• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
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) 2016-2018 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 
42 /* This module contains mode-dependent macro and structure definitions. The
43 file is #included by pcre2_internal.h if PCRE2_CODE_UNIT_WIDTH is defined.
44 These mode-dependent items are kept in a separate file so that they can also be
45 #included multiple times for different code unit widths by pcre2test in order
46 to have access to the hidden structures at all supported widths.
47 
48 Some of the mode-dependent macros are required at different widths for
49 different parts of the pcre2test code (in particular, the included
50 pcre_printint.c file). We undefine them here so that they can be re-defined for
51 multiple inclusions. Not all of these are used in pcre2test, but it's easier
52 just to undefine them all. */
53 
54 #undef ACROSSCHAR
55 #undef BACKCHAR
56 #undef BYTES2CU
57 #undef CHMAX_255
58 #undef CU2BYTES
59 #undef FORWARDCHAR
60 #undef FORWARDCHARTEST
61 #undef GET
62 #undef GET2
63 #undef GETCHAR
64 #undef GETCHARINC
65 #undef GETCHARINCTEST
66 #undef GETCHARLEN
67 #undef GETCHARLENTEST
68 #undef GETCHARTEST
69 #undef GET_EXTRALEN
70 #undef HAS_EXTRALEN
71 #undef IMM2_SIZE
72 #undef MAX_255
73 #undef MAX_MARK
74 #undef MAX_PATTERN_SIZE
75 #undef MAX_UTF_SINGLE_CU
76 #undef NOT_FIRSTCU
77 #undef PUT
78 #undef PUT2
79 #undef PUT2INC
80 #undef PUTCHAR
81 #undef PUTINC
82 #undef TABLE_GET
83 
84 
85 
86 /* -------------------------- MACROS ----------------------------- */
87 
88 /* PCRE keeps offsets in its compiled code as at least 16-bit quantities
89 (always stored in big-endian order in 8-bit mode) by default. These are used,
90 for example, to link from the start of a subpattern to its alternatives and its
91 end. The use of 16 bits per offset limits the size of an 8-bit compiled regex
92 to around 64K, which is big enough for almost everybody. However, I received a
93 request for an even bigger limit. For this reason, and also to make the code
94 easier to maintain, the storing and loading of offsets from the compiled code
95 unit string is now handled by the macros that are defined here.
96 
97 The macros are controlled by the value of LINK_SIZE. This defaults to 2, but
98 values of 3 or 4 are also supported. */
99 
100 /* ------------------- 8-bit support  ------------------ */
101 
102 #if PCRE2_CODE_UNIT_WIDTH == 8
103 
104 #if LINK_SIZE == 2
105 #define PUT(a,n,d)   \
106   (a[n] = (PCRE2_UCHAR)((d) >> 8)), \
107   (a[(n)+1] = (PCRE2_UCHAR)((d) & 255))
108 #define GET(a,n) \
109   (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
110 #define MAX_PATTERN_SIZE (1 << 16)
111 
112 #elif LINK_SIZE == 3
113 #define PUT(a,n,d)       \
114   (a[n] = (PCRE2_UCHAR)((d) >> 16)),    \
115   (a[(n)+1] = (PCRE2_UCHAR)((d) >> 8)), \
116   (a[(n)+2] = (PCRE2_UCHAR)((d) & 255))
117 #define GET(a,n) \
118   (unsigned int)(((a)[n] << 16) | ((a)[(n)+1] << 8) | (a)[(n)+2])
119 #define MAX_PATTERN_SIZE (1 << 24)
120 
121 #elif LINK_SIZE == 4
122 #define PUT(a,n,d)        \
123   (a[n] = (PCRE2_UCHAR)((d) >> 24)),     \
124   (a[(n)+1] = (PCRE2_UCHAR)((d) >> 16)), \
125   (a[(n)+2] = (PCRE2_UCHAR)((d) >> 8)),  \
126   (a[(n)+3] = (PCRE2_UCHAR)((d) & 255))
127 #define GET(a,n) \
128   (unsigned int)(((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
129 #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */
130 
131 #else
132 #error LINK_SIZE must be 2, 3, or 4
133 #endif
134 
135 
136 /* ------------------- 16-bit support  ------------------ */
137 
138 #elif PCRE2_CODE_UNIT_WIDTH == 16
139 
140 #if LINK_SIZE == 2
141 #undef LINK_SIZE
142 #define LINK_SIZE 1
143 #define PUT(a,n,d)   \
144   (a[n] = (PCRE2_UCHAR)(d))
145 #define GET(a,n) \
146   (a[n])
147 #define MAX_PATTERN_SIZE (1 << 16)
148 
149 #elif LINK_SIZE == 3 || LINK_SIZE == 4
150 #undef LINK_SIZE
151 #define LINK_SIZE 2
152 #define PUT(a,n,d)   \
153   (a[n] = (PCRE2_UCHAR)((d) >> 16)), \
154   (a[(n)+1] = (PCRE2_UCHAR)((d) & 65535))
155 #define GET(a,n) \
156   (unsigned int)(((a)[n] << 16) | (a)[(n)+1])
157 #define MAX_PATTERN_SIZE (1 << 30)  /* Keep it positive */
158 
159 #else
160 #error LINK_SIZE must be 2, 3, or 4
161 #endif
162 
163 
164 /* ------------------- 32-bit support  ------------------ */
165 
166 #elif PCRE2_CODE_UNIT_WIDTH == 32
167 #undef LINK_SIZE
168 #define LINK_SIZE 1
169 #define PUT(a,n,d)   \
170   (a[n] = (d))
171 #define GET(a,n) \
172   (a[n])
173 #define MAX_PATTERN_SIZE (1 << 30)  /* Keep it positive */
174 
175 #else
176 #error Unsupported compiling mode
177 #endif
178 
179 
180 /* --------------- Other mode-specific macros ----------------- */
181 
182 /* PCRE uses some other (at least) 16-bit quantities that do not change when
183 the size of offsets changes. There are used for repeat counts and for other
184 things such as capturing parenthesis numbers in back references.
