 /* Code points < 256 are matched against a bitmap, if one is present. If not,
80 we still carry on, because there may be ranges that start below 256 in the
81 additional data. */
82
83 if (c < 256)
84 {
85 if ((*data & XCL_HASPROP) == 0)
86 {
87 if ((*data & XCL_MAP) == 0) return negated;
88 return (((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0;
89 }
90 if ((*data & XCL_MAP) != 0 &&
91 (((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0)
92 return !negated; /* char found */
93 }
94
95 /* First skip the bit map if present. Then match against the list of Unicode
96 properties or large chars or ranges that end with a large char. We won't ever
97 encounter XCL_PROP or XCL_NOTPROP when UTF support is not compiled. */
98
99 if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(PCRE2_UCHAR);
100
101 while ((t = *data++) != XCL_END)
102 {
103 uint32_t x, y;
104 if (t == XCL_SINGLE)
105 {
106 #ifdef SUPPORT_UNICODE
107 if (utf)
108 {
109 GETCHARINC(x, data); /* macro generates multiple statements */
110 }
111 else
112 #endif
113 x = *data++;
114 if (c == x) return !negated;
115 }
116 else if (t == XCL_RANGE)
117 {
118 #ifdef SUPPORT_UNICODE
119 if (utf)
120 {
121 GETCHARINC(x, data); /* macro generates multiple statements */
122 GETCHARINC(y, data); /* macro generates multiple statements */
123 }
124 else
125 #endif
126 {
127 x = *data++;
128 y = *data++;
129 }
130 if (c >= x && c <= y) return !negated;
131 }
132
133 #ifdef SUPPORT_UNICODE
134 else /* XCL_PROP & XCL_NOTPROP */
135 {
136 int chartype;
137 const ucd_record *prop = GET_UCD(c);
138 BOOL isprop = t == XCL_PROP;
139 BOOL ok;
140
141 switch(*data)
142 {
143 case PT_ANY:
144 if (isprop) return !negated;
145 break;
146
147 case PT_LAMP:
148 chartype = prop->chartype;
149 if ((chartype == ucp_Lu || chartype == ucp_Ll ||
150 chartype == ucp_Lt) == isprop) return !negated;
151 break;
152
153 case PT_GC:
154 if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop)
155 return !negated;
156 break;
157
158 case PT_PC:
159 if ((data[1] == prop->chartype) == isprop) return !negated;
160 break;
161
162 case PT_SC:
163 if ((data[1] == prop->script) == isprop) return !negated;
164 break;
165
166 case PT_SCX:
167 ok = (data[1] == prop->script ||
168 MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), data[1]) != 0);
169 if (ok == isprop) return !negated;
170 break;
171
172 case PT_ALNUM:
173 chartype = prop->chartype;
174 if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
175 PRIV(ucp_gentype)[chartype] == ucp_N) == isprop)
176 return !negated;
177 break;
178
179 /* Perl space used to exclude VT, but from Perl 5.18 it is included,
180 which means that Perl space and POSIX space are now identical. PCRE
181 was changed at release 8.34. */
182
183 case PT_SPACE: /* Perl space */
184 case PT_PXSPACE: /* POSIX space */
185 switch(c)
186 {
187 HSPACE_CASES:
188 VSPACE_CASES:
189 if (isprop) return !negated;
190 break;
191
192 default:
193 if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == isprop)
194 return !negated;
195 break;
196 }
197 break;
198
199 case PT_WORD:
200 chartype = prop->chartype;
201 if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
202 PRIV(ucp_gentype)[chartype] == ucp_N ||
203 chartype == ucp_Mn || chartype == ucp_Pc) == isprop)
204 return !negated;
205 break;
206
207 case PT_UCNC:
208 if (c < 0xa0)
209 {
210 if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
211 c == CHAR_GRAVE_ACCENT) == isprop)
212 return !negated;
213 }
214 else
215 {
216 if ((c < 0xd800 || c > 0xdfff) == isprop)
217 return !negated;
218 }
219 break;
220
221 case PT_BIDICL:
222 if ((UCD_BIDICLASS_PROP(prop) == data[1]) == isprop)
223 return !negated;
224 break;
225
226 case PT_BOOL:
227 ok = MAPBIT(PRIV(ucd_boolprop_sets) +
228 UCD_BPROPS_PROP(prop), data[1]) != 0;
229 if (ok == isprop) return !negated;
230 break;
231
232 /* The following three properties can occur only in an XCLASS, as there
233 is no \p or \P coding for them. */
234
235 /* Graphic character. Implement this as not Z (space or separator) and
236 not C (other), except for Cf (format) with a few exceptions. This seems
237 to be what Perl does. The exceptional characters are:
238
239 U+061C Arabic Letter Mark
240 U+180E Mongolian Vowel Separator
241 U+2066 - U+2069 Various "isolate"s
242 */
243
244 case PT_PXGRAPH:
245 chartype = prop->chartype;
246 if ((PRIV(ucp_gentype)[chartype] != ucp_Z &&
247 (PRIV(ucp_gentype)[chartype] != ucp_C ||
248 (chartype == ucp_Cf &&
249 c != 0x061c && c != 0x180e && (c < 0x2066 || c > 0x2069))
250 )) == isprop)
251 return !negated;
252 break;
253
254 /* Printable character: same as graphic, with the addition of Zs, i.e.
255 not Zl and not Zp, and U+180E. */
256
257 case PT_PXPRINT:
258 chartype = prop->chartype;
259 if ((chartype != ucp_Zl &&
260 chartype != ucp_Zp &&
261 (PRIV(ucp_gentype)[chartype] != ucp_C ||
262 (chartype == ucp_Cf &&
263 c != 0x061c && (c < 0x2066 || c > 0x2069))
264 )) == isprop)
265 return !negated;
266 break;
267
268 /* Punctuation: all Unicode punctuation, plus ASCII characters that
269 Unicode treats as symbols rather than punctuation, for Perl
270 compatibility (these are $+<=>^`|~). */
271
272 case PT_PXPUNCT:
273 chartype = prop->chartype;
274 if ((PRIV(ucp_gentype)[chartype] == ucp_P ||
275 (c < 128 && PRIV(ucp_gentype)[chartype] == ucp_S)) == isprop)
276 return !negated;
277 break;
278
279 /* Perl has two sets of hex digits */
280
281 case PT_PXXDIGIT:
282 if (((c >= CHAR_0 && c <= CHAR_9) ||
283 (c >= CHAR_A && c <= CHAR_F) ||
284 (c >= CHAR_a && c <= CHAR_f) ||
285 (c >= 0xff10 && c <= 0xff19) || /* Fullwidth digits */
286 (c >= 0xff21 && c <= 0xff26) || /* Fullwidth letters */
287 (c >= 0xff41 && c <= 0xff46)) == isprop)
288 return !negated;
289 break;
290
291 /* This should never occur, but compilers may mutter if there is no
292 default. */
293
294 default:
295 return FALSE;
296 }
297
298 data += 2;
299 }
300 #else
301 (void)utf; /* Avoid compiler warning */
302 #endif /* SUPPORT_UNICODE */
303 }
304
305 return negated; /* char did not match */
306 }
307
308 /* End of pcre2_xclass.c */
309
