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) 2016-2022 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 |
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 | const ucd_record *prop = GET_UCD(c); |
137 | BOOL isprop = t == XCL_PROP; |
138 | BOOL ok; |
139 | |
140 | switch(*data) |
141 | { |
142 | case PT_ANY: |
143 | if (isprop) return !negated; |
144 | break; |
145 | |
146 | case PT_LAMP: |
147 | if ((prop->chartype == ucp_Lu || prop->chartype == ucp_Ll || |
148 | prop->chartype == ucp_Lt) == isprop) return !negated; |
149 | break; |
150 | |
151 | case PT_GC: |
152 | if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop) |
153 | return !negated; |
154 | break; |
155 | |
156 | case PT_PC: |
157 | if ((data[1] == prop->chartype) == isprop) return !negated; |
158 | break; |
159 | |
160 | case PT_SC: |
161 | if ((data[1] == prop->script) == isprop) return !negated; |
162 | break; |
163 | |
164 | case PT_SCX: |
165 | ok = (data[1] == prop->script || |
166 | MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), data[1]) != 0); |
167 | if (ok == isprop) return !negated; |
168 | break; |
169 | |
170 | case PT_ALNUM: |
171 | if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
172 | PRIV(ucp_gentype)[prop->chartype] == ucp_N) == isprop) |
173 | return !negated; |
174 | break; |
175 | |
176 | /* Perl space used to exclude VT, but from Perl 5.18 it is included, |
177 | which means that Perl space and POSIX space are now identical. PCRE |
178 | was changed at release 8.34. */ |
179 | |
180 | case PT_SPACE: /* Perl space */ |
181 | case PT_PXSPACE: /* POSIX space */ |
182 | switch(c) |
183 | { |
184 | HSPACE_CASES: |
185 | VSPACE_CASES: |
186 | if (isprop) return !negated; |
187 | break; |
188 | |
189 | default: |
190 | if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == isprop) |
191 | return !negated; |
192 | break; |
193 | } |
194 | break; |
195 | |
196 | case PT_WORD: |
197 | if ((PRIV(ucp_gentype)[prop->chartype] == ucp_L || |
198 | PRIV(ucp_gentype)[prop->chartype] == ucp_N || c == CHAR_UNDERSCORE) |
199 | == isprop) |
200 | return !negated; |
201 | break; |
202 | |
203 | case PT_UCNC: |
204 | if (c < 0xa0) |
205 | { |
206 | if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT || |
207 | c == CHAR_GRAVE_ACCENT) == isprop) |
208 | return !negated; |
209 | } |
210 | else |
211 | { |
212 | if ((c < 0xd800 || c > 0xdfff) == isprop) |
213 | return !negated; |
214 | } |
215 | break; |
216 | |
217 | case PT_BIDICL: |
218 | if ((UCD_BIDICLASS_PROP(prop) == data[1]) == isprop) |
219 | return !negated; |
220 | break; |
221 | |
222 | case PT_BOOL: |
223 | ok = MAPBIT(PRIV(ucd_boolprop_sets) + |
224 | UCD_BPROPS_PROP(prop), data[1]) != 0; |
225 | if (ok == isprop) return !negated; |
226 | break; |
227 | |
228 | /* The following three properties can occur only in an XCLASS, as there |
229 | is no \p or \P coding for them. */ |
230 | |
231 | /* Graphic character. Implement this as not Z (space or separator) and |
232 | not C (other), except for Cf (format) with a few exceptions. This seems |
233 | to be what Perl does. The exceptional characters are: |
234 | |
235 | U+061C Arabic Letter Mark |
236 | U+180E Mongolian Vowel Separator |
237 | U+2066 - U+2069 Various "isolate"s |
238 | */ |
239 | |
240 | case PT_PXGRAPH: |
241 | if ((PRIV(ucp_gentype)[prop->chartype] != ucp_Z && |
242 | (PRIV(ucp_gentype)[prop->chartype] != ucp_C || |
243 | (prop->chartype == ucp_Cf && |
244 | c != 0x061c && c != 0x180e && (c < 0x2066 || c > 0x2069)) |
245 | )) == isprop) |
246 | return !negated; |
247 | break; |
248 | |
249 | /* Printable character: same as graphic, with the addition of Zs, i.e. |
250 | not Zl and not Zp, and U+180E. */ |
251 | |
252 | case PT_PXPRINT: |
253 | if ((prop->chartype != ucp_Zl && |
254 | prop->chartype != ucp_Zp && |
255 | (PRIV(ucp_gentype)[prop->chartype] != ucp_C || |
256 | (prop->chartype == ucp_Cf && |
257 | c != 0x061c && (c < 0x2066 || c > 0x2069)) |
258 | )) == isprop) |
259 | return !negated; |
260 | break; |
261 | |
262 | /* Punctuation: all Unicode punctuation, plus ASCII characters that |
263 | Unicode treats as symbols rather than punctuation, for Perl |
264 | compatibility (these are $+<=>^`|~). */ |
265 | |
266 | case PT_PXPUNCT: |
267 | if ((PRIV(ucp_gentype)[prop->chartype] == ucp_P || |
268 | (c < 128 && PRIV(ucp_gentype)[prop->chartype] == ucp_S)) == isprop) |
269 | return !negated; |
270 | break; |
271 | |
272 | /* This should never occur, but compilers may mutter if there is no |
273 | default. */ |
274 | |
275 | default: |
276 | return FALSE; |
277 | } |
278 | |
279 | data += 2; |
280 | } |
281 | #else |
282 | (void)utf; /* Avoid compiler warning */ |
283 | #endif /* SUPPORT_UNICODE */ |
284 | } |
285 | |
286 | return negated; /* char did not match */ |
287 | } |
288 | |
289 | /* End of pcre2_xclass.c */ |
290 | |