185 
186 Define the number of code units required to hold a 16-bit count/offset, and
187 macros to load and store such a value. For reasons that I do not understand,
188 the expression in the 8-bit GET2 macro is treated by gcc as a signed
189 expression, even when a is declared as unsigned. It seems that any kind of
190 arithmetic results in a signed value. Hence the cast. */
191 
192 #if PCRE2_CODE_UNIT_WIDTH == 8
193 #define IMM2_SIZE 2
194 #define GET2(a,n) (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
195 #define PUT2(a,n,d) a[n] = (d) >> 8, a[(n)+1] = (d) & 255
196 
197 #else  /* Code units are 16 or 32 bits */
198 #define IMM2_SIZE 1
199 #define GET2(a,n) a[n]
200 #define PUT2(a,n,d) a[n] = d
201 #endif
202 
203 /* Other macros that are different for 8-bit mode. The MAX_255 macro checks
204 whether its argument, which is assumed to be one code unit, is less than 256.
205 The CHMAX_255 macro does not assume one code unit. The maximum length of a MARK
206 name must fit in one code unit; currently it is set to 255 or 65535. The
207 TABLE_GET macro is used to access elements of tables containing exactly 256
208 items. When code points can be greater than 255, a check is needed before
209 accessing these tables. */
210 
211 #if PCRE2_CODE_UNIT_WIDTH == 8
212 #define MAX_255(c) TRUE
213 #define MAX_MARK ((1u << 8) - 1)
214 #ifdef SUPPORT_UNICODE
215 #define SUPPORT_WIDE_CHARS
216 #define CHMAX_255(c) ((c) <= 255u)
217 #else
218 #define CHMAX_255(c) TRUE
219 #endif  /* SUPPORT_UNICODE */
220 #define TABLE_GET(c, table, default) ((table)[c])
221 
222 #else  /* Code units are 16 or 32 bits */
223 #define CHMAX_255(c) ((c) <= 255u)
224 #define MAX_255(c) ((c) <= 255u)
225 #define MAX_MARK ((1u << 16) - 1)
226 #define SUPPORT_WIDE_CHARS
227 #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
228 #endif
229 
230 
231 
232 /* ----------------- Character-handling macros ----------------- */
233 
234 /* There is a proposed future special "UTF-21" mode, in which only the lowest
235 21 bits of a 32-bit character are interpreted as UTF, with the remaining 11
236 high-order bits available to the application for other uses. In preparation for
237 the future implementation of this mode, there are macros that load a data item
238 and, if in this special mode, mask it to 21 bits. These macros all have names
239 starting with UCHAR21. In all other modes, including the normal 32-bit
240 library, the macros all have the same simple definitions. When the new mode is
241 implemented, it is expected that these definitions will be varied appropriately
242 using #ifdef when compiling the library that supports the special mode. */
243 
244 #define UCHAR21(eptr)        (*(eptr))
245 #define UCHAR21TEST(eptr)    (*(eptr))
246 #define UCHAR21INC(eptr)     (*(eptr)++)
247 #define UCHAR21INCTEST(eptr) (*(eptr)++)
248 
249 /* When UTF encoding is being used, a character is no longer just a single
250 byte in 8-bit mode or a single short in 16-bit mode. The macros for character
251 handling generate simple sequences when used in the basic mode, and more
252 complicated ones for UTF characters. GETCHARLENTEST and other macros are not
253 used when UTF is not supported. To make sure they can never even appear when
254 UTF support is omitted, we don't even define them. */
255 
256 #ifndef SUPPORT_UNICODE
257 
258 /* #define MAX_UTF_SINGLE_CU */
259 /* #define HAS_EXTRALEN(c) */
260 /* #define GET_EXTRALEN(c) */
261 /* #define NOT_FIRSTCU(c) */
262 #define GETCHAR(c, eptr) c = *eptr;
263 #define GETCHARTEST(c, eptr) c = *eptr;
264 #define GETCHARINC(c, eptr) c = *eptr++;
265 #define GETCHARINCTEST(c, eptr) c = *eptr++;
266 #define GETCHARLEN(c, eptr, len) c = *eptr;
267 #define PUTCHAR(c, p) (*p = c, 1)
268 /* #define GETCHARLENTEST(c, eptr, len) */
269 /* #define BACKCHAR(eptr) */
270 /* #define FORWARDCHAR(eptr) */
271 /* #define FORWARCCHARTEST(eptr,end) */
272 /* #define ACROSSCHAR(condition, eptr, action) */
273 
274 #else   /* SUPPORT_UNICODE */
275 
276 /* ------------------- 8-bit support  ------------------ */
277 
278 #if PCRE2_CODE_UNIT_WIDTH == 8
279 #define MAYBE_UTF_MULTI          /* UTF chars may use multiple code units */
280 
281 /* The largest UTF code point that can be encoded as a single code unit. */
282 
283 #define MAX_UTF_SINGLE_CU 127
284 
