ViewVC logotype

Contents of /code/trunk/doc/pcreapi.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 686 - (show annotations)
Tue Sep 6 16:06:36 2011 UTC (10 years, 1 month ago) by ph10
File size: 109376 byte(s)
More documentation updates.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 .B #include <pcre.h>
8 .PP
9 .SM
10 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11 .ti +5n
12 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13 .ti +5n
14 .B const unsigned char *\fItableptr\fP);
15 .PP
16 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17 .ti +5n
18 .B int *\fIerrorcodeptr\fP,
19 .ti +5n
20 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21 .ti +5n
22 .B const unsigned char *\fItableptr\fP);
23 .PP
24 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25 .ti +5n
26 .B const char **\fIerrptr\fP);
27 .PP
28 .B void pcre_free_study(pcre_extra *\fIextra\fP);
29 .PP
30 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
31 .ti +5n
32 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
33 .ti +5n
34 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
35 .
36 .
38 .rs
39 .sp
40 .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
41 .PP
42 .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
43 .PP
44 .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
45 .ti +5n
46 .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
47 .PP
48 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
49 .ti +5n
50 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
51 .ti +5n
52 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
53 .ti +5n
54 .B int *\fIworkspace\fP, int \fIwscount\fP);
55 .PP
56 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
57 .ti +5n
58 .B const char *\fIsubject\fP, int *\fIovector\fP,
59 .ti +5n
60 .B int \fIstringcount\fP, const char *\fIstringname\fP,
61 .ti +5n
62 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
63 .PP
64 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
65 .ti +5n
66 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
67 .ti +5n
68 .B int \fIbuffersize\fP);
69 .PP
70 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
71 .ti +5n
72 .B const char *\fIsubject\fP, int *\fIovector\fP,
73 .ti +5n
74 .B int \fIstringcount\fP, const char *\fIstringname\fP,
75 .ti +5n
76 .B const char **\fIstringptr\fP);
77 .PP
78 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
79 .ti +5n
80 .B const char *\fIname\fP);
81 .PP
82 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
83 .ti +5n
84 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
85 .PP
86 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
87 .ti +5n
88 .B int \fIstringcount\fP, int \fIstringnumber\fP,
89 .ti +5n
90 .B const char **\fIstringptr\fP);
91 .PP
92 .B int pcre_get_substring_list(const char *\fIsubject\fP,
93 .ti +5n
94 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
95 .PP
96 .B void pcre_free_substring(const char *\fIstringptr\fP);
97 .PP
98 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
99 .PP
100 .B const unsigned char *pcre_maketables(void);
101 .PP
102 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
103 .ti +5n
104 .B int \fIwhat\fP, void *\fIwhere\fP);
105 .PP
106 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
107 .B *\fIfirstcharptr\fP);
108 .PP
109 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
110 .PP
111 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
112 .PP
113 .B char *pcre_version(void);
114 .
115 .
117 .rs
118 .sp
119 .B void *(*pcre_malloc)(size_t);
120 .PP
121 .B void (*pcre_free)(void *);
122 .PP
123 .B void *(*pcre_stack_malloc)(size_t);
124 .PP
125 .B void (*pcre_stack_free)(void *);
126 .PP
127 .B int (*pcre_callout)(pcre_callout_block *);
128 .
129 .
131 .rs
132 .sp
133 PCRE has its own native API, which is described in this document. There are
134 also some wrapper functions that correspond to the POSIX regular expression
135 API, but they do not give access to all the functionality. They are described
136 in the
137 .\" HREF
138 \fBpcreposix\fP
139 .\"
140 documentation. Both of these APIs define a set of C function calls. A C++
141 wrapper is also distributed with PCRE. It is documented in the
142 .\" HREF
143 \fBpcrecpp\fP
144 .\"
145 page.
146 .P
147 The native API C function prototypes are defined in the header file
148 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
149 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
150 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
151 and PCRE_MINOR to contain the major and minor release numbers for the library.
152 Applications can use these to include support for different releases of PCRE.
153 .P
154 In a Windows environment, if you want to statically link an application program
155 against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
156 including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
157 \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
158 \fB__declspec(dllimport)\fP, with unwanted results.
159 .P
160 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
161 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
162 in a Perl-compatible manner. A sample program that demonstrates the simplest
163 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
164 source distribution. A listing of this program is given in the
165 .\" HREF
166 \fBpcredemo\fP
167 .\"
168 documentation, and the
169 .\" HREF
170 \fBpcresample\fP
171 .\"
172 documentation describes how to compile and run it.
173 .P
174 Just-in-time compiler support is an optional feature of PCRE that can be built
175 in appropriate hardware environments. It greatly speeds up the matching
176 performance of many patterns. Simple programs can easily request that it be
177 used if available, by setting an option that is ignored when it is not
178 relevant. More complicated programs might need to make use of the functions
179 \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
180 \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
181 These functions are discussed in the
182 .\" HREF
183 \fBpcrejit\fP
184 .\"
185 documentation.
186 .P
187 A second matching function, \fBpcre_dfa_exec()\fP, which is not
188 Perl-compatible, is also provided. This uses a different algorithm for the
189 matching. The alternative algorithm finds all possible matches (at a given
190 point in the subject), and scans the subject just once (unless there are
191 lookbehind assertions). However, this algorithm does not return captured
192 substrings. A description of the two matching algorithms and their advantages
193 and disadvantages is given in the
194 .\" HREF
195 \fBpcrematching\fP
196 .\"
197 documentation.
198 .P
199 In addition to the main compiling and matching functions, there are convenience
200 functions for extracting captured substrings from a subject string that is
201 matched by \fBpcre_exec()\fP. They are:
202 .sp
203 \fBpcre_copy_substring()\fP
204 \fBpcre_copy_named_substring()\fP
205 \fBpcre_get_substring()\fP
206 \fBpcre_get_named_substring()\fP
207 \fBpcre_get_substring_list()\fP
208 \fBpcre_get_stringnumber()\fP
209 \fBpcre_get_stringtable_entries()\fP
210 .sp
211 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
212 provided, to free the memory used for extracted strings.
213 .P
214 The function \fBpcre_maketables()\fP is used to build a set of character tables
215 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
216 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
217 specialist use. Most commonly, no special tables are passed, in which case
218 internal tables that are generated when PCRE is built are used.
219 .P
220 The function \fBpcre_fullinfo()\fP is used to find out information about a
221 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
222 some of the available information, but is retained for backwards compatibility.
223 The function \fBpcre_version()\fP returns a pointer to a string containing the
224 version of PCRE and its date of release.
225 .P
226 The function \fBpcre_refcount()\fP maintains a reference count in a data block
227 containing a compiled pattern. This is provided for the benefit of
228 object-oriented applications.
229 .P
230 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
231 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
232 respectively. PCRE calls the memory management functions via these variables,
233 so a calling program can replace them if it wishes to intercept the calls. This
234 should be done before calling any PCRE functions.
235 .P
236 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
237 indirections to memory management functions. These special functions are used
238 only when PCRE is compiled to use the heap for remembering data, instead of
239 recursive function calls, when running the \fBpcre_exec()\fP function. See the
240 .\" HREF
241 \fBpcrebuild\fP
242 .\"
243 documentation for details of how to do this. It is a non-standard way of
244 building PCRE, for use in environments that have limited stacks. Because of the
245 greater use of memory management, it runs more slowly. Separate functions are
246 provided so that special-purpose external code can be used for this case. When
247 used, these functions are always called in a stack-like manner (last obtained,
248 first freed), and always for memory blocks of the same size. There is a
249 discussion about PCRE's stack usage in the
250 .\" HREF
251 \fBpcrestack\fP
252 .\"
253 documentation.
254 .P
255 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
256 by the caller to a "callout" function, which PCRE will then call at specified
257 points during a matching operation. Details are given in the
258 .\" HREF
259 \fBpcrecallout\fP
260 .\"
261 documentation.
262 .
263 .
264 .\" HTML <a name="newlines"></a>
266 .rs
267 .sp
268 PCRE supports five different conventions for indicating line breaks in
269 strings: a single CR (carriage return) character, a single LF (linefeed)
270 character, the two-character sequence CRLF, any of the three preceding, or any
271 Unicode newline sequence. The Unicode newline sequences are the three just
272 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
273 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
274 (paragraph separator, U+2029).
275 .P
276 Each of the first three conventions is used by at least one operating system as
277 its standard newline sequence. When PCRE is built, a default can be specified.
278 The default default is LF, which is the Unix standard. When PCRE is run, the
279 default can be overridden, either when a pattern is compiled, or when it is
280 matched.
281 .P
282 At compile time, the newline convention can be specified by the \fIoptions\fP
283 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
284 start of the pattern itself; this overrides any other settings. See the
285 .\" HREF
286 \fBpcrepattern\fP
287 .\"
288 page for details of the special character sequences.
289 .P
290 In the PCRE documentation the word "newline" is used to mean "the character or
291 pair of characters that indicate a line break". The choice of newline
292 convention affects the handling of the dot, circumflex, and dollar
293 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
294 recognized line ending sequence, the match position advancement for a
295 non-anchored pattern. There is more detail about this in the
296 .\" HTML <a href="#execoptions">
297 .\" </a>
298 section on \fBpcre_exec()\fP options
299 .\"
300 below.
301 .P
302 The choice of newline convention does not affect the interpretation of
303 the \en or \er escape sequences, nor does it affect what \eR matches, which is
304 controlled in a similar way, but by separate options.
305 .
306 .
308 .rs
309 .sp
310 The PCRE functions can be used in multi-threading applications, with the
311 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
312 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
313 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
314 .P
315 The compiled form of a regular expression is not altered during matching, so
316 the same compiled pattern can safely be used by several threads at once.
317 .P
318 If the just-in-time optimization feature is being used, it needs separate
319 memory stack areas for each thread. See the
320 .\" HREF
321 \fBpcrejit\fP
322 .\"
323 documentation for more details.
324 .
325 .
327 .rs
328 .sp
329 The compiled form of a regular expression can be saved and re-used at a later
330 time, possibly by a different program, and even on a host other than the one on
331 which it was compiled. Details are given in the
332 .\" HREF
333 \fBpcreprecompile\fP
334 .\"
335 documentation. However, compiling a regular expression with one version of PCRE
336 for use with a different version is not guaranteed to work and may cause
337 crashes.
