/src/php-src/ext/pcre/pcre2lib/pcre2_xclass.c

Source
/*************************************************
*      Perl-Compatible Regular Expressions       *
*************************************************/

/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.

                       Written by Philip Hazel
     Original API code Copyright (c) 1997-2012 University of Cambridge
          New API code Copyright (c) 2016-2023 University of Cambridge

-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

    * Redistributions of source code must retain the above copyright notice,
      this list of conditions and the following disclaimer.

    * Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.

    * Neither the name of the University of Cambridge nor the names of its
      contributors may be used to endorse or promote products derived from
      this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/

/* This module contains an internal function that is used to match an extended
class. It is used by pcre2_auto_possessify() and by both pcre2_match() and
pcre2_def_match(). */


#ifdef HAVE_CONFIG_H
#include "config.h"
#endif


#include "pcre2_internal.h"

/*************************************************
*       Match character against an XCLASS        *
*************************************************/

/* This function is called to match a character against an extended class that
might contain codepoints above 255 and/or Unicode properties.

Arguments:
  c           the character
  data        points to the flag code unit of the XCLASS data
  utf         TRUE if in UTF mode

Returns:      TRUE if character matches, else FALSE
*/

BOOL
PRIV(xclass)(uint32_t c, PCRE2_SPTR data, BOOL utf)
{
PCRE2_UCHAR t;
BOOL negated = (*data & XCL_NOT) != 0;

#if PCRE2_CODE_UNIT_WIDTH == 8
/* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */
utf = TRUE;
#endif

/* Code points < 256 are matched against a bitmap, if one is present. If not,
we still carry on, because there may be ranges that start below 256 in the
additional data. */

if (c < 256)
  {
  if ((*data & XCL_HASPROP) == 0)
    {
    if ((*data & XCL_MAP) == 0) return negated;
    return (((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0;
    }
  if ((*data & XCL_MAP) != 0 &&
    (((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0)
    return !negated; /* char found */
  }

/* First skip the bit map if present. Then match against the list of Unicode
properties or large chars or ranges that end with a large char. We won't ever
encounter XCL_PROP or XCL_NOTPROP when UTF support is not compiled. */

if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(PCRE2_UCHAR);

while ((t = *data++) != XCL_END)
  {
  uint32_t x, y;
  if (t == XCL_SINGLE)
    {
#ifdef SUPPORT_UNICODE
    if (utf)
      {
      GETCHARINC(x, data); /* macro generates multiple statements */
      }
    else
#endif
    x = *data++;
    if (c == x) return !negated;
    }
  else if (t == XCL_RANGE)
    {
#ifdef SUPPORT_UNICODE
    if (utf)
      {
      GETCHARINC(x, data); /* macro generates multiple statements */
      GETCHARINC(y, data); /* macro generates multiple statements */
      }
    else
#endif
      {
      x = *data++;
      y = *data++;
      }
    if (c >= x && c <= y) return !negated;
    }

#ifdef SUPPORT_UNICODE
  else  /* XCL_PROP & XCL_NOTPROP */
    {
    int chartype;
    const ucd_record *prop = GET_UCD(c);
    BOOL isprop = t == XCL_PROP;
    BOOL ok;

    switch(*data)
      {
      case PT_ANY:
      if (isprop) return !negated;
      break;

      case PT_LAMP:
      chartype = prop->chartype;
      if ((chartype == ucp_Lu || chartype == ucp_Ll ||
           chartype == ucp_Lt) == isprop) return !negated;
      break;

      case PT_GC:
      if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop)
        return !negated;
      break;

      case PT_PC:
      if ((data[1] == prop->chartype) == isprop) return !negated;
      break;

      case PT_SC:
      if ((data[1] == prop->script) == isprop) return !negated;
      break;

      case PT_SCX:
      ok = (data[1] == prop->script ||
            MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), data[1]) != 0);
      if (ok == isprop) return !negated;
      break;

      case PT_ALNUM:
      chartype = prop->chartype;
      if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
           PRIV(ucp_gentype)[chartype] == ucp_N) == isprop)
        return !negated;
      break;