285 /* Tests whether the code point needs extra characters to decode. */
286 
287 #define HAS_EXTRALEN(c) HASUTF8EXTRALEN(c)
288 
289 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
290 Otherwise it has an undefined behaviour. */
291 
292 #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3fu])
293 
294 /* Returns TRUE, if the given value is not the first code unit of a UTF
295 sequence. */
296 
297 #define NOT_FIRSTCU(c) (((c) & 0xc0u) == 0x80u)
298 
299 /* Get the next UTF-8 character, not advancing the pointer. This is called when
300 we know we are in UTF-8 mode. */
301 
302 #define GETCHAR(c, eptr) \
303   c = *eptr; \
304   if (c >= 0xc0u) GETUTF8(c, eptr);
305 
306 /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
307 pointer. */
308 
309 #define GETCHARTEST(c, eptr) \
310   c = *eptr; \
311   if (utf && c >= 0xc0u) GETUTF8(c, eptr);
312 
313 /* Get the next UTF-8 character, advancing the pointer. This is called when we
314 know we are in UTF-8 mode. */
315 
316 #define GETCHARINC(c, eptr) \
317   c = *eptr++; \
318   if (c >= 0xc0u) GETUTF8INC(c, eptr);
319 
320 /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
321 This is called when we don't know if we are in UTF-8 mode. */
322 
323 #define GETCHARINCTEST(c, eptr) \
324   c = *eptr++; \
325   if (utf && c >= 0xc0u) GETUTF8INC(c, eptr);
326 
327 /* Get the next UTF-8 character, not advancing the pointer, incrementing length
328 if there are extra bytes. This is called when we know we are in UTF-8 mode. */
329 
330 #define GETCHARLEN(c, eptr, len) \
331   c = *eptr; \
332   if (c >= 0xc0u) GETUTF8LEN(c, eptr, len);
333 
334 /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
335 pointer, incrementing length if there are extra bytes. This is called when we
336 do not know if we are in UTF-8 mode. */
337 
338 #define GETCHARLENTEST(c, eptr, len) \
339   c = *eptr; \
340   if (utf && c >= 0xc0u) GETUTF8LEN(c, eptr, len);
341 
342 /* If the pointer is not at the start of a character, move it back until
343 it is. This is called only in UTF-8 mode - we don't put a test within the macro
344 because almost all calls are already within a block of UTF-8 only code. */
345 
346 #define BACKCHAR(eptr) while((*eptr & 0xc0u) == 0x80u) eptr--
347 
348 /* Same as above, just in the other direction. */
349 #define FORWARDCHAR(eptr) while((*eptr & 0xc0u) == 0x80u) eptr++
350 #define FORWARDCHARTEST(eptr,end) while(eptr < end && (*eptr & 0xc0u) == 0x80u) eptr++
351 
352 /* Same as above, but it allows a fully customizable form. */
353 #define ACROSSCHAR(condition, eptr, action) \
354   while((condition) && ((*eptr) & 0xc0u) == 0x80u) action
355 
356 /* Deposit a character into memory, returning the number of code units. */
357 
358 #define PUTCHAR(c, p) ((utf && c > MAX_UTF_SINGLE_CU)? \
359   PRIV(ord2utf)(c,p) : (*p = c, 1))
360 
361 
362 /* ------------------- 16-bit support  ------------------ */
363 
364 #elif PCRE2_CODE_UNIT_WIDTH == 16
365 #define MAYBE_UTF_MULTI          /* UTF chars may use multiple code units */
366 
367 /* The largest UTF code point that can be encoded as a single code unit. */
368 
369 #define MAX_UTF_SINGLE_CU 65535
370 
371 /* Tests whether the code point needs extra characters to decode. */
372 
373 #define HAS_EXTRALEN(c) (((c) & 0xfc00u) == 0xd800u)
374 
375 /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
376 Otherwise it has an undefined behaviour. */
377 
378 #define GET_EXTRALEN(c) 1
379 
380 /* Returns TRUE, if the given value is not the first code unit of a UTF
381 sequence. */
382 
383 #define NOT_FIRSTCU(c) (((c) & 0xfc00u) == 0xdc00u)
384 
385 /* Base macro to pick up the low surrogate of a UTF-16 character, not
386 advancing the pointer. */
387 
388 #define GETUTF16(c, eptr) \
389    { c = (((c & 0x3ffu) << 10) | (eptr[1] & 0x3ffu)) + 0x10000u; }
390 
391 /* Get the next UTF-16 character, not advancing the pointer. This is called when
392 we know we are in UTF-16 mode. */
393 
394 #define GETCHAR(c, eptr) \
395   c = *eptr; \
396   if ((c & 0xfc00u) == 0xd800u) GETUTF16(c, eptr);
397 
398 /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
399 pointer. */
400 
401 #define GETCHARTEST(c, eptr) \
402   c = *eptr; \
403   if (utf && (c & 0xfc00u) == 0xd800u) GETUTF16(c, eptr);
404 
405 /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