338 .
339 .
341 .rs
342 .sp
343 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
344 .PP
345 The function \fBpcre_config()\fP makes it possible for a PCRE client to
346 discover which optional features have been compiled into the PCRE library. The
347 .\" HREF
348 \fBpcrebuild\fP
349 .\"
350 documentation has more details about these optional features.
351 .P
352 The first argument for \fBpcre_config()\fP is an integer, specifying which
353 information is required; the second argument is a pointer to a variable into
354 which the information is placed. The following information is available:
355 .sp
357 .sp
358 The output is an integer that is set to one if UTF-8 support is available;
359 otherwise it is set to zero.
360 .sp
362 .sp
363 The output is an integer that is set to one if support for Unicode character
364 properties is available; otherwise it is set to zero.
365 .sp
367 .sp
368 The output is an integer that is set to one if support for just-in-time
369 compiling is available; otherwise it is set to zero.
370 .sp
372 .sp
373 The output is an integer whose value specifies the default character sequence
374 that is recognized as meaning "newline". The four values that are supported
375 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
376 Though they are derived from ASCII, the same values are returned in EBCDIC
377 environments. The default should normally correspond to the standard sequence
378 for your operating system.
379 .sp
381 .sp
382 The output is an integer whose value indicates what character sequences the \eR
383 escape sequence matches by default. A value of 0 means that \eR matches any
384 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
385 or CRLF. The default can be overridden when a pattern is compiled or matched.
386 .sp
388 .sp
389 The output is an integer that contains the number of bytes used for internal
390 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
391 allow larger regular expressions to be compiled, at the expense of slower
392 matching. The default value of 2 is sufficient for all but the most massive
393 patterns, since it allows the compiled pattern to be up to 64K in size.
394 .sp
396 .sp
397 The output is an integer that contains the threshold above which the POSIX
398 interface uses \fBmalloc()\fP for output vectors. Further details are given in
399 the
400 .\" HREF
401 \fBpcreposix\fP
402 .\"
403 documentation.
404 .sp
406 .sp
407 The output is a long integer that gives the default limit for the number of
408 internal matching function calls in a \fBpcre_exec()\fP execution. Further
409 details are given with \fBpcre_exec()\fP below.
410 .sp
412 .sp
413 The output is a long integer that gives the default limit for the depth of
414 recursion when calling the internal matching function in a \fBpcre_exec()\fP
415 execution. Further details are given with \fBpcre_exec()\fP below.
416 .sp
418 .sp
419 The output is an integer that is set to one if internal recursion when running
420 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
421 to remember their state. This is the usual way that PCRE is compiled. The
422 output is zero if PCRE was compiled to use blocks of data on the heap instead
423 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
424 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
425 avoiding the use of the stack.
426 .
427 .
429 .rs
430 .sp
431 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
432 .ti +5n
433 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
434 .ti +5n
435 .B const unsigned char *\fItableptr\fP);
436 .sp
437 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
438 .ti +5n
439 .B int *\fIerrorcodeptr\fP,
440 .ti +5n
441 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
442 .ti +5n
443 .B const unsigned char *\fItableptr\fP);
444 .P
445 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
446 called to compile a pattern into an internal form. The only difference between
447 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
448 \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
449 too much repetition, we refer just to \fBpcre_compile()\fP below, but the
450 information applies equally to \fBpcre_compile2()\fP.
451 .P
452 The pattern is a C string terminated by a binary zero, and is passed in the
453 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
454 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
455 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
456 for a structure whose contents are not externally defined. It is up to the
457 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
458 .P
459 Although the compiled code of a PCRE regex is relocatable, that is, it does not
460 depend on memory location, the complete \fBpcre\fP data block is not
461 fully relocatable, because it may contain a copy of the \fItableptr\fP
462 argument, which is an address (see below).
463 .P
464 The \fIoptions\fP argument contains various bit settings that affect the
465 compilation. It should be zero if no options are required. The available
466 options are described below. Some of them (in particular, those that are
467 compatible with Perl, but some others as well) can also be set and unset from
468 within the pattern (see the detailed description in the
469 .\" HREF
470 \fBpcrepattern\fP
471 .\"
472 documentation). For those options that can be different in different parts of
473 the pattern, the contents of the \fIoptions\fP argument specifies their
474 settings at the start of compilation and execution. The PCRE_ANCHORED,
476 PCRE_NO_START_OPT options can be set at the time of matching as well as at
477 compile time.
478 .P
479 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
480 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
481 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
482 error message. This is a static string that is part of the library. You must
483 not try to free it. Normally, the offset from the start of the pattern to the
484 byte that was being processed when the error was discovered is placed in the
485 variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
486 immediate error is given). However, for an invalid UTF-8 string, the offset is
487 that of the first byte of the failing character. Also, some errors are not
488 detected until checks are carried out when the whole pattern has been scanned;
489 in these cases the offset passed back is the length of the pattern.
490 .P
491 Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
492 sometimes point into the middle of a UTF-8 character.
493 .P
494 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
495 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
496 returned via this argument in the event of an error. This is in addition to the
497 textual error message. Error codes and messages are listed below.
498 .P
499 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
500 character tables that are built when PCRE is compiled, using the default C
501 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
502 call to \fBpcre_maketables()\fP. This value is stored with the compiled
503 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
504 passed to it. For more discussion, see the section on locale support below.
505 .P
506 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
507 .sp
508 pcre *re;
509 const char *error;
510 int erroffset;
511 re = pcre_compile(
512 "^A.*Z", /* the pattern */
513 0, /* default options */
514 &error, /* for error message */
515 &erroffset, /* for error offset */
516 NULL); /* use default character tables */
517 .sp
518 The following names for option bits are defined in the \fBpcre.h\fP header
519 file:
520 .sp
522 .sp
523 If this bit is set, the pattern is forced to be "anchored", that is, it is
524 constrained to match only at the first matching point in the string that is
525 being searched (the "subject string"). This effect can also be achieved by
526 appropriate constructs in the pattern itself, which is the only way to do it in
527 Perl.
528 .sp
530 .sp
531 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
532 all with number 255, before each pattern item. For discussion of the callout
533 facility, see the
534 .\" HREF
535 \fBpcrecallout\fP
536 .\"
537 documentation.
538 .sp
541 .sp
542 These options (which are mutually exclusive) control what the \eR escape
543 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
544 match any Unicode newline sequence. The default is specified when PCRE is
545 built. It can be overridden from within the pattern, or by setting an option
546 when a compiled pattern is matched.
547 .sp
549 .sp
550 If this bit is set, letters in the pattern match both upper and lower case
551 letters. It is equivalent to Perl's /i option, and it can be changed within a
552 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
553 concept of case for characters whose values are less than 128, so caseless
554 matching is always possible. For characters with higher values, the concept of
555 case is supported if PCRE is compiled with Unicode property support, but not
556 otherwise. If you want to use caseless matching for characters 128 and above,
557 you must ensure that PCRE is compiled with Unicode property support as well as
558 with UTF-8 support.
559 .sp
561 .sp
562 If this bit is set, a dollar metacharacter in the pattern matches only at the
563 end of the subject string. Without this option, a dollar also matches
564 immediately before a newline at the end of the string (but not before any other
565 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
566 There is no equivalent to this option in Perl, and no way to set it within a
567 pattern.
568 .sp
570 .sp
571 If this bit is set, a dot metacharacter in the pattern matches a character of
572 any value, including one that indicates a newline. However, it only ever
573 matches one character, even if newlines are coded as CRLF. Without this option,
574 a dot does not match when the current position is at a newline. This option is
575 equivalent to Perl's /s option, and it can be changed within a pattern by a
576 (?s) option setting. A negative class such as [^a] always matches newline
577 characters, independent of the setting of this option.
578 .sp
580 .sp
581 If this bit is set, names used to identify capturing subpatterns need not be
582 unique. This can be helpful for certain types of pattern when it is known that
583 only one instance of the named subpattern can ever be matched. There are more
584 details of named subpatterns below; see also the
585 .\" HREF
586 \fBpcrepattern\fP
587 .\"
588 documentation.
589 .sp
591 .sp
592 If this bit is set, whitespace data characters in the pattern are totally
593 ignored except when escaped or inside a character class. Whitespace does not
594 include the VT character (code 11). In addition, characters between an
595 unescaped # outside a character class and the next newline, inclusive, are also
596 ignored. This is equivalent to Perl's /x option, and it can be changed within a
597 pattern by a (?x) option setting.
598 .P
599 Which characters are interpreted as newlines is controlled by the options
600 passed to \fBpcre_compile()\fP or by a special sequence at the start of the
601 pattern, as described in the section entitled
602 .\" HTML <a href="pcrepattern.html#newlines">
603 .\" </a>
604 "Newline conventions"
605 .\"
606 in the \fBpcrepattern\fP documentation. Note that the end of this type of
607 comment is a literal newline sequence in the pattern; escape sequences that
608 happen to represent a newline do not count.
609 .P
610 This option makes it possible to include comments inside complicated patterns.
611 Note, however, that this applies only to data characters. Whitespace characters
612 may never appear within special character sequences in a pattern, for example
613 within the sequence (?( that introduces a conditional subpattern.
614 .sp
616 .sp
617 This option was invented in order to turn on additional functionality of PCRE
618 that is incompatible with Perl, but it is currently of very little use. When
619 set, any backslash in a pattern that is followed by a letter that has no
620 special meaning causes an error, thus reserving these combinations for future
621 expansion. By default, as in Perl, a backslash followed by a letter with no
622 special meaning is treated as a literal. (Perl can, however, be persuaded to
623 give an error for this, by running it with the -w option.) There are at present
624 no other features controlled by this option. It can also be set by a (?X)
625 option setting within a pattern.
626 .sp
628 .sp
629 If this option is set, an unanchored pattern is required to match before or at
630 the first newline in the subject string, though the matched text may continue
631 over the newline.
632 .sp
634 .sp
635 If this option is set, PCRE's behaviour is changed in some ways so that it is
636 compatible with JavaScript rather than Perl. The changes are as follows:
637 .P
638 (1) A lone closing square bracket in a pattern causes a compile-time error,
639 because this is illegal in JavaScript (by default it is treated as a data
640 character). Thus, the pattern AB]CD becomes illegal when this option is set.