      /* Perl space used to exclude VT, but from Perl 5.18 it is included,
      which means that Perl space and POSIX space are now identical. PCRE
      was changed at release 8.34. */

      case PT_SPACE:    /* Perl space */
      case PT_PXSPACE:  /* POSIX space */
      switch(c)
        {
        HSPACE_CASES:
        VSPACE_CASES:
        if (isprop) return !negated;
        break;

        default:
        if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == isprop)
          return !negated;
        break;
        }
      break;

      case PT_WORD:
      chartype = prop->chartype;
      if ((PRIV(ucp_gentype)[chartype] == ucp_L ||
           PRIV(ucp_gentype)[chartype] == ucp_N ||
           chartype == ucp_Mn || chartype == ucp_Pc) == isprop)
        return !negated;
      break;

      case PT_UCNC:
      if (c < 0xa0)
        {
        if ((c == CHAR_DOLLAR_SIGN || c == CHAR_COMMERCIAL_AT ||
             c == CHAR_GRAVE_ACCENT) == isprop)
          return !negated;
        }
      else
        {
        if ((c < 0xd800 || c > 0xdfff) == isprop)
          return !negated;
        }
      break;

      case PT_BIDICL:
      if ((UCD_BIDICLASS_PROP(prop) == data[1]) == isprop)
        return !negated;
      break;

      case PT_BOOL:
      ok = MAPBIT(PRIV(ucd_boolprop_sets) +
        UCD_BPROPS_PROP(prop), data[1]) != 0;
      if (ok == isprop) return !negated;
      break;

      /* The following three properties can occur only in an XCLASS, as there
      is no \p or \P coding for them. */

      /* Graphic character. Implement this as not Z (space or separator) and
      not C (other), except for Cf (format) with a few exceptions. This seems
      to be what Perl does. The exceptional characters are:

      U+061C           Arabic Letter Mark
      U+180E           Mongolian Vowel Separator
      U+2066 - U+2069  Various "isolate"s
      */

      case PT_PXGRAPH:
      chartype = prop->chartype;
      if ((PRIV(ucp_gentype)[chartype] != ucp_Z &&
            (PRIV(ucp_gentype)[chartype] != ucp_C ||
              (chartype == ucp_Cf &&
                c != 0x061c && c != 0x180e && (c < 0x2066 || c > 0x2069))
         )) == isprop)
        return !negated;
      break;

      /* Printable character: same as graphic, with the addition of Zs, i.e.
      not Zl and not Zp, and U+180E. */

      case PT_PXPRINT:
      chartype = prop->chartype;
      if ((chartype != ucp_Zl &&
           chartype != ucp_Zp &&
            (PRIV(ucp_gentype)[chartype] != ucp_C ||
              (chartype == ucp_Cf &&
                c != 0x061c && (c < 0x2066 || c > 0x2069))
         )) == isprop)
        return !negated;
      break;

      /* Punctuation: all Unicode punctuation, plus ASCII characters that
      Unicode treats as symbols rather than punctuation, for Perl
      compatibility (these are $+<=>^`|~). */

      case PT_PXPUNCT:
      chartype = prop->chartype;
      if ((PRIV(ucp_gentype)[chartype] == ucp_P ||
            (c < 128 && PRIV(ucp_gentype)[chartype] == ucp_S)) == isprop)
        return !negated;
      break;

      /* Perl has two sets of hex digits */

      case PT_PXXDIGIT:
      if (((c >= CHAR_0 && c <= CHAR_9) ||
           (c >= CHAR_A && c <= CHAR_F) ||
           (c >= CHAR_a && c <= CHAR_f) ||
           (c >= 0xff10 && c <= 0xff19) ||  /* Fullwidth digits */
           (c >= 0xff21 && c <= 0xff26) ||  /* Fullwidth letters */
           (c >= 0xff41 && c <= 0xff46)) == isprop)
        return !negated;
      break;