406 the pointer. */
407 
408 #define GETUTF16INC(c, eptr) \
409    { c = (((c & 0x3ffu) << 10) | (*eptr++ & 0x3ffu)) + 0x10000u; }
410 
411 /* Get the next UTF-16 character, advancing the pointer. This is called when we
412 know we are in UTF-16 mode. */
413 
414 #define GETCHARINC(c, eptr) \
415   c = *eptr++; \
416   if ((c & 0xfc00u) == 0xd800u) GETUTF16INC(c, eptr);
417 
418 /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
419 This is called when we don't know if we are in UTF-16 mode. */
420 
421 #define GETCHARINCTEST(c, eptr) \
422   c = *eptr++; \
423   if (utf && (c & 0xfc00u) == 0xd800u) GETUTF16INC(c, eptr);
424 
425 /* Base macro to pick up the low surrogate of a UTF-16 character, not
426 advancing the pointer, incrementing the length. */
427 
428 #define GETUTF16LEN(c, eptr, len) \
429    { c = (((c & 0x3ffu) << 10) | (eptr[1] & 0x3ffu)) + 0x10000u; len++; }
430 
431 /* Get the next UTF-16 character, not advancing the pointer, incrementing
432 length if there is a low surrogate. This is called when we know we are in
433 UTF-16 mode. */
434 
435 #define GETCHARLEN(c, eptr, len) \
436   c = *eptr; \
437   if ((c & 0xfc00u) == 0xd800u) GETUTF16LEN(c, eptr, len);
438 
439 /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
440 pointer, incrementing length if there is a low surrogate. This is called when
441 we do not know if we are in UTF-16 mode. */
442 
443 #define GETCHARLENTEST(c, eptr, len) \
444   c = *eptr; \
445   if (utf && (c & 0xfc00u) == 0xd800u) GETUTF16LEN(c, eptr, len);
446 
447 /* If the pointer is not at the start of a character, move it back until
448 it is. This is called only in UTF-16 mode - we don't put a test within the
449 macro because almost all calls are already within a block of UTF-16 only
450 code. */
451 
452 #define BACKCHAR(eptr) if ((*eptr & 0xfc00u) == 0xdc00u) eptr--
453 
454 /* Same as above, just in the other direction. */
455 #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00u) == 0xdc00u) eptr++
456 #define FORWARDCHARTEST(eptr,end) if (eptr < end && (*eptr & 0xfc00u) == 0xdc00u) eptr++
457 
458 /* Same as above, but it allows a fully customizable form. */
459 #define ACROSSCHAR(condition, eptr, action) \
460   if ((condition) && ((*eptr) & 0xfc00u) == 0xdc00u) action
461 
462 /* Deposit a character into memory, returning the number of code units. */
463 
464 #define PUTCHAR(c, p) ((utf && c > MAX_UTF_SINGLE_CU)? \
465   PRIV(ord2utf)(c,p) : (*p = c, 1))
466 
467 
468 /* ------------------- 32-bit support  ------------------ */
469 
470 #else
471 
472 /* These are trivial for the 32-bit library, since all UTF-32 characters fit
473 into one PCRE2_UCHAR unit. */
474 
475 #define MAX_UTF_SINGLE_CU (0x10ffffu)
476 #define HAS_EXTRALEN(c) (0)
477 #define GET_EXTRALEN(c) (0)
478 #define NOT_FIRSTCU(c) (0)
479 
480 /* Get the next UTF-32 character, not advancing the pointer. This is called when
481 we know we are in UTF-32 mode. */
482 
483 #define GETCHAR(c, eptr) \
484   c = *(eptr);
485 
486 /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
487 pointer. */
488 
489 #define GETCHARTEST(c, eptr) \
490   c = *(eptr);
491 
492 /* Get the next UTF-32 character, advancing the pointer. This is called when we
493 know we are in UTF-32 mode. */
494 
495 #define GETCHARINC(c, eptr) \
496   c = *((eptr)++);
497 
498 /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
499 This is called when we don't know if we are in UTF-32 mode. */
500 
501 #define GETCHARINCTEST(c, eptr) \
502   c = *((eptr)++);
503 
504 /* Get the next UTF-32 character, not advancing the pointer, not incrementing
505 length (since all UTF-32 is of length 1). This is called when we know we are in
506 UTF-32 mode. */
507 
508 #define GETCHARLEN(c, eptr, len) \
509   GETCHAR(c, eptr)
510 
511 /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
512 pointer, not incrementing the length (since all UTF-32 is of length 1).
513 This is called when we do not know if we are in UTF-32 mode. */
514 
515 #define GETCHARLENTEST(c, eptr, len) \
516   GETCHARTEST(c, eptr)
517 
518 /* If the pointer is not at the start of a character, move it back until
519 it is. This is called only in UTF-32 mode - we don't put a test within the
520 macro because almost all calls are already within a block of UTF-32 only
521 code.