641 .P
642 (2) At run time, a back reference to an unset subpattern group matches an empty
643 string (by default this causes the current matching alternative to fail). A
644 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
645 an "a" in the subject), whereas it fails by default, for Perl compatibility.
646 .sp
648 .sp
649 By default, PCRE treats the subject string as consisting of a single line of
650 characters (even if it actually contains newlines). The "start of line"
651 metacharacter (^) matches only at the start of the string, while the "end of
652 line" metacharacter ($) matches only at the end of the string, or before a
653 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
654 Perl.
655 .P
656 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
657 match immediately following or immediately before internal newlines in the
658 subject string, respectively, as well as at the very start and end. This is
659 equivalent to Perl's /m option, and it can be changed within a pattern by a
660 (?m) option setting. If there are no newlines in a subject string, or no
661 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
662 .sp
668 .sp
669 These options override the default newline definition that was chosen when PCRE
670 was built. Setting the first or the second specifies that a newline is
671 indicated by a single character (CR or LF, respectively). Setting
672 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
673 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
674 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
675 that any Unicode newline sequence should be recognized. The Unicode newline
676 sequences are the three just mentioned, plus the single characters VT (vertical
677 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
678 separator, U+2028), and PS (paragraph separator, U+2029). The last two are
679 recognized only in UTF-8 mode.
680 .P
681 The newline setting in the options word uses three bits that are treated
682 as a number, giving eight possibilities. Currently only six are used (default
683 plus the five values above). This means that if you set more than one newline
684 option, the combination may or may not be sensible. For example,
686 other combinations may yield unused numbers and cause an error.
687 .P
688 The only time that a line break in a pattern is specially recognized when
689 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
690 and so are ignored in this mode. Also, an unescaped # outside a character class
691 indicates a comment that lasts until after the next line break sequence. In
692 other circumstances, line break sequences in patterns are treated as literal
693 data.
694 .P
695 The newline option that is set at compile time becomes the default that is used
696 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
697 .sp
699 .sp
700 If this option is set, it disables the use of numbered capturing parentheses in
701 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
702 were followed by ?: but named parentheses can still be used for capturing (and
703 they acquire numbers in the usual way). There is no equivalent of this option
704 in Perl.
705 .sp
707 .sp
708 This is an option that acts at matching time; that is, it is really an option
709 for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
710 it is remembered with the compiled pattern and assumed at matching time. For
711 details see the discussion of PCRE_NO_START_OPTIMIZE
712 .\" HTML <a href="#execoptions">
713 .\" </a>
714 below.
715 .\"
716 .sp
718 .sp
719 This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
720 \ew, and some of the POSIX character classes. By default, only ASCII characters
721 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
722 classify characters. More details are given in the section on
723 .\" HTML <a href="pcre.html#genericchartypes">
724 .\" </a>
725 generic character types
726 .\"
727 in the
728 .\" HREF
729 \fBpcrepattern\fP
730 .\"
731 page. If you set PCRE_UCP, matching one of the items it affects takes much
732 longer. The option is available only if PCRE has been compiled with Unicode
733 property support.
734 .sp
736 .sp
737 This option inverts the "greediness" of the quantifiers so that they are not
738 greedy by default, but become greedy if followed by "?". It is not compatible
739 with Perl. It can also be set by a (?U) option setting within the pattern.
740 .sp
742 .sp
743 This option causes PCRE to regard both the pattern and the subject as strings
744 of UTF-8 characters instead of single-byte character strings. However, it is
745 available only when PCRE is built to include UTF-8 support. If not, the use
746 of this option provokes an error. Details of how this option changes the
747 behaviour of PCRE are given in the
748 .\" HREF
749 \fBpcreunicode\fP
750 .\"
751 page.
752 .sp
754 .sp
755 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
756 automatically checked. There is a discussion about the
757 .\" HTML <a href="pcre.html#utf8strings">
758 .\" </a>
759 validity of UTF-8 strings
760 .\"
761 in the main
762 .\" HREF
763 \fBpcre\fP
764 .\"
765 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
766 returns an error. If you already know that your pattern is valid, and you want
767 to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
768 option. When it is set, the effect of passing an invalid UTF-8 string as a
769 pattern is undefined. It may cause your program to crash. Note that this option
770 can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
771 the UTF-8 validity checking of subject strings.
772 .
773 .
775 .rs
776 .sp
777 The following table lists the error codes than may be returned by
778 \fBpcre_compile2()\fP, along with the error messages that may be returned by
779 both compiling functions. As PCRE has developed, some error codes have fallen
780 out of use. To avoid confusion, they have not been re-used.
781 .sp
782 0 no error
783 1 \e at end of pattern
784 2 \ec at end of pattern
785 3 unrecognized character follows \e
786 4 numbers out of order in {} quantifier
787 5 number too big in {} quantifier
788 6 missing terminating ] for character class
789 7 invalid escape sequence in character class
790 8 range out of order in character class
791 9 nothing to repeat
792 10 [this code is not in use]
793 11 internal error: unexpected repeat
794 12 unrecognized character after (? or (?-
795 13 POSIX named classes are supported only within a class
796 14 missing )
797 15 reference to non-existent subpattern
798 16 erroffset passed as NULL
799 17 unknown option bit(s) set
800 18 missing ) after comment
801 19 [this code is not in use]
802 20 regular expression is too large
803 21 failed to get memory
804 22 unmatched parentheses
805 23 internal error: code overflow
806 24 unrecognized character after (?<
807 25 lookbehind assertion is not fixed length
808 26 malformed number or name after (?(
809 27 conditional group contains more than two branches
810 28 assertion expected after (?(
811 29 (?R or (?[+-]digits must be followed by )
812 30 unknown POSIX class name
813 31 POSIX collating elements are not supported
814 32 this version of PCRE is not compiled with PCRE_UTF8 support
815 33 [this code is not in use]
816 34 character value in \ex{...} sequence is too large
817 35 invalid condition (?(0)
818 36 \eC not allowed in lookbehind assertion
819 37 PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
820 38 number after (?C is > 255
821 39 closing ) for (?C expected
822 40 recursive call could loop indefinitely
823 41 unrecognized character after (?P
824 42 syntax error in subpattern name (missing terminator)
825 43 two named subpatterns have the same name
826 44 invalid UTF-8 string
827 45 support for \eP, \ep, and \eX has not been compiled
828 46 malformed \eP or \ep sequence
829 47 unknown property name after \eP or \ep
830 48 subpattern name is too long (maximum 32 characters)
831 49 too many named subpatterns (maximum 10000)
832 50 [this code is not in use]
833 51 octal value is greater than \e377 (not in UTF-8 mode)
834 52 internal error: overran compiling workspace
835 53 internal error: previously-checked referenced subpattern
836 not found
837 54 DEFINE group contains more than one branch
838 55 repeating a DEFINE group is not allowed
839 56 inconsistent NEWLINE options
840 57 \eg is not followed by a braced, angle-bracketed, or quoted
841 name/number or by a plain number
842 58 a numbered reference must not be zero
843 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
844 60 (*VERB) not recognized
845 61 number is too big
846 62 subpattern name expected
847 63 digit expected after (?+
848 64 ] is an invalid data character in JavaScript compatibility mode
849 65 different names for subpatterns of the same number are
850 not allowed
851 66 (*MARK) must have an argument
852 67 this version of PCRE is not compiled with PCRE_UCP support
853 68 \ec must be followed by an ASCII character
854 69 \ek is not followed by a braced, angle-bracketed, or quoted name
855 .sp
856 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
857 be used if the limits were changed when PCRE was built.
858 .
859 .
860 .\" HTML <a name="studyingapattern"></a>
862 .rs
863 .sp
864 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
865 .ti +5n
866 .B const char **\fIerrptr\fP);
867 .PP
868 If a compiled pattern is going to be used several times, it is worth spending
869 more time analyzing it in order to speed up the time taken for matching. The
870 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
871 argument. If studying the pattern produces additional information that will
872 help speed up matching, \fBpcre_study()\fP returns a pointer to a
873 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
874 results of the study.
875 .P
876 The returned value from \fBpcre_study()\fP can be passed directly to
877 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
878 also contains other fields that can be set by the caller before the block is
879 passed; these are described
880 .\" HTML <a href="#extradata">
881 .\" </a>
882 below
883 .\"
884 in the section on matching a pattern.
885 .P
886 If studying the pattern does not produce any useful information,
887 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
888 wants to pass any of the other fields to \fBpcre_exec()\fP or
889 \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
890 .P
891 The second argument of \fBpcre_study()\fP contains option bits. There is only
892 one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
893 compiler is available, the pattern is further compiled into machine code that
894 executes much faster than the \fBpcre_exec()\fP matching function. If
895 the just-in-time compiler is not available, this option is ignored. All other
896 bits in the \fIoptions\fP argument must be zero.
897 .P
898 JIT compilation is a heavyweight optimization. It can take some time for
899 patterns to be analyzed, and for one-off matches and simple patterns the
900 benefit of faster execution might be offset by a much slower study time.
901 Not all patterns can be optimized by the JIT compiler. For those that cannot be
902 handled, matching automatically falls back to the \fBpcre_exec()\fP
903 interpreter. For more details, see the
904 .\" HREF
905 \fBpcrejit\fP
906 .\"
907 documentation.
908 .P
909 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
910 studying succeeds (even if no data is returned), the variable it points to is
911 set to NULL. Otherwise it is set to point to a textual error message. This is a
912 static string that is part of the library. You must not try to free it. You
913 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
914 sure that it has run successfully.
915 .P
916 When you are finished with a pattern, you can free the memory used for the
917 study data by calling \fBpcre_free_study()\fP. This function was added to the
918 API for release 8.20. For earlier versions, the memory could be freed with
919 \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
920 where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
921 new function when convenient.
922 .P
923 This is a typical way in which \fBpcre_study\fP() is used (except that in a
924 real application there should be tests for errors):
925 .sp
926 int rc;
927 pcre *re;
928 pcre_extra *sd;
929 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
930 sd = pcre_study(
931 re, /* result of pcre_compile() */
932 0, /* no options */
933 &error); /* set to NULL or points to a message */
934 rc = pcre_exec( /* see below for details of pcre_exec() options */
935 re, sd, "subject", 7, 0, 0, ovector, 30);
936 ...