      /* This should never occur, but compilers may mutter if there is no
      default. */

      default:
      return FALSE;
      }

    data += 2;
    }
#else
  (void)utf;  /* Avoid compiler warning */
#endif  /* SUPPORT_UNICODE */
  }

return negated;   /* char did not match */
}

/* End of pcre2_xclass.c */

Coverage Report

Created: 2026-06-02 06:40

Line	Count	Source
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
69		PRIV(xclass)(uint32_t c, PCRE2_SPTR data, BOOL utf)
70	928k	{
71	928k	PCRE2_UCHAR t;
72	928k	BOOL negated = (*data & XCL_NOT) != 0;
73
74	928k	#if PCRE2_CODE_UNIT_WIDTH == 8
75		/* In 8 bit mode, this must always be TRUE. Help the compiler to know that. */
76	928k	utf = TRUE;
77	928k	#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	928k	if (c < 256)
84	922k	{
85	922k	if ((*data & XCL_HASPROP) == 0)
86	130k	{
87	130k	if ((*data & XCL_MAP) == 0) return negated;
88	130k	return (((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0;
89	130k	}
90	791k	if ((*data & XCL_MAP) != 0 &&
91	791k	(((uint8_t *)(data + 1))[c/8] & (1u << (c&7))) != 0)
92	253k	return !negated; /* char found */
93	791k	}
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	544k	if ((*data++ & XCL_MAP) != 0) data += 32 / sizeof(PCRE2_UCHAR);
100
101	927k	while ((t = *data++) != XCL_END)
102	560k	{
103	560k	uint32_t x, y;
104	560k	if (t == XCL_SINGLE)
105	17.8k	{
106	17.8k	#ifdef SUPPORT_UNICODE
107	17.8k	if (utf)
108	17.8k	{
109	17.8k	GETCHARINC(x, data); /* macro generates multiple statements */
110	17.8k	}
111	0	else
112	0	#endif
113	0	x = *data++;
114	17.8k	if (c == x) return !negated;
115	17.8k	}
116	542k	else if (t == XCL_RANGE)
117	77	{
118	77	#ifdef SUPPORT_UNICODE
119	77	if (utf)
120	77	{
121	77	GETCHARINC(x, data); /* macro generates multiple statements */
122	77	GETCHARINC(y, data); /* macro generates multiple statements */
123	77	}
124	0	else
125	0	#endif
126	0	{
127	0	x = *data++;
128	0	y = *data++;
129	0	}
130	77	if (c >= x && c <= y) return !negated;
131	77	}
132
133	542k	#ifdef SUPPORT_UNICODE
134	542k	else /* XCL_PROP & XCL_NOTPROP */
135	542k	{
136	542k	int chartype;
137	542k	const ucd_record *prop = GET_UCD(c);
138	542k	BOOL isprop = t == XCL_PROP;
139	542k	BOOL ok;
140
141	542k	switch(*data)
142	542k	{
143	0	case PT_ANY:
144	0	if (isprop) return !negated;
145	0	break;
146
147	0	case PT_LAMP:
148	0	chartype = prop->chartype;
149	0	if ((chartype == ucp_Lu \|\| chartype == ucp_Ll \|\|
150	0	chartype == ucp_Lt) == isprop) return !negated;
151	0	break;
152
153	476k	case PT_GC:
154	476k	if ((data[1] == PRIV(ucp_gentype)[prop->chartype]) == isprop)
155	134k	return !negated;
156	342k	break;
157
158	342k	case PT_PC:
159	81	if ((data[1] == prop->chartype) == isprop) return !negated;
160	32	break;
161
162	32	case PT_SC:
163	0	if ((data[1] == prop->script) == isprop) return !negated;
164	0	break;
165
166	0	case PT_SCX:
167	0	ok = (data[1] == prop->script \|\|
168	0	MAPBIT(PRIV(ucd_script_sets) + UCD_SCRIPTX_PROP(prop), data[1]) != 0);
169	0	if (ok == isprop) return !negated;
170	0	break;
171
172	0	case PT_ALNUM:
173	0	chartype = prop->chartype;
174	0	if ((PRIV(ucp_gentype)[chartype] == ucp_L \|\|