522 
523 These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
524 
525 #define BACKCHAR(eptr) do { } while (0)
526 
527 /* Same as above, just in the other direction. */
528 
529 #define FORWARDCHAR(eptr) do { } while (0)
530 #define FORWARDCHARTEST(eptr,end) do { } while (0)
531 
532 /* Same as above, but it allows a fully customizable form. */
533 
534 #define ACROSSCHAR(condition, eptr, action) do { } while (0)
535 
536 /* Deposit a character into memory, returning the number of code units. */
537 
538 #define PUTCHAR(c, p) (*p = c, 1)
539 
540 #endif  /* UTF-32 character handling */
541 #endif  /* SUPPORT_UNICODE */
542 
543 
544 /* Mode-dependent macros that have the same definition in all modes. */
545 
546 #define CU2BYTES(x)     ((x)*((PCRE2_CODE_UNIT_WIDTH/8)))
547 #define BYTES2CU(x)     ((x)/((PCRE2_CODE_UNIT_WIDTH/8)))
548 #define PUTINC(a,n,d)   PUT(a,n,d), a += LINK_SIZE
549 #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
550 
551 
552 /* ----------------------- HIDDEN STRUCTURES ----------------------------- */
553 
554 /* NOTE: All these structures *must* start with a pcre2_memctl structure. The
555 code that uses them is simpler because it assumes this. */
556 
557 /* The real general context structure. At present it holds only data for custom
558 memory control. */
559 
560 typedef struct pcre2_real_general_context {
561   pcre2_memctl memctl;
562 } pcre2_real_general_context;
563 
564 /* The real compile context structure */
565 
566 typedef struct pcre2_real_compile_context {
567   pcre2_memctl memctl;
568   int (*stack_guard)(uint32_t, void *);
569   void *stack_guard_data;
570   const uint8_t *tables;
571   PCRE2_SIZE max_pattern_length;
572   uint16_t bsr_convention;
573   uint16_t newline_convention;
574   uint32_t parens_nest_limit;
575   uint32_t extra_options;
576 } pcre2_real_compile_context;
577 
578 /* The real match context structure. */
579 
580 typedef struct pcre2_real_match_context {
581   pcre2_memctl memctl;
582 #ifdef SUPPORT_JIT
583   pcre2_jit_callback jit_callback;
584   void *jit_callback_data;
585 #endif
586   int    (*callout)(pcre2_callout_block *, void *);
587   void    *callout_data;
588   PCRE2_SIZE offset_limit;
589   uint32_t heap_limit;
590   uint32_t match_limit;
591   uint32_t depth_limit;
592 } pcre2_real_match_context;
593 
594 /* The real convert context structure. */
595 
596 typedef struct pcre2_real_convert_context {
597   pcre2_memctl memctl;
598   uint32_t glob_separator;
599   uint32_t glob_escape;
600 } pcre2_real_convert_context;
601 
602 /* The real compiled code structure. The type for the blocksize field is
603 defined specially because it is required in pcre2_serialize_decode() when
604 copying the size from possibly unaligned memory into a variable of the same
605 type. Use a macro rather than a typedef to avoid compiler warnings when this
606 file is included multiple times by pcre2test. LOOKBEHIND_MAX specifies the
607 largest lookbehind that is supported. (OP_REVERSE in a pattern has a 16-bit
608 argument in 8-bit and 16-bit modes, so we need no more than a 16-bit field
609 here.) */
610 
611 #undef  CODE_BLOCKSIZE_TYPE
612 #define CODE_BLOCKSIZE_TYPE size_t
613 
614 #undef  LOOKBEHIND_MAX
615 #define LOOKBEHIND_MAX UINT16_MAX
616 
617 typedef struct pcre2_real_code {
618   pcre2_memctl memctl;            /* Memory control fields */
619   const uint8_t *tables;          /* The character tables */
620   void    *executable_jit;        /* Pointer to JIT code */
621   uint8_t  start_bitmap[32];      /* Bitmap for starting code unit < 256 */
622   CODE_BLOCKSIZE_TYPE blocksize;  /* Total (bytes) that was malloc-ed */
623   uint32_t magic_number;          /* Paranoid and endianness check */
624   uint32_t compile_options;       /* Options passed to pcre2_compile() */
625   uint32_t overall_options;       /* Options after processing the pattern */
626   uint32_t extra_options;         /* Taken from compile_context */
627   uint32_t flags;                 /* Various state flags */
628   uint32_t limit_heap;            /* Limit set in the pattern */
629   uint32_t limit_match;           /* Limit set in the pattern */
630   uint32_t limit_depth;           /* Limit set in the pattern */
631   uint32_t first_codeunit;        /* Starting code unit */
632   uint32_t last_codeunit;         /* This codeunit must be seen */
633   uint16_t bsr_convention;        /* What \R matches */
634   uint16_t newline_convention;    /* What is a newline? */
635   uint16_t max_lookbehind;        /* Longest lookbehind (characters) */
636   uint16_t minlength;             /* Minimum length of match */
637   uint16_t top_bracket;           /* Highest numbered group */
638   uint16_t top_backref;           /* Highest numbered back reference */
639   uint16_t name_entry_size;       /* Size (code units) of table entries */
640   uint16_t name_count;            /* Number of name entries in the table */
641 } pcre2_real_code;
642 
643 /* The real match data structure. Define ovector as large as it can ever