937 pcre_free_study(sd);
938 pcre_free(re);
939 .sp
940 Studying a pattern does two things: first, a lower bound for the length of
941 subject string that is needed to match the pattern is computed. This does not
942 mean that there are any strings of that length that match, but it does
943 guarantee that no shorter strings match. The value is used by
944 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
945 match strings that are shorter than the lower bound. You can find out the value
946 in a calling program via the \fBpcre_fullinfo()\fP function.
947 .P
948 Studying a pattern is also useful for non-anchored patterns that do not have a
949 single fixed starting character. A bitmap of possible starting bytes is
950 created. This speeds up finding a position in the subject at which to start
951 matching.
952 .P
953 These two optimizations apply to both \fBpcre_exec()\fP and
954 \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
955 \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
956 just-in-time compiling is successful. The optimizations can be disabled by
957 setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
958 \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
959 callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
960 to make use of these facilities in cases where matching fails. See the
961 discussion of PCRE_NO_START_OPTIMIZE
962 .\" HTML <a href="#execoptions">
963 .\" </a>
964 below.
965 .\"
966 .
967 .
968 .\" HTML <a name="localesupport"></a>
970 .rs
971 .sp
972 PCRE handles caseless matching, and determines whether characters are letters,
973 digits, or whatever, by reference to a set of tables, indexed by character
974 value. When running in UTF-8 mode, this applies only to characters with codes
975 less than 128. By default, higher-valued codes never match escapes such as \ew
976 or \ed, but they can be tested with \ep if PCRE is built with Unicode character
977 property support. Alternatively, the PCRE_UCP option can be set at compile
978 time; this causes \ew and friends to use Unicode property support instead of
979 built-in tables. The use of locales with Unicode is discouraged. If you are
980 handling characters with codes greater than 128, you should either use UTF-8
981 and Unicode, or use locales, but not try to mix the two.
982 .P
983 PCRE contains an internal set of tables that are used when the final argument
984 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
985 Normally, the internal tables recognize only ASCII characters. However, when
986 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
987 default "C" locale of the local system, which may cause them to be different.
988 .P
989 The internal tables can always be overridden by tables supplied by the
990 application that calls PCRE. These may be created in a different locale from
991 the default. As more and more applications change to using Unicode, the need
992 for this locale support is expected to die away.
993 .P
994 External tables are built by calling the \fBpcre_maketables()\fP function,
995 which has no arguments, in the relevant locale. The result can then be passed
996 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
997 example, to build and use tables that are appropriate for the French locale
998 (where accented characters with values greater than 128 are treated as letters),
999 the following code could be used:
1000 .sp
1001 setlocale(LC_CTYPE, "fr_FR");
1002 tables = pcre_maketables();
1003 re = pcre_compile(..., tables);
1004 .sp
1005 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1006 are using Windows, the name for the French locale is "french".
1007 .P
1008 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1009 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1010 that the memory containing the tables remains available for as long as it is
1011 needed.
1012 .P
1013 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
1014 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
1015 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
1016 pattern, compilation, studying and matching all happen in the same locale, but
1017 different patterns can be compiled in different locales.
1018 .P
1019 It is possible to pass a table pointer or NULL (indicating the use of the
1020 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
1021 this facility could be used to match a pattern in a different locale from the
1022 one in which it was compiled. Passing table pointers at run time is discussed
1023 below in the section on matching a pattern.
1024 .
1025 .
1026 .\" HTML <a name="infoaboutpattern"></a>
1028 .rs
1029 .sp
1030 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1031 .ti +5n
1032 .B int \fIwhat\fP, void *\fIwhere\fP);
1033 .PP
1034 The \fBpcre_fullinfo()\fP function returns information about a compiled
1035 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
1036 nevertheless retained for backwards compability (and is documented below).
1037 .P
1038 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1039 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
1040 the pattern was not studied. The third argument specifies which piece of
1041 information is required, and the fourth argument is a pointer to a variable
1042 to receive the data. The yield of the function is zero for success, or one of
1043 the following negative numbers:
1044 .sp
1045 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1046 the argument \fIwhere\fP was NULL
1047 PCRE_ERROR_BADMAGIC the "magic number" was not found
1048 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
1049 .sp
1050 The "magic number" is placed at the start of each compiled pattern as an simple
1051 check against passing an arbitrary memory pointer. Here is a typical call of
1052 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
1053 .sp
1054 int rc;
1055 size_t length;
1056 rc = pcre_fullinfo(
1057 re, /* result of pcre_compile() */
1058 sd, /* result of pcre_study(), or NULL */
1059 PCRE_INFO_SIZE, /* what is required */
1060 &length); /* where to put the data */
1061 .sp
1062 The possible values for the third argument are defined in \fBpcre.h\fP, and are
1063 as follows:
1064 .sp
1066 .sp
1067 Return the number of the highest back reference in the pattern. The fourth
1068 argument should point to an \fBint\fP variable. Zero is returned if there are
1069 no back references.
1070 .sp
1072 .sp
1073 Return the number of capturing subpatterns in the pattern. The fourth argument
1074 should point to an \fBint\fP variable.
1075 .sp
1077 .sp
1078 Return a pointer to the internal default character tables within PCRE. The
1079 fourth argument should point to an \fBunsigned char *\fP variable. This
1080 information call is provided for internal use by the \fBpcre_study()\fP
1081 function. External callers can cause PCRE to use its internal tables by passing
1082 a NULL table pointer.
1083 .sp
1085 .sp
1086 Return information about the first byte of any matched string, for a
1087 non-anchored pattern. The fourth argument should point to an \fBint\fP
1088 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1089 still recognized for backwards compatibility.)
1090 .P
1091 If there is a fixed first byte, for example, from a pattern such as
1092 (cat|cow|coyote), its value is returned. Otherwise, if either
1093 .sp
1094 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1095 starts with "^", or
1096 .sp
1097 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1098 (if it were set, the pattern would be anchored),
1099 .sp
1100 -1 is returned, indicating that the pattern matches only at the start of a
1101 subject string or after any newline within the string. Otherwise -2 is
1102 returned. For anchored patterns, -2 is returned.
1103 .sp
1105 .sp
1106 If the pattern was studied, and this resulted in the construction of a 256-bit
1107 table indicating a fixed set of bytes for the first byte in any matching
1108 string, a pointer to the table is returned. Otherwise NULL is returned. The
1109 fourth argument should point to an \fBunsigned char *\fP variable.
1110 .sp
1112 .sp
1113 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1114 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1115 explicit match is either a literal CR or LF character, or \er or \en.
1116 .sp
1118 .sp
1119 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1120 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1121 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1122 .sp
1124 .sp
1125 Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1126 just-in-time compiling was successful. The fourth argument should point to an
1127 \fBint\fP variable. A return value of 0 means that JIT support is not available
1128 in this version of PCRE, or that the pattern was not studied with the
1129 PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1130 particular pattern. See the
1131 .\" HREF
1132 \fBpcrejit\fP
1133 .\"
1134 documentation for details of what can and cannot be handled.
1135 .sp
1137 .sp
1138 Return the value of the rightmost literal byte that must exist in any matched
1139 string, other than at its start, if such a byte has been recorded. The fourth
1140 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1141 returned. For anchored patterns, a last literal byte is recorded only if it
1142 follows something of variable length. For example, for the pattern
1143 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1144 is -1.
1145 .sp
1147 .sp
1148 If the pattern was studied and a minimum length for matching subject strings
1149 was computed, its value is returned. Otherwise the returned value is -1. The
1150 value is a number of characters, not bytes (this may be relevant in UTF-8
1151 mode). The fourth argument should point to an \fBint\fP variable. A
1152 non-negative value is a lower bound to the length of any matching string. There
1153 may not be any strings of that length that do actually match, but every string
1154 that does match is at least that long.
1155 .sp
1159 .sp
1160 PCRE supports the use of named as well as numbered capturing parentheses. The
1161 names are just an additional way of identifying the parentheses, which still
1162 acquire numbers. Several convenience functions such as
1163 \fBpcre_get_named_substring()\fP are provided for extracting captured
1164 substrings by name. It is also possible to extract the data directly, by first
1165 converting the name to a number in order to access the correct pointers in the
1166 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1167 you need to use the name-to-number map, which is described by these three
1168 values.
1169 .P
1170 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1171 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1172 entry; both of these return an \fBint\fP value. The entry size depends on the
1173 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1174 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1175 are the number of the capturing parenthesis, most significant byte first. The
1176 rest of the entry is the corresponding name, zero terminated.
1177 .P
1178 The names are in alphabetical order. Duplicate names may appear if (?| is used
1179 to create multiple groups with the same number, as described in the
1180 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1181 .\" </a>
1182 section on duplicate subpattern numbers
1183 .\"
1184 in the
1185 .\" HREF
1186 \fBpcrepattern\fP
1187 .\"
1188 page. Duplicate names for subpatterns with different numbers are permitted only
1189 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1190 table in the order in which they were found in the pattern. In the absence of
1191 (?| this is the order of increasing number; when (?| is used this is not
1192 necessarily the case because later subpatterns may have lower numbers.
1193 .P
1194 As a simple example of the name/number table, consider the following pattern
1195 (assume PCRE_EXTENDED is set, so white space - including newlines - is
1196 ignored):
1197 .sp
1198 .\" JOIN
1199 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1200 (?<month>\ed\ed) - (?<day>\ed\ed) )
1201 .sp
1202 There are four named subpatterns, so the table has four entries, and each entry
1203 in the table is eight bytes long. The table is as follows, with non-printing
1204 bytes shows in hexadecimal, and undefined bytes shown as ??:
1205 .sp
1206 00 01 d a t e 00 ??
1207 00 05 d a y 00 ?? ??
1208 00 04 m o n t h 00
1209 00 02 y e a r 00 ??
1210 .sp
1211 When writing code to extract data from named subpatterns using the
1212 name-to-number map, remember that the length of the entries is likely to be
1213 different for each compiled pattern.
1214 .sp
1216 .sp
1217 Return 1 if the pattern can be used for partial matching with
1218 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1219 \fBint\fP variable. From release 8.00, this always returns 1, because the
1220 restrictions that previously applied to partial matching have been lifted. The
1221 .\" HREF
1222 \fBpcrepartial\fP
1223 .\"
1224 documentation gives details of partial matching.