175	0	PRIV(ucp_gentype)[chartype] == ucp_N) == isprop)
176	0	return !negated;
177	0	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	1.17k	case PT_SPACE: /* Perl space */
184	1.17k	case PT_PXSPACE: /* POSIX space */
185	1.17k	switch(c)
186	1.17k	{
187	4.28k	HSPACE_CASES:
188	4.28k	VSPACE_CASES:
189	2.46k	if (isprop) return !negated;
190	46	break;
191
192	823	default:
193	823	if ((PRIV(ucp_gentype)[prop->chartype] == ucp_Z) == isprop)
194	33	return !negated;
195	790	break;
196	1.17k	}
197	836	break;
198
199	64.3k	case PT_WORD:
200	64.3k	chartype = prop->chartype;
201	64.3k	if ((PRIV(ucp_gentype)[chartype] == ucp_L \|\|
202	46.6k	PRIV(ucp_gentype)[chartype] == ucp_N \|\|
203	24.3k	chartype == ucp_Mn \|\| chartype == ucp_Pc) == isprop)
204	42.7k	return !negated;
205	21.6k	break;
206
207	21.6k	case PT_UCNC:
208	0	if (c < 0xa0)
209	0	{
210	0	if ((c == CHAR_DOLLAR_SIGN \|\| c == CHAR_COMMERCIAL_AT \|\|
211	0	c == CHAR_GRAVE_ACCENT) == isprop)
212	0	return !negated;
213	0	}
214	0	else
215	0	{
216	0	if ((c < 0xd800 \|\| c > 0xdfff) == isprop)
217	0	return !negated;
218	0	}
219	0	break;
220
221	0	case PT_BIDICL:
222	0	if ((UCD_BIDICLASS_PROP(prop) == data[1]) == isprop)
223	0	return !negated;
224	0	break;
225
226	0	case PT_BOOL:
227	0	ok = MAPBIT(PRIV(ucd_boolprop_sets) +
228	0	UCD_BPROPS_PROP(prop), data[1]) != 0;
229	0	if (ok == isprop) return !negated;
230	0	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	0	case PT_PXGRAPH:
245	0	chartype = prop->chartype;
246	0	if ((PRIV(ucp_gentype)[chartype] != ucp_Z &&
247	0	(PRIV(ucp_gentype)[chartype] != ucp_C \|\|
248	0	(chartype == ucp_Cf &&
249	0	c != 0x061c && c != 0x180e && (c < 0x2066 \|\| c > 0x2069))
250	0	)) == isprop)
251	0	return !negated;
252	0	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	0	case PT_PXPRINT:
258	0	chartype = prop->chartype;
259	0	if ((chartype != ucp_Zl &&
260	0	chartype != ucp_Zp &&
261	0	(PRIV(ucp_gentype)[chartype] != ucp_C \|\|
262	0	(chartype == ucp_Cf &&
263	0	c != 0x061c && (c < 0x2066 \|\| c > 0x2069))
264	0	)) == isprop)
265	0	return !negated;
266	0	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	0	case PT_PXPUNCT:
273	0	chartype = prop->chartype;
274	0	if ((PRIV(ucp_gentype)[chartype] == ucp_P \|\|
275	0	(c < 128 && PRIV(ucp_gentype)[chartype] == ucp_S)) == isprop)
276	0	return !negated;
277	0	break;
278
279		/* Perl has two sets of hex digits */
280
281	0	case PT_PXXDIGIT:
282	0	if (((c >= CHAR_0 && c <= CHAR_9) \|\|
283	0	(c >= CHAR_A && c <= CHAR_F) \|\|
284	0	(c >= CHAR_a && c <= CHAR_f) \|\|
285	0	(c >= 0xff10 && c <= 0xff19) \|\| /* Fullwidth digits */
286	0	(c >= 0xff21 && c <= 0xff26) \|\| /* Fullwidth letters */
287	0	(c >= 0xff41 && c <= 0xff46)) == isprop)
288	0	return !negated;
289	0	break;
290
291		/* This should never occur, but compilers may mutter if there is no
292		default. */
293
294	0	default:
295	0	return FALSE;
296	542k	}
297
298	364k	data += 2;
299	364k	}
300		#else
301		(void)utf; /* Avoid compiler warning */
302		#endif /* SUPPORT_UNICODE */
303	560k	}
304
305	367k	return negated; /* char did not match */
306	544k	}
307
308		/* End of pcre2_xclass.c */