644 actually be so that array bound checkers don't grumble. Memory for this
645 structure is obtained by calling pcre2_match_data_create(), which sets the size
646 as the offset of ovector plus a pair of elements for each capturable string, so
647 the size varies from call to call. As the maximum number of capturing
648 subpatterns is 65535 we must allow for 65536 strings to include the overall
649 match. (See also the heapframe structure below.) */
650 
651 typedef struct pcre2_real_match_data {
652   pcre2_memctl     memctl;
653   const pcre2_real_code *code;    /* The pattern used for the match */
654   PCRE2_SPTR       subject;       /* The subject that was matched */
655   PCRE2_SPTR       mark;          /* Pointer to last mark */
656   PCRE2_SIZE       leftchar;      /* Offset to leftmost code unit */
657   PCRE2_SIZE       rightchar;     /* Offset to rightmost code unit */
658   PCRE2_SIZE       startchar;     /* Offset to starting code unit */
659   uint16_t         matchedby;     /* Type of match (normal, JIT, DFA) */
660   uint16_t         oveccount;     /* Number of pairs */
661   int              rc;            /* The return code from the match */
662   PCRE2_SIZE       ovector[131072]; /* Must be last in the structure */
663 } pcre2_real_match_data;
664 
665 
666 /* ----------------------- PRIVATE STRUCTURES ----------------------------- */
667 
668 /* These structures are not needed for pcre2test. */
669 
670 #ifndef PCRE2_PCRE2TEST
671 
672 /* Structures for checking for mutual recursion when scanning compiled or
673 parsed code. */
674 
675 typedef struct recurse_check {
676   struct recurse_check *prev;
677   PCRE2_SPTR group;
678 } recurse_check;
679 
680 typedef struct parsed_recurse_check {
681   struct parsed_recurse_check *prev;
682   uint32_t *groupptr;
683 } parsed_recurse_check;
684 
685 /* Structure for building a cache when filling in recursion offsets. */
686 
687 typedef struct recurse_cache {
688   PCRE2_SPTR group;
689   int groupnumber;
690 } recurse_cache;
691 
692 /* Structure for maintaining a chain of pointers to the currently incomplete
693 branches, for testing for left recursion while compiling. */
694 
695 typedef struct branch_chain {
696   struct branch_chain *outer;
697   PCRE2_UCHAR *current_branch;
698 } branch_chain;
699 
700 /* Structure for building a list of named groups during the first pass of
701 compiling. */
702 
703 typedef struct named_group {
704   PCRE2_SPTR   name;          /* Points to the name in the pattern */
705   uint32_t     number;        /* Group number */
706   uint16_t     length;        /* Length of the name */
707   uint16_t     isdup;         /* TRUE if a duplicate */
708 } named_group;
709 
710 /* Structure for passing "static" information around between the functions
711 doing the compiling, so that they are thread-safe. */
712 
713 typedef struct compile_block {
714   pcre2_real_compile_context *cx;  /* Points to the compile context */
715   const uint8_t *lcc;              /* Points to lower casing table */
716   const uint8_t *fcc;              /* Points to case-flipping table */
717   const uint8_t *cbits;            /* Points to character type table */
718   const uint8_t *ctypes;           /* Points to table of type maps */
719   PCRE2_SPTR start_workspace;      /* The start of working space */
720   PCRE2_SPTR start_code;           /* The start of the compiled code */
721   PCRE2_SPTR start_pattern;        /* The start of the pattern */
722   PCRE2_SPTR end_pattern;          /* The end of the pattern */
723   PCRE2_UCHAR *name_table;         /* The name/number table */
724   PCRE2_SIZE workspace_size;       /* Size of workspace */
725   PCRE2_SIZE small_ref_offset[10]; /* Offsets for \1 to \9 */
726   PCRE2_SIZE erroroffset;          /* Offset of error in pattern */
727   uint16_t names_found;            /* Number of entries so far */
728   uint16_t name_entry_size;        /* Size of each entry */
729   uint16_t parens_depth;           /* Depth of nested parentheses */
730   uint16_t assert_depth;           /* Depth of nested assertions */
731   open_capitem *open_caps;         /* Chain of open capture items */
732   named_group *named_groups;       /* Points to vector in pre-compile */
733   uint32_t named_group_list_size;  /* Number of entries in the list */
734   uint32_t external_options;       /* External (initial) options */
735   uint32_t external_flags;         /* External flag bits to be set */
736   uint32_t bracount;               /* Count of capturing parentheses */
737   uint32_t lastcapture;            /* Last capture encountered */
738   uint32_t *parsed_pattern;        /* Parsed pattern buffer */
739   uint32_t *parsed_pattern_end;    /* Parsed pattern should not get here */
740   uint32_t *groupinfo;             /* Group info vector */
741   uint32_t top_backref;            /* Maximum back reference */
742   uint32_t backref_map;            /* Bitmap of low back refs */
743   uint32_t nltype;                 /* Newline type */