1225 .sp
1227 .sp
1228 Return a copy of the options with which the pattern was compiled. The fourth
1229 argument should point to an \fBunsigned long int\fP variable. These option bits
1230 are those specified in the call to \fBpcre_compile()\fP, modified by any
1231 top-level option settings at the start of the pattern itself. In other words,
1232 they are the options that will be in force when matching starts. For example,
1233 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1235 .P
1236 A pattern is automatically anchored by PCRE if all of its top-level
1237 alternatives begin with one of the following:
1238 .sp
1239 ^ unless PCRE_MULTILINE is set
1240 \eA always
1241 \eG always
1242 .\" JOIN
1243 .* if PCRE_DOTALL is set and there are no back
1244 references to the subpattern in which .* appears
1245 .sp
1246 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1247 \fBpcre_fullinfo()\fP.
1248 .sp
1250 .sp
1251 Return the size of the compiled pattern, that is, the value that was passed as
1252 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1253 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1254 variable.
1255 .sp
1257 .sp
1258 Return the size of the data block pointed to by the \fIstudy_data\fP field in a
1259 \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
1260 zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
1261 The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
1262 that will speed up matching (see the section entitled
1263 .\" HTML <a href="#studyingapattern">
1264 .\" </a>
1265 "Studying a pattern"
1266 .\"
1267 above). The format of the \fIstudy_data\fP block is private, but its length
1268 is made available via this option so that it can be saved and restored (see the
1269 .\" HREF
1270 \fBpcreprecompile\fP
1271 .\"
1272 documentation for details).
1273 .
1274 .
1276 .rs
1277 .sp
1278 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1279 .B *\fIfirstcharptr\fP);
1280 .PP
1281 The \fBpcre_info()\fP function is now obsolete because its interface is too
1282 restrictive to return all the available data about a compiled pattern. New
1283 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1284 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1285 following negative numbers:
1286 .sp
1287 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1288 PCRE_ERROR_BADMAGIC the "magic number" was not found
1289 .sp
1290 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1291 pattern was compiled is placed in the integer it points to (see
1292 PCRE_INFO_OPTIONS above).
1293 .P
1294 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1295 it is used to pass back information about the first character of any matched
1296 string (see PCRE_INFO_FIRSTBYTE above).
1297 .
1298 .
1300 .rs
1301 .sp
1302 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1303 .PP
1304 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1305 data block that contains a compiled pattern. It is provided for the benefit of
1306 applications that operate in an object-oriented manner, where different parts
1307 of the application may be using the same compiled pattern, but you want to free
1308 the block when they are all done.
1309 .P
1310 When a pattern is compiled, the reference count field is initialized to zero.
1311 It is changed only by calling this function, whose action is to add the
1312 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1313 function is the new value. However, the value of the count is constrained to
1314 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1315 it is forced to the appropriate limit value.
1316 .P
1317 Except when it is zero, the reference count is not correctly preserved if a
1318 pattern is compiled on one host and then transferred to a host whose byte-order
1319 is different. (This seems a highly unlikely scenario.)
1320 .
1321 .
1323 .rs
1324 .sp
1325 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1326 .ti +5n
1327 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1328 .ti +5n
1329 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1330 .P
1331 The function \fBpcre_exec()\fP is called to match a subject string against a
1332 compiled pattern, which is passed in the \fIcode\fP argument. If the
1333 pattern was studied, the result of the study should be passed in the
1334 \fIextra\fP argument. This function is the main matching facility of the
1335 library, and it operates in a Perl-like manner. For specialist use there is
1336 also an alternative matching function, which is described
1337 .\" HTML <a href="#dfamatch">
1338 .\" </a>
1339 below
1340 .\"
1341 in the section about the \fBpcre_dfa_exec()\fP function.
1342 .P
1343 In most applications, the pattern will have been compiled (and optionally
1344 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1345 possible to save compiled patterns and study data, and then use them later
1346 in different processes, possibly even on different hosts. For a discussion
1347 about this, see the
1348 .\" HREF
1349 \fBpcreprecompile\fP
1350 .\"
1351 documentation.
1352 .P
1353 Here is an example of a simple call to \fBpcre_exec()\fP:
1354 .sp
1355 int rc;
1356 int ovector[30];
1357 rc = pcre_exec(
1358 re, /* result of pcre_compile() */
1359 NULL, /* we didn't study the pattern */
1360 "some string", /* the subject string */
1361 11, /* the length of the subject string */
1362 0, /* start at offset 0 in the subject */
1363 0, /* default options */
1364 ovector, /* vector of integers for substring information */
1365 30); /* number of elements (NOT size in bytes) */
1366 .
1367 .
1368 .\" HTML <a name="extradata"></a>
1369 .SS "Extra data for \fBpcre_exec()\fR"
1370 .rs
1371 .sp
1372 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1373 data block. The \fBpcre_study()\fP function returns such a block (when it
1374 doesn't return NULL), but you can also create one for yourself, and pass
1375 additional information in it. The \fBpcre_extra\fP block contains the following
1376 fields (not necessarily in this order):
1377 .sp
1378 unsigned long int \fIflags\fP;
1379 void *\fIstudy_data\fP;
1380 void *\fIexecutable_jit\fP;
1381 unsigned long int \fImatch_limit\fP;
1382 unsigned long int \fImatch_limit_recursion\fP;
1383 void *\fIcallout_data\fP;
1384 const unsigned char *\fItables\fP;
1385 unsigned char **\fImark\fP;
1386 .sp
1387 The \fIflags\fP field is a bitmap that specifies which of the other fields
1388 are set. The flag bits are:
1389 .sp
1397 .sp
1398 Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1399 the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1400 returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1401 should not set these yourself, but you may add to the block by setting the
1402 other fields and their corresponding flag bits.
1403 .P
1404 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1405 vast amount of resources when running patterns that are not going to match,
1406 but which have a very large number of possibilities in their search trees. The
1407 classic example is a pattern that uses nested unlimited repeats.
1408 .P
1409 Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1410 calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1411 imposed on the number of times this function is called during a match, which
1412 has the effect of limiting the amount of backtracking that can take place. For
1413 patterns that are not anchored, the count restarts from zero for each position
1414 in the subject string.
1415 .P
1416 When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1417 with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1418 is entirely different. However, there is still the possibility of runaway
1419 matching that goes on for a very long time, and so the \fImatch_limit\fP value
1420 is also used in this case (but in a different way) to limit how long the
1421 matching can continue.
1422 .P
1423 The default value for the limit can be set when PCRE is built; the default
1424 default is 10 million, which handles all but the most extreme cases. You can
1425 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1426 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1427 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1429 .P
1430 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1431 instead of limiting the total number of times that \fBmatch()\fP is called, it
1432 limits the depth of recursion. The recursion depth is a smaller number than the
1433 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1434 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1435 .P
1436 Limiting the recursion depth limits the amount of machine stack that can be
1437 used, or, when PCRE has been compiled to use memory on the heap instead of the
1438 stack, the amount of heap memory that can be used. This limit is not relevant,
1439 and is ignored, if the pattern was successfully studied with
1441 .P
1442 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1443 built; the default default is the same value as the default for
1444 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1445 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1446 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1447 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1448 .P
1449 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1450 and is described in the
1451 .\" HREF
1452 \fBpcrecallout\fP
1453 .\"
1454 documentation.
1455 .P
1456 The \fItables\fP field is used to pass a character tables pointer to
1457 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1458 pattern. A non-NULL value is stored with the compiled pattern only if custom
1459 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1460 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1461 internal tables to be used. This facility is helpful when re-using patterns
1462 that have been saved after compiling with an external set of tables, because
1463 the external tables might be at a different address when \fBpcre_exec()\fP is
1464 called. See the
1465 .\" HREF
1466 \fBpcreprecompile\fP
1467 .\"
1468 documentation for a discussion of saving compiled patterns for later use.
1469 .P
1470 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1471 be set to point to a \fBchar *\fP variable. If the pattern contains any
1472 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1473 a name to pass back, a pointer to the name string (zero terminated) is placed
1474 in the variable pointed to by the \fImark\fP field. The names are within the
1475 compiled pattern; if you wish to retain such a name you must copy it before
1476 freeing the memory of a compiled pattern. If there is no name to pass back, the
1477 variable pointed to by the \fImark\fP field set to NULL. For details of the
1478 backtracking control verbs, see the section entitled
1479 .\" HTML <a href="pcrepattern#backtrackcontrol">
1480 .\" </a>
1481 "Backtracking control"
1482 .\"
1483 in the
1484 .\" HREF
1485 \fBpcrepattern\fP
1486 .\"
1487 documentation.
1488 .
1489 .
1490 .\" HTML <a name="execoptions"></a>
1491 .SS "Option bits for \fBpcre_exec()\fP"
1492 .rs
1493 .sp
1494 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1495 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1499 .P
1500 If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1501 the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1503 particular that partial matching is not supported. If an unsupported option is
1504 used, JIT execution is disabled and the normal interpretive code in
1505 \fBpcre_exec()\fP is run.
1506 .sp
1508 .sp
1509 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1510 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1511 to be anchored by virtue of its contents, it cannot be made unachored at
1512 matching time.
1513 .sp
1516 .sp
1517 These options (which are mutually exclusive) control what the \eR escape
1518 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1519 match any Unicode newline sequence. These options override the choice that was
1520 made or defaulted when the pattern was compiled.
1521 .sp
1527 .sp
1528 These options override the newline definition that was chosen or defaulted when
1529 the pattern was compiled. For details, see the description of
1530 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1531 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1532 the way the match position is advanced after a match failure for an unanchored
1533 pattern.
1534 .P
1536 match attempt for an unanchored pattern fails when the current position is at a
1537 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1538 characters, the match position is advanced by two characters instead of one, in
1539 other words, to after the CRLF.
1540 .P
1541 The above rule is a compromise that makes the most common cases work as
1542 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1543 set), it does not match the string "\er\enA" because, after failing at the
1544 start, it skips both the CR and the LF before retrying. However, the pattern
1545 [\er\en]A does match that string, because it contains an explicit CR or LF
1546 reference, and so advances only by one character after the first failure.
