ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 683 - (show annotations)
Tue Sep 6 10:37:15 2011 UTC (9 years, 6 months ago) by ph10
File size: 108023 byte(s)
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 .PP
115 .B void *(*pcre_malloc)(size_t);
116 .PP
117 .B void (*pcre_free)(void *);
118 .PP
119 .B void *(*pcre_stack_malloc)(size_t);
120 .PP
121 .B void (*pcre_stack_free)(void *);
122 .PP
123 .B int (*pcre_callout)(pcre_callout_block *);
124 .
125 .
127 .rs
128 .sp
129 PCRE has its own native API, which is described in this document. There are
130 also some wrapper functions that correspond to the POSIX regular expression
131 API, but they do not give access to all the functionality. They are described
132 in the
133 .\" HREF
134 \fBpcreposix\fP
135 .\"
136 documentation. Both of these APIs define a set of C function calls. A C++
137 wrapper is also distributed with PCRE. It is documented in the
138 .\" HREF
139 \fBpcrecpp\fP
140 .\"
141 page.
142 .P
143 The native API C function prototypes are defined in the header file
144 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
145 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
146 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
147 and PCRE_MINOR to contain the major and minor release numbers for the library.
148 Applications can use these to include support for different releases of PCRE.
149 .P
150 In a Windows environment, if you want to statically link an application program
151 against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
152 including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
153 \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
154 \fB__declspec(dllimport)\fP, with unwanted results.
155 .P
156 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
157 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
158 in a Perl-compatible manner. A sample program that demonstrates the simplest
159 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
160 source distribution. A listing of this program is given in the
161 .\" HREF
162 \fBpcredemo\fP
163 .\"
164 documentation, and the
165 .\" HREF
166 \fBpcresample\fP
167 .\"
168 documentation describes how to compile and run it.
169 .P
170 Just-in-time compiler support is an optional feature of PCRE that can be built
171 in appropriate hardware environments. It greatly speeds up the matching
172 performance of many patterns. Simple programs can request its use if available.
173 More complicated programs might need to make use of the
174 \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
175 \fBpcre_assign_jit_stack()\fP functions in order to control its memory usage.
176 These functions are discussed in the
177 .\" HREF
178 \fBpcrejit\fP
179 .\"
180 documentation.
181 .P
182 A second matching function, \fBpcre_dfa_exec()\fP, which is not
183 Perl-compatible, is also provided. This uses a different algorithm for the
184 matching. The alternative algorithm finds all possible matches (at a given
185 point in the subject), and scans the subject just once (unless there are
186 lookbehind assertions). However, this algorithm does not return captured
187 substrings. A description of the two matching algorithms and their advantages
188 and disadvantages is given in the
189 .\" HREF
190 \fBpcrematching\fP
191 .\"
192 documentation.
193 .P
194 In addition to the main compiling and matching functions, there are convenience
195 functions for extracting captured substrings from a subject string that is
196 matched by \fBpcre_exec()\fP. They are:
197 .sp
198 \fBpcre_copy_substring()\fP
199 \fBpcre_copy_named_substring()\fP
200 \fBpcre_get_substring()\fP
201 \fBpcre_get_named_substring()\fP
202 \fBpcre_get_substring_list()\fP
203 \fBpcre_get_stringnumber()\fP
204 \fBpcre_get_stringtable_entries()\fP
205 .sp
206 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
207 provided, to free the memory used for extracted strings.
208 .P
209 The function \fBpcre_maketables()\fP is used to build a set of character tables
210 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
211 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
212 specialist use. Most commonly, no special tables are passed, in which case
213 internal tables that are generated when PCRE is built are used.
214 .P
215 The function \fBpcre_fullinfo()\fP is used to find out information about a
216 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
217 some of the available information, but is retained for backwards compatibility.
218 The function \fBpcre_version()\fP returns a pointer to a string containing the
219 version of PCRE and its date of release.
220 .P
221 The function \fBpcre_refcount()\fP maintains a reference count in a data block
222 containing a compiled pattern. This is provided for the benefit of
223 object-oriented applications.
224 .P
225 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
226 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
227 respectively. PCRE calls the memory management functions via these variables,
228 so a calling program can replace them if it wishes to intercept the calls. This
229 should be done before calling any PCRE functions.
230 .P
231 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
232 indirections to memory management functions. These special functions are used
233 only when PCRE is compiled to use the heap for remembering data, instead of
234 recursive function calls, when running the \fBpcre_exec()\fP function. See the
235 .\" HREF
236 \fBpcrebuild\fP
237 .\"
238 documentation for details of how to do this. It is a non-standard way of
239 building PCRE, for use in environments that have limited stacks. Because of the
240 greater use of memory management, it runs more slowly. Separate functions are
241 provided so that special-purpose external code can be used for this case. When
242 used, these functions are always called in a stack-like manner (last obtained,
243 first freed), and always for memory blocks of the same size. There is a
244 discussion about PCRE's stack usage in the
245 .\" HREF
246 \fBpcrestack\fP
247 .\"
248 documentation.
249 .P
250 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
251 by the caller to a "callout" function, which PCRE will then call at specified
252 points during a matching operation. Details are given in the
253 .\" HREF
254 \fBpcrecallout\fP
255 .\"
256 documentation.
257 .
258 .
259 .\" HTML <a name="newlines"></a>
261 .rs
262 .sp
263 PCRE supports five different conventions for indicating line breaks in
264 strings: a single CR (carriage return) character, a single LF (linefeed)
265 character, the two-character sequence CRLF, any of the three preceding, or any
266 Unicode newline sequence. The Unicode newline sequences are the three just
267 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
268 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
269 (paragraph separator, U+2029).