744   uint32_t nllen;                  /* Newline string length */
745   uint32_t class_range_start;      /* Overall class range start */
746   uint32_t class_range_end;        /* Overall class range end */
747   PCRE2_UCHAR nl[4];               /* Newline string when fixed length */
748   int  max_lookbehind;             /* Maximum lookbehind (characters) */
749   int  req_varyopt;                /* "After variable item" flag for reqbyte */
750   BOOL had_accept;                 /* (*ACCEPT) encountered */
751   BOOL had_pruneorskip;            /* (*PRUNE) or (*SKIP) encountered */
752   BOOL had_recurse;                /* Had a recursion or subroutine call */
753   BOOL dupnames;                   /* Duplicate names exist */
754 } compile_block;
755 
756 /* Structure for keeping the properties of the in-memory stack used
757 by the JIT matcher. */
758 
759 typedef struct pcre2_real_jit_stack {
760   pcre2_memctl memctl;
761   void* stack;
762 } pcre2_real_jit_stack;
763 
764 /* Structure for items in a linked list that represents an explicit recursive
765 call within the pattern when running pcre_dfa_match(). */
766 
767 typedef struct dfa_recursion_info {
768   struct dfa_recursion_info *prevrec;
769   PCRE2_SPTR subject_position;
770   uint32_t group_num;
771 } dfa_recursion_info;
772 
773 /* Structure for "stack" frames that are used for remembering backtracking
774 positions during matching. As these are used in a vector, with the ovector item
775 being extended, the size of the structure must be a multiple of PCRE2_SIZE. The
776 only way to check this at compile time is to force an error by generating an
777 array with a negative size. By putting this in a typedef (which is never used),
778 we don't generate any code when all is well. */
779 
780 typedef struct heapframe {
781 
782   /* The first set of fields are variables that have to be preserved over calls
783   to RRMATCH(), but which do not need to be copied to new frames. */
784 
785   PCRE2_SPTR ecode;          /* The current position in the pattern */
786   PCRE2_SPTR temp_sptr[2];   /* Used for short-term PCRE_SPTR values */
787   PCRE2_SIZE length;         /* Used for character, string, or code lengths */
788   PCRE2_SIZE back_frame;     /* Amount to subtract on RRETURN */
789   PCRE2_SIZE temp_size;      /* Used for short-term PCRE2_SIZE values */
790   uint32_t rdepth;           /* "Recursion" depth */
791   uint32_t group_frame_type; /* Type information for group frames */
792   uint32_t temp_32[4];       /* Used for short-term 32-bit or BOOL values */
793   uint8_t return_id;         /* Where to go on in internal "return" */
794   uint8_t op;                /* Processing opcode */
795 
796   /* At this point, the structure is 16-bit aligned. On most architectures
797   the alignment requirement for a pointer will ensure that the eptr field below
798   is 32-bit or 64-bit aligned. However, on m68k it is fine to have a pointer
799   that is 16-bit aligned. We must therefore ensure that what comes between here
800   and eptr is an odd multiple of 16 bits so as to get back into 32-bit
801   alignment. This happens naturally when PCRE2_UCHAR is 8 bits wide, but needs
802   fudges in the other cases. In the 32-bit case the padding comes first so that
803   the occu field itself is 32-bit aligned. Without the padding, this structure
804   is no longer a multiple of PCRE2_SIZE on m68k, and the check below fails. */
805 
806 #if PCRE2_CODE_UNIT_WIDTH == 8
807   PCRE2_UCHAR occu[6];       /* Used for other case code units */
808 #elif PCRE2_CODE_UNIT_WIDTH == 16
809   PCRE2_UCHAR occu[2];       /* Used for other case code units */
810   uint8_t unused[2];         /* Ensure 32-bit alignment (see above) */
811 #else
812   uint8_t unused[2];         /* Ensure 32-bit alignment (see above) */
813   PCRE2_UCHAR occu[1];       /* Used for other case code units */
814 #endif
815 
816   /* The rest have to be copied from the previous frame whenever a new frame
817   becomes current. The final field is specified as a large vector so that
818   runtime array bound checks don't catch references to it. However, for any
819   specific call to pcre2_match() the memory allocated for each frame structure
820   allows for exactly the right size ovector for the number of capturing
821   parentheses. (See also the comment for pcre2_real_match_data above.) */
822 
823   PCRE2_SPTR eptr;           /* MUST BE FIRST */
824   PCRE2_SPTR start_match;    /* Can be adjusted by \K */
825   PCRE2_SPTR mark;           /* Most recent mark on the success path */
826   uint32_t current_recurse;  /* Current (deepest) recursion number */
827   uint32_t capture_last;     /* Most recent capture */
828   PCRE2_SIZE last_group_offset;  /* Saved offset to most recent group frame */
829   PCRE2_SIZE offset_top;     /* Offset after highest capture */
830   PCRE2_SIZE ovector[131072]; /* Must be last in the structure */
831 } heapframe;
832 
833 /* This typedef is a check that the size of the heapframe structure is a
834 multiple of PCRE2_SIZE. See various comments above. */