1547 .P
1548 An explicit match for CR of LF is either a literal appearance of one of those
1549 characters, or one of the \er or \en escape sequences. Implicit matches such as
1550 [^X] do not count, nor does \es (which includes CR and LF in the characters
1551 that it matches).
1552 .P
1553 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1554 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1555 .sp
1557 .sp
1558 This option specifies that first character of the subject string is not the
1559 beginning of a line, so the circumflex metacharacter should not match before
1560 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1561 never to match. This option affects only the behaviour of the circumflex
1562 metacharacter. It does not affect \eA.
1563 .sp
1565 .sp
1566 This option specifies that the end of the subject string is not the end of a
1567 line, so the dollar metacharacter should not match it nor (except in multiline
1568 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1569 compile time) causes dollar never to match. This option affects only the
1570 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1571 .sp
1573 .sp
1574 An empty string is not considered to be a valid match if this option is set. If
1575 there are alternatives in the pattern, they are tried. If all the alternatives
1576 match the empty string, the entire match fails. For example, if the pattern
1577 .sp
1578 a?b?
1579 .sp
1580 is applied to a string not beginning with "a" or "b", it matches an empty
1581 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1582 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1583 .sp
1585 .sp
1586 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1587 the start of the subject is permitted. If the pattern is anchored, such a match
1588 can occur only if the pattern contains \eK.
1589 .P
1590 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1591 does make a special case of a pattern match of the empty string within its
1592 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1593 emulate Perl's behaviour after matching a null string by first trying the match
1594 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1595 if that fails, by advancing the starting offset (see below) and trying an
1596 ordinary match again. There is some code that demonstrates how to do this in
1597 the
1598 .\" HREF
1599 \fBpcredemo\fP
1600 .\"
1601 sample program. In the most general case, you have to check to see if the
1602 newline convention recognizes CRLF as a newline, and if so, and the current
1603 character is CR followed by LF, advance the starting offset by two characters
1604 instead of one.
1605 .sp
1607 .sp
1608 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1609 a match, in order to speed up the process. For example, if it is known that an
1610 unanchored match must start with a specific character, it searches the subject
1611 for that character, and fails immediately if it cannot find it, without
1612 actually running the main matching function. This means that a special item
1613 such as (*COMMIT) at the start of a pattern is not considered until after a
1614 suitable starting point for the match has been found. When callouts or (*MARK)
1615 items are in use, these "start-up" optimizations can cause them to be skipped
1616 if the pattern is never actually used. The start-up optimizations are in effect
1617 a pre-scan of the subject that takes place before the pattern is run.
1618 .P
1619 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1620 causing performance to suffer, but ensuring that in cases where the result is
1621 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1622 are considered at every possible starting position in the subject string. If
1623 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1624 time.
1625 .P
1626 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1627 Consider the pattern
1628 .sp
1630 .sp
1631 When this is compiled, PCRE records the fact that a match must start with the
1632 character "A". Suppose the subject string is "DEFABC". The start-up
1633 optimization scans along the subject, finds "A" and runs the first match
1634 attempt from there. The (*COMMIT) item means that the pattern must match the
1635 current starting position, which in this case, it does. However, if the same
1636 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1637 subject string does not happen. The first match attempt is run starting from
1638 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1639 the overall result is "no match". If the pattern is studied, more start-up
1640 optimizations may be used. For example, a minimum length for the subject may be
1641 recorded. Consider the pattern
1642 .sp
1643 (*MARK:A)(X|Y)
1644 .sp
1645 The minimum length for a match is one character. If the subject is "ABC", there
1646 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1647 If the pattern is studied, the final attempt does not take place, because PCRE
1648 knows that the subject is too short, and so the (*MARK) is never encountered.
1649 In this case, studying the pattern does not affect the overall match result,
1650 which is still "no match", but it does affect the auxiliary information that is
1651 returned.
1652 .sp
1654 .sp
1655 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1656 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1657 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1658 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1659 strings in the
1660 .\" HTML <a href="pcre.html#utf8strings">
1661 .\" </a>
1662 section on UTF-8 support
1663 .\"
1664 in the main
1665 .\" HREF
1666 \fBpcre\fP
1667 .\"
1668 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1669 the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1670 a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1671 both cases, information about the precise nature of the error may also be
1672 returned (see the descriptions of these errors in the section entitled \fIError
1673 return values from\fP \fBpcre_exec()\fP
1674 .\" HTML <a href="#errorlist">
1675 .\" </a>
1676 below).
1677 .\"
1678 If \fIstartoffset\fP contains a value that does not point to the start of a
1679 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1680 returned.
1681 .P
1682 If you already know that your subject is valid, and you want to skip these
1683 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1684 calling \fBpcre_exec()\fP. You might want to do this for the second and
1685 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1686 all the matches in a single subject string. However, you should be sure that
1687 the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1688 end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1689 invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1690 undefined. Your program may crash.
1691 .sp
1694 .sp
1695 These options turn on the partial matching feature. For backwards
1696 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1697 occurs if the end of the subject string is reached successfully, but there are
1698 not enough subject characters to complete the match. If this happens when
1699 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1700 testing any remaining alternatives. Only if no complete match can be found is
1701 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1702 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1703 but only if no complete match can be found.
1704 .P
1705 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1706 partial match is found, \fBpcre_exec()\fP immediately returns
1707 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1708 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1709 important that an alternative complete match.
1710 .P
1711 In both cases, the portion of the string that was inspected when the partial
1712 match was found is set as the first matching string. There is a more detailed
1713 discussion of partial and multi-segment matching, with examples, in the
1714 .\" HREF
1715 \fBpcrepartial\fP
1716 .\"
1717 documentation.
1718 .
1719 .
1720 .SS "The string to be matched by \fBpcre_exec()\fP"
1721 .rs
1722 .sp
1723 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1724 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1725 in \fIstartoffset\fP. If this is negative or greater than the length of the
1726 subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1727 offset is zero, the search for a match starts at the beginning of the subject,
1728 and this is by far the most common case. In UTF-8 mode, the byte offset must
1729 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1730 pattern string, the subject may contain binary zero bytes.
1731 .P
1732 A non-zero starting offset is useful when searching for another match in the
1733 same subject by calling \fBpcre_exec()\fP again after a previous success.
1734 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1735 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1736 lookbehind. For example, consider the pattern
1737 .sp
1738 \eBiss\eB
1739 .sp
1740 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1741 the current position in the subject is not a word boundary.) When applied to
1742 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1743 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1744 subject, namely "issipi", it does not match, because \eB is always false at the
1745 start of the subject, which is deemed to be a word boundary. However, if
1746 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1747 set to 4, it finds the second occurrence of "iss" because it is able to look
1748 behind the starting point to discover that it is preceded by a letter.
1749 .P
1750 Finding all the matches in a subject is tricky when the pattern can match an
1751 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1752 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1753 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1754 and trying an ordinary match again. There is some code that demonstrates how to
1755 do this in the
1756 .\" HREF
1757 \fBpcredemo\fP
1758 .\"
1759 sample program. In the most general case, you have to check to see if the
1760 newline convention recognizes CRLF as a newline, and if so, and the current
1761 character is CR followed by LF, advance the starting offset by two characters
1762 instead of one.
1763 .P
1764 If a non-zero starting offset is passed when the pattern is anchored, one
1765 attempt to match at the given offset is made. This can only succeed if the
1766 pattern does not require the match to be at the start of the subject.
1767 .
1768 .
1769 .SS "How \fBpcre_exec()\fP returns captured substrings"
1770 .rs
1771 .sp
1772 In general, a pattern matches a certain portion of the subject, and in
1773 addition, further substrings from the subject may be picked out by parts of the
1774 pattern. Following the usage in Jeffrey Friedl's book, this is called
1775 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1776 a fragment of a pattern that picks out a substring. PCRE supports several other
1777 kinds of parenthesized subpattern that do not cause substrings to be captured.
1778 .P
1779 Captured substrings are returned to the caller via a vector of integers whose
1780 address is passed in \fIovector\fP. The number of elements in the vector is
1781 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1782 argument is NOT the size of \fIovector\fP in bytes.
1783 .P
1784 The first two-thirds of the vector is used to pass back captured substrings,
1785 each substring using a pair of integers. The remaining third of the vector is
1786 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1787 and is not available for passing back information. The number passed in
1788 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1789 rounded down.
1790 .P
1791 When a match is successful, information about captured substrings is returned
1792 in pairs of integers, starting at the beginning of \fIovector\fP, and
1793 continuing up to two-thirds of its length at the most. The first element of
1794 each pair is set to the byte offset of the first character in a substring, and
1795 the second is set to the byte offset of the first character after the end of a
1796 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1797 mode. They are not character counts.
1798 .P
1799 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1800 portion of the subject string matched by the entire pattern. The next pair is
1801 used for the first capturing subpattern, and so on. The value returned by
1802 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1803 For example, if two substrings have been captured, the returned value is 3. If
1804 there are no capturing subpatterns, the return value from a successful match is
1805 1, indicating that just the first pair of offsets has been set.
1806 .P
1807 If a capturing subpattern is matched repeatedly, it is the last portion of the
1808 string that it matched that is returned.
1809 .P
1810 If the vector is too small to hold all the captured substring offsets, it is
1811 used as far as possible (up to two-thirds of its length), and the function
1812 returns a value of zero. If neither the actual string matched not any captured
1813 substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
1814 passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
1815 back references and the \fIovector\fP is not big enough to remember the related
1816 substrings, PCRE has to get additional memory for use during matching. Thus it
1817 is usually advisable to supply an \fIovector\fP of reasonable size.
1818 .P
1819 There are some cases where zero is returned (indicating vector overflow) when
1820 in fact the vector is exactly the right size for the final match. For example,
1821 consider the pattern
1822 .sp
1823 (a)(?:(b)c|bd)
1824 .sp
1825 If a vector of 6 elements (allowing for only 1 captured substring) is given
1826 with subject string "abd", \fBpcre_exec()\fP will try to set the second
1827 captured string, thereby recording a vector overflow, before failing to match
1828 "c" and backing up to try the second alternative. The zero return, however,
1829 does correctly indicate that the maximum number of slots (namely 2) have been
1830 filled. In similar cases where there is temporary overflow, but the final
1831 number of used slots is actually less than the maximum, a non-zero value is
1832 returned.