270 .P
271 Each of the first three conventions is used by at least one operating system as
272 its standard newline sequence. When PCRE is built, a default can be specified.
273 The default default is LF, which is the Unix standard. When PCRE is run, the
274 default can be overridden, either when a pattern is compiled, or when it is
275 matched.
276 .P
277 At compile time, the newline convention can be specified by the \fIoptions\fP
278 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
279 start of the pattern itself; this overrides any other settings. See the
280 .\" HREF
281 \fBpcrepattern\fP
282 .\"
283 page for details of the special character sequences.
284 .P
285 In the PCRE documentation the word "newline" is used to mean "the character or
286 pair of characters that indicate a line break". The choice of newline
287 convention affects the handling of the dot, circumflex, and dollar
288 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
289 recognized line ending sequence, the match position advancement for a
290 non-anchored pattern. There is more detail about this in the
291 .\" HTML <a href="#execoptions">
292 .\" </a>
293 section on \fBpcre_exec()\fP options
294 .\"
295 below.
296 .P
297 The choice of newline convention does not affect the interpretation of
298 the \en or \er escape sequences, nor does it affect what \eR matches, which is
299 controlled in a similar way, but by separate options.
300 .
301 .
303 .rs
304 .sp
305 The PCRE functions can be used in multi-threading applications, with the
306 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
307 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
308 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
309 .P
310 The compiled form of a regular expression is not altered during matching, so
311 the same compiled pattern can safely be used by several threads at once.
312 .P
313 If the just-in-time optimization feature is being used, it needs separate
314 memory stack areas for each thread. See the
315 .\" HREF
316 \fBpcrejit\fP
317 .\"
318 documentation for more details.
319 .
320 .
322 .rs
323 .sp
324 The compiled form of a regular expression can be saved and re-used at a later
325 time, possibly by a different program, and even on a host other than the one on
326 which it was compiled. Details are given in the
327 .\" HREF
328 \fBpcreprecompile\fP
329 .\"
330 documentation. However, compiling a regular expression with one version of PCRE
331 for use with a different version is not guaranteed to work and may cause
332 crashes.
333 .
334 .
336 .rs
337 .sp
338 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
339 .PP
340 The function \fBpcre_config()\fP makes it possible for a PCRE client to
341 discover which optional features have been compiled into the PCRE library. The
342 .\" HREF
343 \fBpcrebuild\fP
344 .\"
345 documentation has more details about these optional features.
346 .P
347 The first argument for \fBpcre_config()\fP is an integer, specifying which
348 information is required; the second argument is a pointer to a variable into
349 which the information is placed. The following information is available:
350 .sp
352 .sp
353 The output is an integer that is set to one if UTF-8 support is available;
354 otherwise it is set to zero.
355 .sp
357 .sp
358 The output is an integer that is set to one if support for Unicode character
359 properties is available; otherwise it is set to zero.
360 .sp
362 .sp
363 The output is an integer that is set to one if support for just-in-time
364 compiling is available; otherwise it is set to zero.
365 .sp
367 .sp
368 The output is an integer whose value specifies the default character sequence
369 that is recognized as meaning "newline". The four values that are supported
370 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
371 Though they are derived from ASCII, the same values are returned in EBCDIC
372 environments. The default should normally correspond to the standard sequence
373 for your operating system.
374 .sp
376 .sp
377 The output is an integer whose value indicates what character sequences the \eR
378 escape sequence matches by default. A value of 0 means that \eR matches any
379 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
380 or CRLF. The default can be overridden when a pattern is compiled or matched.
381 .sp
383 .sp
384 The output is an integer that contains the number of bytes used for internal
385 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
386 allow larger regular expressions to be compiled, at the expense of slower
387 matching. The default value of 2 is sufficient for all but the most massive
388 patterns, since it allows the compiled pattern to be up to 64K in size.
389 .sp
391 .sp
392 The output is an integer that contains the threshold above which the POSIX
393 interface uses \fBmalloc()\fP for output vectors. Further details are given in
394 the
395 .\" HREF
396 \fBpcreposix\fP
397 .\"
398 documentation.
399 .sp
401 .sp
402 The output is a long integer that gives the default limit for the number of
403 internal matching function calls in a \fBpcre_exec()\fP execution. Further
404 details are given with \fBpcre_exec()\fP below.
405 .sp
407 .sp
408 The output is a long integer that gives the default limit for the depth of
409 recursion when calling the internal matching function in a \fBpcre_exec()\fP
410 execution. Further details are given with \fBpcre_exec()\fP below.
411 .sp
413 .sp
414 The output is an integer that is set to one if internal recursion when running
415 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
416 to remember their state. This is the usual way that PCRE is compiled. The
417 output is zero if PCRE was compiled to use blocks of data on the heap instead
418 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
419 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
420 avoiding the use of the stack.
421 .
422 .
424 .rs
425 .sp
426 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
427 .ti +5n
428 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
429 .ti +5n
430 .B const unsigned char *\fItableptr\fP);
431 .sp
432 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
433 .ti +5n
434 .B int *\fIerrorcodeptr\fP,
435 .ti +5n
436 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
437 .ti +5n
438 .B const unsigned char *\fItableptr\fP);
439 .P
440 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
441 called to compile a pattern into an internal form. The only difference between
442 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
443 \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
444 too much repetition, we refer just to \fBpcre_compile()\fP below, but the
445 information applies equally to \fBpcre_compile2()\fP.