835 
836 typedef char check_heapframe_size[
837   ((sizeof(heapframe) % sizeof(PCRE2_SIZE)) == 0)? (+1):(-1)];
838 
839 /* Structure for passing "static" information around between the functions
840 doing traditional NFA matching (pcre2_match() and friends). */
841 
842 typedef struct match_block {
843   pcre2_memctl memctl;            /* For general use */
844   PCRE2_SIZE frame_vector_size;   /* Size of a backtracking frame */
845   heapframe *match_frames;        /* Points to vector of frames */
846   heapframe *match_frames_top;    /* Points after the end of the vector */
847   heapframe *stack_frames;        /* The original vector on the stack */
848   PCRE2_SIZE heap_limit;          /* As it says */
849   uint32_t match_limit;           /* As it says */
850   uint32_t match_limit_depth;     /* As it says */
851   uint32_t match_call_count;      /* Number of times a new frame is created */
852   BOOL hitend;                    /* Hit the end of the subject at some point */
853   BOOL hasthen;                   /* Pattern contains (*THEN) */
854   const uint8_t *lcc;             /* Points to lower casing table */
855   const uint8_t *fcc;             /* Points to case-flipping table */
856   const uint8_t *ctypes;          /* Points to table of type maps */
857   PCRE2_SIZE start_offset;        /* The start offset value */
858   PCRE2_SIZE end_offset_top;      /* Highwater mark at end of match */
859   uint16_t partial;               /* PARTIAL options */
860   uint16_t bsr_convention;        /* \R interpretation */
861   uint16_t name_count;            /* Number of names in name table */
862   uint16_t name_entry_size;       /* Size of entry in names table */
863   PCRE2_SPTR name_table;          /* Table of group names */
864   PCRE2_SPTR start_code;          /* For use when recursing */
865   PCRE2_SPTR start_subject;       /* Start of the subject string */
866   PCRE2_SPTR end_subject;         /* End of the subject string */
867   PCRE2_SPTR end_match_ptr;       /* Subject position at end match */
868   PCRE2_SPTR start_used_ptr;      /* Earliest consulted character */
869   PCRE2_SPTR last_used_ptr;       /* Latest consulted character */
870   PCRE2_SPTR mark;                /* Mark pointer to pass back on success */
871   PCRE2_SPTR nomatch_mark;        /* Mark pointer to pass back on failure */
872   PCRE2_SPTR verb_ecode_ptr;      /* For passing back info */
873   PCRE2_SPTR verb_skip_ptr;       /* For passing back a (*SKIP) name */
874   uint32_t verb_current_recurse;  /* Current recurse when (*VERB) happens */
875   uint32_t moptions;              /* Match options */
876   uint32_t poptions;              /* Pattern options */
877   uint32_t skip_arg_count;        /* For counting SKIP_ARGs */
878   uint32_t ignore_skip_arg;       /* For re-run when SKIP arg name not found */
879   uint32_t nltype;                /* Newline type */
880   uint32_t nllen;                 /* Newline string length */
881   PCRE2_UCHAR nl[4];              /* Newline string when fixed */
882   pcre2_callout_block *cb;        /* Points to a callout block */
883   void  *callout_data;            /* To pass back to callouts */
884   int (*callout)(pcre2_callout_block *,void *);  /* Callout function or NULL */
885 } match_block;
886 
887 /* A similar structure is used for the same purpose by the DFA matching
888 functions. */
889 
890 typedef struct dfa_match_block {
891   pcre2_memctl memctl;            /* For general use */
892   PCRE2_SPTR start_code;          /* Start of the compiled pattern */
893   PCRE2_SPTR start_subject ;      /* Start of the subject string */
894   PCRE2_SPTR end_subject;         /* End of subject string */
895   PCRE2_SPTR start_used_ptr;      /* Earliest consulted character */
896   PCRE2_SPTR last_used_ptr;       /* Latest consulted character */
897   const uint8_t *tables;          /* Character tables */
898   PCRE2_SIZE start_offset;        /* The start offset value */
899   PCRE2_SIZE heap_limit;          /* As it says */
900   PCRE2_SIZE heap_used;           /* As it says */
901   uint32_t match_limit;           /* As it says */
902   uint32_t match_limit_depth;     /* As it says */
903   uint32_t match_call_count;      /* Number of calls of internal function */
904   uint32_t moptions;              /* Match options */
905   uint32_t poptions;              /* Pattern options */
906   uint32_t nltype;                /* Newline type */
907   uint32_t nllen;                 /* Newline string length */
908   PCRE2_UCHAR nl[4];              /* Newline string when fixed */
909   uint16_t bsr_convention;        /* \R interpretation */
910   pcre2_callout_block *cb;        /* Points to a callout block */
911   void *callout_data;             /* To pass back to callouts */
912   int (*callout)(pcre2_callout_block *,void *);  /* Callout function or NULL */
913   dfa_recursion_info *recursive;  /* Linked list of recursion data */
914 } dfa_match_block;
915 
916 #endif  /* PCRE2_PCRE2TEST */
917 
918 /* End of pcre2_intmodedep.h */
919