1833 .P
1834 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1835 subpatterns there are in a compiled pattern. The smallest size for
1836 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1837 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1838 .P
1839 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1840 the subject when subpattern \fIn\fP has not been used at all. For example, if
1841 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1842 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1843 happens, both values in the offset pairs corresponding to unused subpatterns
1844 are set to -1.
1845 .P
1846 Offset values that correspond to unused subpatterns at the end of the
1847 expression are also set to -1. For example, if the string "abc" is matched
1848 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1849 return from the function is 2, because the highest used capturing subpattern
1850 number is 1, and the offsets for for the second and third capturing subpatterns
1851 (assuming the vector is large enough, of course) are set to -1.
1852 .P
1853 \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1854 correspond to capturing parentheses in the pattern are never changed. That is,
1855 if a pattern contains \fIn\fP capturing parentheses, no more than
1856 \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1857 elements (in the first two-thirds) retain whatever values they previously had.
1858 .P
1859 Some convenience functions are provided for extracting the captured substrings
1860 as separate strings. These are described below.
1861 .
1862 .
1863 .\" HTML <a name="errorlist"></a>
1864 .SS "Error return values from \fBpcre_exec()\fP"
1865 .rs
1866 .sp
1867 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1868 defined in the header file:
1869 .sp
1871 .sp
1872 The subject string did not match the pattern.
1873 .sp
1875 .sp
1876 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1877 NULL and \fIovecsize\fP was not zero.
1878 .sp
1880 .sp
1881 An unrecognized bit was set in the \fIoptions\fP argument.
1882 .sp
1884 .sp
1885 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1886 the case when it is passed a junk pointer and to detect when a pattern that was
1887 compiled in an environment of one endianness is run in an environment with the
1888 other endianness. This is the error that PCRE gives when the magic number is
1889 not present.
1890 .sp
1892 .sp
1893 While running the pattern match, an unknown item was encountered in the
1894 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1895 of the compiled pattern.
1896 .sp
1898 .sp
1899 If a pattern contains back references, but the \fIovector\fP that is passed to
1900 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1901 gets a block of memory at the start of matching to use for this purpose. If the
1902 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1903 automatically freed at the end of matching.
1904 .P
1905 This error is also given if \fBpcre_stack_malloc()\fP fails in
1906 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1907 \fB--disable-stack-for-recursion\fP.
1908 .sp
1910 .sp
1911 This error is used by the \fBpcre_copy_substring()\fP,
1912 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1913 below). It is never returned by \fBpcre_exec()\fP.
1914 .sp
1916 .sp
1917 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1918 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1919 above.
1920 .sp
1922 .sp
1923 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1924 use by callout functions that want to yield a distinctive error code. See the
1925 .\" HREF
1926 \fBpcrecallout\fP
1927 .\"
1928 documentation for details.
1929 .sp
1931 .sp
1932 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1933 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1934 (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1935 UTF-8 character is placed in the first element, and a reason code is placed in
1936 the second element. The reason codes are listed in the
1937 .\" HTML <a href="#badutf8reasons">
1938 .\" </a>
1939 following section.
1940 .\"
1941 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1942 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1943 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1944 .sp
1946 .sp
1947 The UTF-8 byte sequence that was passed as a subject was checked and found to
1948 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
1949 \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1950 end of the subject.
1951 .sp
1953 .sp
1954 The subject string did not match, but it did match partially. See the
1955 .\" HREF
1956 \fBpcrepartial\fP
1957 .\"
1958 documentation for details of partial matching.
1959 .sp
1961 .sp
1962 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1963 option was used with a compiled pattern containing items that were not
1964 supported for partial matching. From release 8.00 onwards, there are no
1965 restrictions on partial matching.
1966 .sp
1968 .sp
1969 An unexpected internal error has occurred. This error could be caused by a bug
1970 in PCRE or by overwriting of the compiled pattern.
1971 .sp
1973 .sp
1974 This error is given if the value of the \fIovecsize\fP argument is negative.
1975 .sp
1977 .sp
1978 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1979 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1980 description above.
1981 .sp
1983 .sp
1984 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1985 .sp
1987 .sp
1988 The value of \fIstartoffset\fP was negative or greater than the length of the
1989 subject, that is, the value in \fIlength\fP.
1990 .sp
1992 .sp
1993 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
1994 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
1995 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
1996 fact sufficient to detect this case, but this special error code for
1997 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
1998 retained for backwards compatibility.
1999 .sp
2001 .sp
2002 This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2003 the pattern. Specifically, it means that either the whole pattern or a
2004 subpattern has been called recursively for the second time at the same position
2005 in the subject string. Some simple patterns that might do this are detected and
2006 faulted at compile time, but more complicated cases, in particular mutual
2007 recursions between two different subpatterns, cannot be detected until run
2008 time.
2009 .sp
2011 .sp
2012 This error is returned when a pattern that was successfully studied using the
2013 PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2014 the just-in-time processing stack is not large enough. See the
2015 .\" HREF
2016 \fBpcrejit\fP
2017 .\"
2018 documentation for more details.
2019 .P
2020 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
2021 .
2022 .
2023 .\" HTML <a name="badutf8reasons"></a>
2024 .SS "Reason codes for invalid UTF-8 strings"
2025 .rs
2026 .sp
2027 When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2028 PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2029 least 2, the offset of the start of the invalid UTF-8 character is placed in
2030 the first output vector element (\fIovector[0]\fP) and a reason code is placed
2031 in the second element (\fIovector[1]\fP). The reason codes are given names in
2032 the \fBpcre.h\fP header file:
2033 .sp
2039 .sp
2040 The string ends with a truncated UTF-8 character; the code specifies how many
2041 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2042 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2043 allows for up to 6 bytes, and this is checked first; hence the possibility of
2044 4 or 5 missing bytes.
2045 .sp
2050 PCRE_UTF8_ERR10
2051 .sp
2052 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2053 character do not have the binary value 0b10 (that is, either the most
2054 significant bit is 0, or the next bit is 1).
2055 .sp
2056 PCRE_UTF8_ERR11
2057 PCRE_UTF8_ERR12
2058 .sp
2059 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2060 these code points are excluded by RFC 3629.
2061 .sp
2062 PCRE_UTF8_ERR13
2063 .sp
2064 A 4-byte character has a value greater than 0x10fff; these code points are
2065 excluded by RFC 3629.
2066 .sp
2067 PCRE_UTF8_ERR14
2068 .sp
2069 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2070 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2071 from UTF-8.
2072 .sp
2073 PCRE_UTF8_ERR15
2074 PCRE_UTF8_ERR16
2075 PCRE_UTF8_ERR17
2076 PCRE_UTF8_ERR18
2077 PCRE_UTF8_ERR19
2078 .sp
2079 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2080 value that can be represented by fewer bytes, which is invalid. For example,
2081 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2082 one byte.
2083 .sp
2084 PCRE_UTF8_ERR20
2085 .sp
2086 The two most significant bits of the first byte of a character have the binary
2087 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2088 byte can only validly occur as the second or subsequent byte of a multi-byte
2089 character.
2090 .sp
2091 PCRE_UTF8_ERR21
2092 .sp
2093 The first byte of a character has the value 0xfe or 0xff. These values can
2094 never occur in a valid UTF-8 string.
2095 .
2096 .
2098 .rs
2099 .sp
2100 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2101 .ti +5n
2102 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
2103 .ti +5n
2104 .B int \fIbuffersize\fP);
2105 .PP
2106 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2107 .ti +5n
2108 .B int \fIstringcount\fP, int \fIstringnumber\fP,
2109 .ti +5n
2110 .B const char **\fIstringptr\fP);
2111 .PP
2112 .B int pcre_get_substring_list(const char *\fIsubject\fP,
2113 .ti +5n
2114 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
2115 .PP
2116 Captured substrings can be accessed directly by using the offsets returned by
2117 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
2118 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
2119 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2120 as new, separate, zero-terminated strings. These functions identify substrings
2121 by number. The next section describes functions for extracting named
2122 substrings.
2123 .P
2124 A substring that contains a binary zero is correctly extracted and has a
2125 further zero added on the end, but the result is not, of course, a C string.
2126 However, you can process such a string by referring to the length that is
2127 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2128 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2129 for handling strings containing binary zeros, because the end of the final
2130 string is not independently indicated.
2131 .P
2132 The first three arguments are the same for all three of these functions:
2133 \fIsubject\fP is the subject string that has just been successfully matched,
2134 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
2135 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
2136 captured by the match, including the substring that matched the entire regular
2137 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
2138 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
2139 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
2140 number of elements in the vector divided by three.
2141 .P
2142 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
2143 extract a single substring, whose number is given as \fIstringnumber\fP. A
2144 value of zero extracts the substring that matched the entire pattern, whereas
2145 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2146 the string is placed in \fIbuffer\fP, whose length is given by
2147 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2148 obtained via \fBpcre_malloc\fP, and its address is returned via
2149 \fIstringptr\fP. The yield of the function is the length of the string, not
2150 including the terminating zero, or one of these error codes:
2151 .sp
2153 .sp
2154 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2155 memory failed for \fBpcre_get_substring()\fP.
2156 .sp
2158 .sp
2159 There is no substring whose number is \fIstringnumber\fP.
2160 .P
2161 The \fBpcre_get_substring_list()\fP function extracts all available substrings
2162 and builds a list of pointers to them. All this is done in a single block of
2163 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2164 is returned via \fIlistptr\fP, which is also the start of the list of string
2165 pointers. The end of the list is marked by a NULL pointer. The yield of the
2166 function is zero if all went well, or the error code
2167 .sp
2169 .sp
2170 if the attempt to get the memory block failed.
2171 .P
2172 When any of these functions encounter a substring that is unset, which can
2173 happen when capturing subpattern number \fIn+1\fP matches some part of the
2174 subject, but subpattern \fIn\fP has not been used at all, they return an empty
2175 string. This can be distinguished from a genuine zero-length substring by
2176 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2177 substrings.