446 .P
447 The pattern is a C string terminated by a binary zero, and is passed in the
448 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
449 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
450 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
451 for a structure whose contents are not externally defined. It is up to the
452 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
453 .P
454 Although the compiled code of a PCRE regex is relocatable, that is, it does not
455 depend on memory location, the complete \fBpcre\fP data block is not
456 fully relocatable, because it may contain a copy of the \fItableptr\fP
457 argument, which is an address (see below).
458 .P
459 The \fIoptions\fP argument contains various bit settings that affect the
460 compilation. It should be zero if no options are required. The available
461 options are described below. Some of them (in particular, those that are
462 compatible with Perl, but some others as well) can also be set and unset from
463 within the pattern (see the detailed description in the
464 .\" HREF
465 \fBpcrepattern\fP
466 .\"
467 documentation). For those options that can be different in different parts of
468 the pattern, the contents of the \fIoptions\fP argument specifies their
469 settings at the start of compilation and execution. The PCRE_ANCHORED,
471 PCRE_NO_START_OPT options can be set at the time of matching as well as at
472 compile time.
473 .P
474 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
475 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
476 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
477 error message. This is a static string that is part of the library. You must
478 not try to free it. Normally, the offset from the start of the pattern to the
479 byte that was being processed when the error was discovered is placed in the
480 variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
481 immediate error is given). However, for an invalid UTF-8 string, the offset is
482 that of the first byte of the failing character. Also, some errors are not
483 detected until checks are carried out when the whole pattern has been scanned;
484 in these cases the offset passed back is the length of the pattern.
485 .P
486 Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
487 sometimes point into the middle of a UTF-8 character.
488 .P
489 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
490 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
491 returned via this argument in the event of an error. This is in addition to the
492 textual error message. Error codes and messages are listed below.
493 .P
494 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
495 character tables that are built when PCRE is compiled, using the default C
496 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
497 call to \fBpcre_maketables()\fP. This value is stored with the compiled
498 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
499 passed to it. For more discussion, see the section on locale support below.
500 .P
501 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
502 .sp
503 pcre *re;
504 const char *error;
505 int erroffset;
506 re = pcre_compile(
507 "^A.*Z", /* the pattern */
508 0, /* default options */
509 &error, /* for error message */
510 &erroffset, /* for error offset */
511 NULL); /* use default character tables */
512 .sp
513 The following names for option bits are defined in the \fBpcre.h\fP header
514 file:
515 .sp
517 .sp
518 If this bit is set, the pattern is forced to be "anchored", that is, it is
519 constrained to match only at the first matching point in the string that is
520 being searched (the "subject string"). This effect can also be achieved by
521 appropriate constructs in the pattern itself, which is the only way to do it in
522 Perl.
523 .sp
525 .sp
526 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
527 all with number 255, before each pattern item. For discussion of the callout
528 facility, see the
529 .\" HREF
530 \fBpcrecallout\fP
531 .\"
532 documentation.
533 .sp
536 .sp
537 These options (which are mutually exclusive) control what the \eR escape
538 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
539 match any Unicode newline sequence. The default is specified when PCRE is
540 built. It can be overridden from within the pattern, or by setting an option
541 when a compiled pattern is matched.
542 .sp
544 .sp
545 If this bit is set, letters in the pattern match both upper and lower case
546 letters. It is equivalent to Perl's /i option, and it can be changed within a
547 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
548 concept of case for characters whose values are less than 128, so caseless
549 matching is always possible. For characters with higher values, the concept of
550 case is supported if PCRE is compiled with Unicode property support, but not
551 otherwise. If you want to use caseless matching for characters 128 and above,
552 you must ensure that PCRE is compiled with Unicode property support as well as
553 with UTF-8 support.
554 .sp
556 .sp
557 If this bit is set, a dollar metacharacter in the pattern matches only at the
558 end of the subject string. Without this option, a dollar also matches
559 immediately before a newline at the end of the string (but not before any other
560 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
561 There is no equivalent to this option in Perl, and no way to set it within a
562 pattern.
563 .sp
565 .sp
566 If this bit is set, a dot metacharacter in the pattern matches a character of
567 any value, including one that indicates a newline. However, it only ever
568 matches one character, even if newlines are coded as CRLF. Without this option,
569 a dot does not match when the current position is at a newline. This option is
570 equivalent to Perl's /s option, and it can be changed within a pattern by a
571 (?s) option setting. A negative class such as [^a] always matches newline
572 characters, independent of the setting of this option.
573 .sp
575 .sp
576 If this bit is set, names used to identify capturing subpatterns need not be
577 unique. This can be helpful for certain types of pattern when it is known that
578 only one instance of the named subpattern can ever be matched. There are more
579 details of named subpatterns below; see also the
580 .\" HREF
581 \fBpcrepattern\fP
582 .\"
583 documentation.
584 .sp
586 .sp
587 If this bit is set, whitespace data characters in the pattern are totally
588 ignored except when escaped or inside a character class. Whitespace does not
589 include the VT character (code 11). In addition, characters between an
590 unescaped # outside a character class and the next newline, inclusive, are also
591 ignored. This is equivalent to Perl's /x option, and it can be changed within a
592 pattern by a (?x) option setting.
593 .P
594 Which characters are interpreted as newlines is controlled by the options
595 passed to \fBpcre_compile()\fP or by a special sequence at the start of the
596 pattern, as described in the section entitled
597 .\" HTML <a href="pcrepattern.html#newlines">
598 .\" </a>
599 "Newline conventions"
600 .\"
601 in the \fBpcrepattern\fP documentation. Note that the end of this type of
602 comment is a literal newline sequence in the pattern; escape sequences that
603 happen to represent a newline do not count.
604 .P
605 This option makes it possible to include comments inside complicated patterns.
606 Note, however, that this applies only to data characters. Whitespace characters
607 may never appear within special character sequences in a pattern, for example
608 within the sequence (?( that introduces a conditional subpattern.
609 .sp
611 .sp
612 This option was invented in order to turn on additional functionality of PCRE
613 that is incompatible with Perl, but it is currently of very little use. When
614 set, any backslash in a pattern that is followed by a letter that has no
615 special meaning causes an error, thus reserving these combinations for future
616 expansion. By default, as in Perl, a backslash followed by a letter with no
617 special meaning is treated as a literal. (Perl can, however, be persuaded to
618 give an error for this, by running it with the -w option.) There are at present
619 no other features controlled by this option. It can also be set by a (?X)
620 option setting within a pattern.
621 .sp
623 .sp
624 If this option is set, an unanchored pattern is required to match before or at
625 the first newline in the subject string, though the matched text may continue
626 over the newline.
627 .sp
629 .sp
630 If this option is set, PCRE's behaviour is changed in some ways so that it is
631 compatible with JavaScript rather than Perl. The changes are as follows:
632 .P
633 (1) A lone closing square bracket in a pattern causes a compile-time error,
634 because this is illegal in JavaScript (by default it is treated as a data
635 character). Thus, the pattern AB]CD becomes illegal when this option is set.
636 .P
637 (2) At run time, a back reference to an unset subpattern group matches an empty
638 string (by default this causes the current matching alternative to fail). A
639 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
640 an "a" in the subject), whereas it fails by default, for Perl compatibility.
641 .sp
643 .sp
644 By default, PCRE treats the subject string as consisting of a single line of
645 characters (even if it actually contains newlines). The "start of line"
646 metacharacter (^) matches only at the start of the string, while the "end of
647 line" metacharacter ($) matches only at the end of the string, or before a
648 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
649 Perl.
650 .P
651 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
652 match immediately following or immediately before internal newlines in the
653 subject string, respectively, as well as at the very start and end. This is
654 equivalent to Perl's /m option, and it can be changed within a pattern by a
655 (?m) option setting. If there are no newlines in a subject string, or no
656 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
657 .sp
663 .sp
664 These options override the default newline definition that was chosen when PCRE
665 was built. Setting the first or the second specifies that a newline is
666 indicated by a single character (CR or LF, respectively). Setting
667 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
668 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
669 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
670 that any Unicode newline sequence should be recognized. The Unicode newline
671 sequences are the three just mentioned, plus the single characters VT (vertical
672 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
673 separator, U+2028), and PS (paragraph separator, U+2029). The last two are
674 recognized only in UTF-8 mode.
675 .P
676 The newline setting in the options word uses three bits that are treated
677 as a number, giving eight possibilities. Currently only six are used (default
678 plus the five values above). This means that if you set more than one newline
679 option, the combination may or may not be sensible. For example,
681 other combinations may yield unused numbers and cause an error.
682 .P
683 The only time that a line break in a pattern is specially recognized when
684 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
685 and so are ignored in this mode. Also, an unescaped # outside a character class
686 indicates a comment that lasts until after the next line break sequence. In
687 other circumstances, line break sequences in patterns are treated as literal
688 data.
689 .P
690 The newline option that is set at compile time becomes the default that is used
691 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
692 .sp
694 .sp
695 If this option is set, it disables the use of numbered capturing parentheses in
696 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
697 were followed by ?: but named parentheses can still be used for capturing (and
698 they acquire numbers in the usual way). There is no equivalent of this option
699 in Perl.
700 .sp
702 .sp
703 This is an option that acts at matching time; that is, it is really an option
704 for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
705 it is remembered with the compiled pattern and assumed at matching time. For
706 details see the discussion of PCRE_NO_START_OPTIMIZE
707 .\" HTML <a href="#execoptions">
708 .\" </a>
709 below.
710 .\"
711 .sp
713 .sp
714 This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
715 \ew, and some of the POSIX character classes. By default, only ASCII characters
716 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
717 classify characters. More details are given in the section on
718 .\" HTML <a href="pcre.html#genericchartypes">
719 .\" </a>
720 generic character types
721 .\"
722 in the
723 .\" HREF
724 \fBpcrepattern\fP
725 .\"
726 page. If you set PCRE_UCP, matching one of the items it affects takes much
727 longer. The option is available only if PCRE has been compiled with Unicode
728 property support.
729 .sp
731 .sp
732 This option inverts the "greediness" of the quantifiers so that they are not
733 greedy by default, but become greedy if followed by "?". It is not compatible
734 with Perl. It can also be set by a (?U) option setting within the pattern.
735 .sp
737 .sp
738 This option causes PCRE to regard both the pattern and the subject as strings
739 of UTF-8 characters instead of single-byte character strings. However, it is
740 available only when PCRE is built to include UTF-8 support. If not, the use
741 of this option provokes an error. Details of how this option changes the
742 behaviour of PCRE are given in the
743 .\" HREF
744 \fBpcreunicode\fP
745 .\"
746 page.
747 .sp
749 .sp
750 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
751 automatically checked. There is a discussion about the
752 .\" HTML <a href="pcre.html#utf8strings">
753 .\" </a>
754 validity of UTF-8 strings
755 .\"
756 in the main
757 .\" HREF
758 \fBpcre\fP
759 .\"
760 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
761 returns an error. If you already know that your pattern is valid, and you want
762 to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
763 option. When it is set, the effect of passing an invalid UTF-8 string as a
764 pattern is undefined. It may cause your program to crash. Note that this option
765 can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
766 the UTF-8 validity checking of subject strings.