2178 .P
2179 The two convenience functions \fBpcre_free_substring()\fP and
2180 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2181 a previous call of \fBpcre_get_substring()\fP or
2182 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2183 the function pointed to by \fBpcre_free\fP, which of course could be called
2184 directly from a C program. However, PCRE is used in some situations where it is
2185 linked via a special interface to another programming language that cannot use
2186 \fBpcre_free\fP directly; it is for these cases that the functions are
2187 provided.
2188 .
2189 .
2191 .rs
2192 .sp
2193 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2194 .ti +5n
2195 .B const char *\fIname\fP);
2196 .PP
2197 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2198 .ti +5n
2199 .B const char *\fIsubject\fP, int *\fIovector\fP,
2200 .ti +5n
2201 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2202 .ti +5n
2203 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2204 .PP
2205 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2206 .ti +5n
2207 .B const char *\fIsubject\fP, int *\fIovector\fP,
2208 .ti +5n
2209 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2210 .ti +5n
2211 .B const char **\fIstringptr\fP);
2212 .PP
2213 To extract a substring by name, you first have to find associated number.
2214 For example, for this pattern
2215 .sp
2216 (a+)b(?<xxx>\ed+)...
2217 .sp
2218 the number of the subpattern called "xxx" is 2. If the name is known to be
2219 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2220 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2221 pattern, and the second is the name. The yield of the function is the
2222 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2223 that name.
2224 .P
2225 Given the number, you can extract the substring directly, or use one of the
2226 functions described in the previous section. For convenience, there are also
2227 two functions that do the whole job.
2228 .P
2229 Most of the arguments of \fBpcre_copy_named_substring()\fP and
2230 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2231 functions that extract by number. As these are described in the previous
2232 section, they are not re-described here. There are just two differences:
2233 .P
2234 First, instead of a substring number, a substring name is given. Second, there
2235 is an extra argument, given at the start, which is a pointer to the compiled
2236 pattern. This is needed in order to gain access to the name-to-number
2237 translation table.
2238 .P
2239 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2240 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2241 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2242 the behaviour may not be what you want (see the next section).
2243 .P
2244 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2245 subpatterns with the same number, as described in the
2246 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2247 .\" </a>
2248 section on duplicate subpattern numbers
2249 .\"
2250 in the
2251 .\" HREF
2252 \fBpcrepattern\fP
2253 .\"
2254 page, you cannot use names to distinguish the different subpatterns, because
2255 names are not included in the compiled code. The matching process uses only
2256 numbers. For this reason, the use of different names for subpatterns of the
2257 same number causes an error at compile time.
2258 .
2259 .
2261 .rs
2262 .sp
2263 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2264 .ti +5n
2265 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2266 .PP
2267 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2268 are not required to be unique. (Duplicate names are always allowed for
2269 subpatterns with the same number, created by using the (?| feature. Indeed, if
2270 such subpatterns are named, they are required to use the same names.)
2271 .P
2272 Normally, patterns with duplicate names are such that in any one match, only
2273 one of the named subpatterns participates. An example is shown in the
2274 .\" HREF
2275 \fBpcrepattern\fP
2276 .\"
2277 documentation.
2278 .P
2279 When duplicates are present, \fBpcre_copy_named_substring()\fP and
2280 \fBpcre_get_named_substring()\fP return the first substring corresponding to
2281 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2282 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2283 returns one of the numbers that are associated with the name, but it is not
2284 defined which it is.
2285 .P
2286 If you want to get full details of all captured substrings for a given name,
2287 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2288 argument is the compiled pattern, and the second is the name. The third and
2289 fourth are pointers to variables which are updated by the function. After it
2290 has run, they point to the first and last entries in the name-to-number table
2291 for the given name. The function itself returns the length of each entry, or
2292 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2293 described above in the section entitled \fIInformation about a pattern\fP
2294 .\" HTML <a href="#infoaboutpattern">
2295 .\" </a>
2296 above.
2297 .\"
2298 Given all the relevant entries for the name, you can extract each of their
2299 numbers, and hence the captured data, if any.
2300 .
2301 .
2303 .rs
2304 .sp
2305 The traditional matching function uses a similar algorithm to Perl, which stops
2306 when it finds the first match, starting at a given point in the subject. If you
2307 want to find all possible matches, or the longest possible match, consider
2308 using the alternative matching function (see below) instead. If you cannot use
2309 the alternative function, but still need to find all possible matches, you
2310 can kludge it up by making use of the callout facility, which is described in
2311 the
2312 .\" HREF
2313 \fBpcrecallout\fP
2314 .\"
2315 documentation.
2316 .P
2317 What you have to do is to insert a callout right at the end of the pattern.
2318 When your callout function is called, extract and save the current matched
2319 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2320 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2321 will yield PCRE_ERROR_NOMATCH.
2322 .
2323 .
2324 .\" HTML <a name="dfamatch"></a>
2326 .rs
2327 .sp
2328 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2329 .ti +5n
2330 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2331 .ti +5n
2332 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2333 .ti +5n
2334 .B int *\fIworkspace\fP, int \fIwscount\fP);
2335 .P
2336 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2337 a compiled pattern, using a matching algorithm that scans the subject string
2338 just once, and does not backtrack. This has different characteristics to the
2339 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2340 patterns are not supported. Nevertheless, there are times when this kind of
2341 matching can be useful. For a discussion of the two matching algorithms, and a
2342 list of features that \fBpcre_dfa_exec()\fP does not support, see the
2343 .\" HREF
2344 \fBpcrematching\fP
2345 .\"
2346 documentation.
2347 .P
2348 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2349 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2350 different way, and this is described below. The other common arguments are used
2351 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2352 here.
2353 .P
2354 The two additional arguments provide workspace for the function. The workspace
2355 vector should contain at least 20 elements. It is used for keeping track of
2356 multiple paths through the pattern tree. More workspace will be needed for
2357 patterns and subjects where there are a lot of potential matches.
2358 .P
2359 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2360 .sp
2361 int rc;
2362 int ovector[10];
2363 int wspace[20];
2364 rc = pcre_dfa_exec(
2365 re, /* result of pcre_compile() */
2366 NULL, /* we didn't study the pattern */
2367 "some string", /* the subject string */
2368 11, /* the length of the subject string */
2369 0, /* start at offset 0 in the subject */
2370 0, /* default options */
2371 ovector, /* vector of integers for substring information */
2372 10, /* number of elements (NOT size in bytes) */
2373 wspace, /* working space vector */
2374 20); /* number of elements (NOT size in bytes) */
2375 .
2376 .SS "Option bits for \fBpcre_dfa_exec()\fP"
2377 .rs
2378 .sp
2379 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2380 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2384 All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2385 so their description is not repeated here.
2386 .sp
2389 .sp
2390 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2391 details are slightly different. When PCRE_PARTIAL_HARD is set for
2392 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2393 is reached and there is still at least one matching possibility that requires
2394 additional characters. This happens even if some complete matches have also
2395 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2396 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2397 there have been no complete matches, but there is still at least one matching
2398 possibility. The portion of the string that was inspected when the longest
2399 partial match was found is set as the first matching string in both cases.
2400 There is a more detailed discussion of partial and multi-segment matching, with
2401 examples, in the
2402 .\" HREF
2403 \fBpcrepartial\fP
2404 .\"
2405 documentation.
2406 .sp
2408 .sp
2409 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2410 soon as it has found one match. Because of the way the alternative algorithm
2411 works, this is necessarily the shortest possible match at the first possible
2412 matching point in the subject string.
2413 .sp
2415 .sp
2416 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2417 again, with additional subject characters, and have it continue with the same
2418 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2419 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2420 before because data about the match so far is left in them after a partial
2421 match. There is more discussion of this facility in the
2422 .\" HREF
2423 \fBpcrepartial\fP
2424 .\"
2425 documentation.
2426 .
2427 .
2428 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2429 .rs
2430 .sp
2431 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2432 substring in the subject. Note, however, that all the matches from one run of
2433 the function start at the same point in the subject. The shorter matches are
2434 all initial substrings of the longer matches. For example, if the pattern
2435 .sp
2436 <.*>
2437 .sp
2438 is matched against the string
2439 .sp
2440 This is <something> <something else> <something further> no more
2441 .sp
2442 the three matched strings are
2443 .sp
2444 <something>
2445 <something> <something else>
2446 <something> <something else> <something further>
2447 .sp
2448 On success, the yield of the function is a number greater than zero, which is
2449 the number of matched substrings. The substrings themselves are returned in
2450 \fIovector\fP. Each string uses two elements; the first is the offset to the
2451 start, and the second is the offset to the end. In fact, all the strings have
2452 the same start offset. (Space could have been saved by giving this only once,
2453 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2454 returns data, even though the meaning of the strings is different.)
2455 .P
2456 The strings are returned in reverse order of length; that is, the longest
2457 matching string is given first. If there were too many matches to fit into
2458 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2459 the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2460 the entire \fIovector\fP for returning matched strings.
2461 .
2462 .
2463 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2464 .rs
2465 .sp
2466 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2467 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2468 described
2469 .\" HTML <a href="#errorlist">
2470 .\" </a>
2471 above.
2472 .\"
2473 There are in addition the following errors that are specific to
2474 \fBpcre_dfa_exec()\fP:
2475 .sp
2477 .sp
2478 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2479 that it does not support, for instance, the use of \eC or a back reference.
2480 .sp
2482 .sp
2483 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2484 uses a back reference for the condition, or a test for recursion in a specific
2485 group. These are not supported.
2486 .sp
2488 .sp
2489 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2490 block that contains a setting of the \fImatch_limit\fP or
2491 \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2492 meaningless for DFA matching).
2493 .sp
2495 .sp
2496 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2497 \fIworkspace\fP vector.
2498 .sp
2500 .sp
2501 When a recursive subpattern is processed, the matching function calls itself
2502 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2503 error is given if the output vector is not large enough. This should be
2504 extremely rare, as a vector of size 1000 is used.
2505 .
2506 .
2507 .SH "SEE ALSO"
2508 .rs
2509 .sp
2510 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2511 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2512 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2513 .
2514 .
2516 .rs
2517 .sp
2518 .nf
2519 Philip Hazel
2520 University Computing Service
2521 Cambridge CB2 3QH, England.
2522 .fi
2523 .
2524 .
2526 .rs
2527 .sp
2528 .nf
2529 Last updated: 06 September 2011
2530 Copyright (c) 1997-2011 University of Cambridge.
2531 .fi


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5