767 .
768 .
770 .rs
771 .sp
772 The following table lists the error codes than may be returned by
773 \fBpcre_compile2()\fP, along with the error messages that may be returned by
774 both compiling functions. As PCRE has developed, some error codes have fallen
775 out of use. To avoid confusion, they have not been re-used.
776 .sp
777 0 no error
778 1 \e at end of pattern
779 2 \ec at end of pattern
780 3 unrecognized character follows \e
781 4 numbers out of order in {} quantifier
782 5 number too big in {} quantifier
783 6 missing terminating ] for character class
784 7 invalid escape sequence in character class
785 8 range out of order in character class
786 9 nothing to repeat
787 10 [this code is not in use]
788 11 internal error: unexpected repeat
789 12 unrecognized character after (? or (?-
790 13 POSIX named classes are supported only within a class
791 14 missing )
792 15 reference to non-existent subpattern
793 16 erroffset passed as NULL
794 17 unknown option bit(s) set
795 18 missing ) after comment
796 19 [this code is not in use]
797 20 regular expression is too large
798 21 failed to get memory
799 22 unmatched parentheses
800 23 internal error: code overflow
801 24 unrecognized character after (?<
802 25 lookbehind assertion is not fixed length
803 26 malformed number or name after (?(
804 27 conditional group contains more than two branches
805 28 assertion expected after (?(
806 29 (?R or (?[+-]digits must be followed by )
807 30 unknown POSIX class name
808 31 POSIX collating elements are not supported
809 32 this version of PCRE is not compiled with PCRE_UTF8 support
810 33 [this code is not in use]
811 34 character value in \ex{...} sequence is too large
812 35 invalid condition (?(0)
813 36 \eC not allowed in lookbehind assertion
814 37 PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
815 38 number after (?C is > 255
816 39 closing ) for (?C expected
817 40 recursive call could loop indefinitely
818 41 unrecognized character after (?P
819 42 syntax error in subpattern name (missing terminator)
820 43 two named subpatterns have the same name
821 44 invalid UTF-8 string
822 45 support for \eP, \ep, and \eX has not been compiled
823 46 malformed \eP or \ep sequence
824 47 unknown property name after \eP or \ep
825 48 subpattern name is too long (maximum 32 characters)
826 49 too many named subpatterns (maximum 10000)
827 50 [this code is not in use]
828 51 octal value is greater than \e377 (not in UTF-8 mode)
829 52 internal error: overran compiling workspace
830 53 internal error: previously-checked referenced subpattern
831 not found
832 54 DEFINE group contains more than one branch
833 55 repeating a DEFINE group is not allowed
834 56 inconsistent NEWLINE options
835 57 \eg is not followed by a braced, angle-bracketed, or quoted
836 name/number or by a plain number
837 58 a numbered reference must not be zero
838 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
839 60 (*VERB) not recognized
840 61 number is too big
841 62 subpattern name expected
842 63 digit expected after (?+
843 64 ] is an invalid data character in JavaScript compatibility mode
844 65 different names for subpatterns of the same number are
845 not allowed
846 66 (*MARK) must have an argument
847 67 this version of PCRE is not compiled with PCRE_UCP support
848 68 \ec must be followed by an ASCII character
849 69 \ek is not followed by a braced, angle-bracketed, or quoted name
850 .sp
851 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
852 be used if the limits were changed when PCRE was built.
853 .
854 .
855 .\" HTML <a name="studyingapattern"></a>
857 .rs
858 .sp
859 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
860 .ti +5n
861 .B const char **\fIerrptr\fP);
862 .PP
863 If a compiled pattern is going to be used several times, it is worth spending
864 more time analyzing it in order to speed up the time taken for matching. The
865 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
866 argument. If studying the pattern produces additional information that will
867 help speed up matching, \fBpcre_study()\fP returns a pointer to a
868 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
869 results of the study.
870 .P
871 The returned value from \fBpcre_study()\fP can be passed directly to
872 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
873 also contains other fields that can be set by the caller before the block is
874 passed; these are described
875 .\" HTML <a href="#extradata">
876 .\" </a>
877 below
878 .\"
879 in the section on matching a pattern.
880 .P
881 If studying the pattern does not produce any useful information,
882 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
883 wants to pass any of the other fields to \fBpcre_exec()\fP or
884 \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
885 .P
886 The second argument of \fBpcre_study()\fP contains option bits. There is only
887 one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
888 compiler is available, the pattern is further compiled into machine code that
889 executes much faster than the \fBpcre_exec()\fP matching function. If
890 the just-in-time compiler is not available, this option is ignored. All other
891 bits in the \fIoptions\fP argument must be zero.
892 .P
893 JIT compilation is a heavyweight optimization. It can take some time for
894 patterns to be analyzed, and for one-off matches and simple patterns the
895 benefit of faster execution might be offset by a much slower study time.
896 Not all patterns can be optimized by the JIT compiler. For those that cannot be
897 handled, matching automatically falls back to the \fBpcre_exec()\fP
898 interpreter. For more details, see the
899 .\" HREF
900 \fBpcrejit\fP
901 .\"
902 documentation.
903 .P
904 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
905 studying succeeds (even if no data is returned), the variable it points to is
906 set to NULL. Otherwise it is set to point to a textual error message. This is a
907 static string that is part of the library. You must not try to free it. You
908 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
909 sure that it has run successfully.
910 .P
911 When you are finished with a pattern, you can free the memory used for the
912 study data by calling \fBpcre_free_study()\fP. This function was added to the
913 API for release 8.20. For earlier versions, the memory could be freed with
914 \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
915 where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
916 new function when convenient.
917 .P
918 This is a typical way in which \fBpcre_study\fP() is used (except that in a
919 real application there should be tests for errors):
920 .sp
921 int rc;
922 pcre *re;
923 pcre_extra *sd;
924 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
925 sd = pcre_study(
926 re, /* result of pcre_compile() */
927 0, /* no options */
928 &error); /* set to NULL or points to a message */
929 rc = pcre_exec( /* see below for details of pcre_exec() options */
930 re, sd, "subject", 7, 0, 0, ovector, 30);
931 ...
932 pcre_free_study(sd);
933 pcre_free(re);
934 .sp
935 Studying a pattern does two things: first, a lower bound for the length of
936 subject string that is needed to match the pattern is computed. This does not
937 mean that there are any strings of that length that match, but it does
938 guarantee that no shorter strings match. The value is used by
939 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
940 match strings that are shorter than the lower bound. You can find out the value
941 in a calling program via the \fBpcre_fullinfo()\fP function.
942 .P
943 Studying a pattern is also useful for non-anchored patterns that do not have a
944 single fixed starting character. A bitmap of possible starting bytes is
945 created. This speeds up finding a position in the subject at which to start
946 matching.
947 .P
948 These two optimizations apply to both \fBpcre_exec()\fP and
949 \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
950 \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
951 just-in-time compiling is successful. The optimizations can be disabled by
952 setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
953 \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
954 callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
955 to make use of these facilities in cases where matching fails. See the
956 discussion of PCRE_NO_START_OPTIMIZE
957 .\" HTML <a href="#execoptions">
958 .\" </a>
959 below.
960 .\"
961 .
962 .
963 .\" HTML <a name="localesupport"></a>
965 .rs
966 .sp
967 PCRE handles caseless matching, and determines whether characters are letters,
968 digits, or whatever, by reference to a set of tables, indexed by character
969 value. When running in UTF-8 mode, this applies only to characters with codes
970 less than 128. By default, higher-valued codes never match escapes such as \ew
971 or \ed, but they can be tested with \ep if PCRE is built with Unicode character
972 property support. Alternatively, the PCRE_UCP option can be set at compile
973 time; this causes \ew and friends to use Unicode property support instead of
974 built-in tables. The use of locales with Unicode is discouraged. If you are
975 handling characters with codes greater than 128, you should either use UTF-8
976 and Unicode, or use locales, but not try to mix the two.
977 .P
978 PCRE contains an internal set of tables that are used when the final argument
979 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
980 Normally, the internal tables recognize only ASCII characters. However, when
981 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
982 default "C" locale of the local system, which may cause them to be different.
983 .P
984 The internal tables can always be overridden by tables supplied by the
985 application that calls PCRE. These may be created in a different locale from
986 the default. As more and more applications change to using Unicode, the need
987 for this locale support is expected to die away.
988 .P
989 External tables are built by calling the \fBpcre_maketables()\fP function,
990 which has no arguments, in the relevant locale. The result can then be passed
991 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
992 example, to build and use tables that are appropriate for the French locale
993 (where accented characters with values greater than 128 are treated as letters),
994 the following code could be used:
995 .sp
996 setlocale(LC_CTYPE, "fr_FR");
997 tables = pcre_maketables();
998 re = pcre_compile(..., tables);
999 .sp
1000 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1001 are using Windows, the name for the French locale is "french".
1002 .P
1003 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1004 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1005 that the memory containing the tables remains available for as long as it is
1006 needed.
1007 .P
1008 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
1009 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
1010 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
1011 pattern, compilation, studying and matching all happen in the same locale, but
1012 different patterns can be compiled in different locales.
1013 .P
1014 It is possible to pass a table pointer or NULL (indicating the use of the
1015 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
1016 this facility could be used to match a pattern in a different locale from the
1017 one in which it was compiled. Passing table pointers at run time is discussed
1018 below in the section on matching a pattern.
1019 .
1020 .
1021 .\" HTML <a name="infoaboutpattern"></a>
1023 .rs
1024 .sp
1025 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1026 .ti +5n
1027 .B int \fIwhat\fP, void *\fIwhere\fP);
1028 .PP
1029 The \fBpcre_fullinfo()\fP function returns information about a compiled
1030 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
1031 nevertheless retained for backwards compability (and is documented below).
1032 .P
1033 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1034 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
1035 the pattern was not studied. The third argument specifies which piece of
1036 information is required, and the fourth argument is a pointer to a variable
1037 to receive the data. The yield of the function is zero for success, or one of
1038 the following negative numbers:
1039 .sp
1040 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1041 the argument \fIwhere\fP was NULL
1042 PCRE_ERROR_BADMAGIC the "magic number" was not found
1043 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
1044 .sp
1045 The "magic number" is placed at the start of each compiled pattern as an simple
1046 check against passing an arbitrary memory pointer. Here is a typical call of
1047 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
1048 .sp
1049 int rc;
1050 size_t length;
1051 rc = pcre_fullinfo(
1052 re, /* result of pcre_compile() */
1053 sd, /* result of pcre_study(), or NULL */
1054 PCRE_INFO_SIZE, /* what is required */
1055 &length); /* where to put the data */
1056 .sp
1057 The possible values for the third argument are defined in \fBpcre.h\fP, and are
1058 as follows:
1059 .sp
1061 .sp
1062 Return the number of the highest back reference in the pattern. The fourth
1063 argument should point to an \fBint\fP variable. Zero is returned if there are
1064 no back references.
1065 .sp
1067 .sp
1068 Return the number of capturing subpatterns in the pattern. The fourth argument
1069 should point to an \fBint\fP variable.
1070 .sp
1072 .sp
1073 Return a pointer to the internal default character tables within PCRE. The
1074 fourth argument should point to an \fBunsigned char *\fP variable. This
1075 information call is provided for internal use by the \fBpcre_study()\fP
1076 function. External callers can cause PCRE to use its internal tables by passing
1077 a NULL table pointer.
1078 .sp
1080 .sp
1081 Return information about the first byte of any matched string, for a
1082 non-anchored pattern. The fourth argument should point to an \fBint\fP
1083 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1084 still recognized for backwards compatibility.)
1085 .P
1086 If there is a fixed first byte, for example, from a pattern such as
1087 (cat|cow|coyote), its value is returned. Otherwise, if either
1088 .sp
1089 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1090 starts with "^", or
1091 .sp
1092 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1093 (if it were set, the pattern would be anchored),
1094 .sp
1095 -1 is returned, indicating that the pattern matches only at the start of a
1096 subject string or after any newline within the string. Otherwise -2 is
1097 returned. For anchored patterns, -2 is returned.
1098 .sp
1100 .sp
1101 If the pattern was studied, and this resulted in the construction of a 256-bit
1102 table indicating a fixed set of bytes for the first byte in any matching
1103 string, a pointer to the table is returned. Otherwise NULL is returned. The
1104 fourth argument should point to an \fBunsigned char *\fP variable.
1105 .sp
1107 .sp
1108 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1109 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1110 explicit match is either a literal CR or LF character, or \er or \en.
1111 .sp
1113 .sp
1114 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1115 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1116 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1117 .sp
1119 .sp
1120 Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1121 just-in-time compiling was successful. The fourth argument should point to an
1122 \fBint\fP variable. A return value of 0 means that JIT support is not available
1123 in this version of PCRE, or that the pattern was not studied with the
1124 PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1125 particular pattern. See the
1126 .\" HREF
1127 \fBpcrejit\fP
1128 .\"
1129 documentation for details of what can and cannot be handled.
1130 .sp
1132 .sp
1133 Return the value of the rightmost literal byte that must exist in any matched
1134 string, other than at its start, if such a byte has been recorded. The fourth
1135 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1136 returned. For anchored patterns, a last literal byte is recorded only if it
1137 follows something of variable length. For example, for the pattern
1138 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1139 is -1.
1140 .sp
1142 .sp
1143 If the pattern was studied and a minimum length for matching subject strings
1144 was computed, its value is returned. Otherwise the returned value is -1. The
1145 value is a number of characters, not bytes (this may be relevant in UTF-8
1146 mode). The fourth argument should point to an \fBint\fP variable. A
1147 non-negative value is a lower bound to the length of any matching string. There
1148 may not be any strings of that length that do actually match, but every string
1149 that does match is at least that long.
1150 .sp
1154 .sp
1155 PCRE supports the use of named as well as numbered capturing parentheses. The
1156 names are just an additional way of identifying the parentheses, which still
1157 acquire numbers. Several convenience functions such as
1158 \fBpcre_get_named_substring()\fP are provided for extracting captured
1159 substrings by name. It is also possible to extract the data directly, by first
1160 converting the name to a number in order to access the correct pointers in the
1161 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1162 you need to use the name-to-number map, which is described by these three
1163 values.
1164 .P
1165 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1166 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1167 entry; both of these return an \fBint\fP value. The entry size depends on the
1168 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1169 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1170 are the number of the capturing parenthesis, most significant byte first. The
1171 rest of the entry is the corresponding name, zero terminated.
1172 .P
1173 The names are in alphabetical order. Duplicate names may appear if (?| is used
1174 to create multiple groups with the same number, as described in the
1175 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1176 .\" </a>
1177 section on duplicate subpattern numbers
1178 .\"
1179 in the
1180 .\" HREF
1181 \fBpcrepattern\fP
1182 .\"
1183 page. Duplicate names for subpatterns with different numbers are permitted only
1184 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1185 table in the order in which they were found in the pattern. In the absence of
1186 (?| this is the order of increasing number; when (?| is used this is not
1187 necessarily the case because later subpatterns may have lower numbers.
1188 .P
1189 As a simple example of the name/number table, consider the following pattern
1190 (assume PCRE_EXTENDED is set, so white space - including newlines - is
1191 ignored):
1192 .sp
1193 .\" JOIN
1194 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1195 (?<month>\ed\ed) - (?<day>\ed\ed) )
1196 .sp
1197 There are four named subpatterns, so the table has four entries, and each entry
1198 in the table is eight bytes long. The table is as follows, with non-printing
1199 bytes shows in hexadecimal, and undefined bytes shown as ??:
1200 .sp
1201 00 01 d a t e 00 ??
1202 00 05 d a y 00 ?? ??
1203 00 04 m o n t h 00
1204 00 02 y e a r 00 ??
1205 .sp
1206 When writing code to extract data from named subpatterns using the
1207 name-to-number map, remember that the length of the entries is likely to be
1208 different for each compiled pattern.
1209 .sp
1211 .sp
1212 Return 1 if the pattern can be used for partial matching with
1213 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1214 \fBint\fP variable. From release 8.00, this always returns 1, because the
1215 restrictions that previously applied to partial matching have been lifted. The
1216 .\" HREF
1217 \fBpcrepartial\fP
1218 .\"
1219 documentation gives details of partial matching.
1220 .sp
1222 .sp
1223 Return a copy of the options with which the pattern was compiled. The fourth
1224 argument should point to an \fBunsigned long int\fP variable. These option bits
1225 are those specified in the call to \fBpcre_compile()\fP, modified by any
1226 top-level option settings at the start of the pattern itself. In other words,
1227 they are the options that will be in force when matching starts. For example,
1228 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1230 .P
1231 A pattern is automatically anchored by PCRE if all of its top-level
1232 alternatives begin with one of the following:
1233 .sp
1234 ^ unless PCRE_MULTILINE is set
1235 \eA always
1236 \eG always
1237 .\" JOIN
1238 .* if PCRE_DOTALL is set and there are no back
1239 references to the subpattern in which .* appears
1240 .sp
1241 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1242 \fBpcre_fullinfo()\fP.
1243 .sp
1245 .sp
1246 Return the size of the compiled pattern, that is, the value that was passed as
1247 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1248 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1249 variable.
1250 .sp
1252 .sp
1253 Return the size of the data block pointed to by the \fIstudy_data\fP field in a
1254 \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
1255 zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
1256 The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
1257 that will speed up matching (see the section entitled
1258 .\" HTML <a href="#studyingapattern">
1259 .\" </a>
1260 "Studying a pattern"
1261 .\"
1262 above). The format of the \fIstudy_data\fP block is private, but its length
1263 is made available via this option so that it can be saved and restored (see the
1264 .\" HREF
1265 \fBpcreprecompile\fP
1266 .\"
1267 documentation for details).
1268 .
1269 .
1271 .rs
1272 .sp
1273 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1274 .B *\fIfirstcharptr\fP);
1275 .PP
1276 The \fBpcre_info()\fP function is now obsolete because its interface is too
1277 restrictive to return all the available data about a compiled pattern. New
1278 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1279 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1280 following negative numbers:
1281 .sp
1282 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1283 PCRE_ERROR_BADMAGIC the "magic number" was not found
1284 .sp
1285 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1286 pattern was compiled is placed in the integer it points to (see
1287 PCRE_INFO_OPTIONS above).
1288 .P
1289 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1290 it is used to pass back information about the first character of any matched
1291 string (see PCRE_INFO_FIRSTBYTE above).
1292 .
1293 .
1295 .rs
1296 .sp
1297 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1298 .PP
1299 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1300 data block that contains a compiled pattern. It is provided for the benefit of
1301 applications that operate in an object-oriented manner, where different parts
1302 of the application may be using the same compiled pattern, but you want to free
1303 the block when they are all done.
1304 .P
1305 When a pattern is compiled, the reference count field is initialized to zero.
1306 It is changed only by calling this function, whose action is to add the
1307 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1308 function is the new value. However, the value of the count is constrained to
1309 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1310 it is forced to the appropriate limit value.
1311 .P
1312 Except when it is zero, the reference count is not correctly preserved if a
1313 pattern is compiled on one host and then transferred to a host whose byte-order
1314 is different. (This seems a highly unlikely scenario.)
1315 .
1316 .
1318 .rs
1319 .sp
1320 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1321 .ti +5n
1322 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1323 .ti +5n
1324 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1325 .P
1326 The function \fBpcre_exec()\fP is called to match a subject string against a
1327 compiled pattern, which is passed in the \fIcode\fP argument. If the
1328 pattern was studied, the result of the study should be passed in the
1329 \fIextra\fP argument. This function is the main matching facility of the
1330 library, and it operates in a Perl-like manner. For specialist use there is
1331 also an alternative matching function, which is described
1332 .\" HTML <a href="#dfamatch">
1333 .\" </a>
1334 below
1335 .\"
1336 in the section about the \fBpcre_dfa_exec()\fP function.
1337 .P
1338 In most applications, the pattern will have been compiled (and optionally
1339 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1340 possible to save compiled patterns and study data, and then use them later
1341 in different processes, possibly even on different hosts. For a discussion
1342 about this, see the
1343 .\" HREF
1344 \fBpcreprecompile\fP
1345 .\"
1346 documentation.
1347 .P
1348 Here is an example of a simple call to \fBpcre_exec()\fP:
1349 .sp
1350 int rc;
1351 int ovector[30];
1352 rc = pcre_exec(
1353 re, /* result of pcre_compile() */
1354 NULL, /* we didn't study the pattern */
1355 "some string", /* the subject string */
1356 11, /* the length of the subject string */
1357 0, /* start at offset 0 in the subject */
1358 0, /* default options */
1359 ovector, /* vector of integers for substring information */
1360 30); /* number of elements (NOT size in bytes) */
1361 .
1362 .
1363 .\" HTML <a name="extradata"></a>
1364 .SS "Extra data for \fBpcre_exec()\fR"
1365 .rs
1366 .sp
1367 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1368 data block. The \fBpcre_study()\fP function returns such a block (when it
1369 doesn't return NULL), but you can also create one for yourself, and pass
1370 additional information in it. The \fBpcre_extra\fP block contains the following
1371 fields (not necessarily in this order):
1372 .sp
1373 unsigned long int \fIflags\fP;
1374 void *\fIstudy_data\fP;
1375 void *\fIexecutable_jit\fP;
1376 unsigned long int \fImatch_limit\fP;
1377 unsigned long int \fImatch_limit_recursion\fP;
1378 void *\fIcallout_data\fP;
1379 const unsigned char *\fItables\fP;
1380 unsigned char **\fImark\fP;
1381 .sp
1382 The \fIflags\fP field is a bitmap that specifies which of the other fields
1383 are set. The flag bits are:
1384 .sp
1392 .sp
1393 Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1394 the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1395 returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1396 should not set these yourself, but you may add to the block by setting the
1397 other fields and their corresponding flag bits.
1398 .P
1399 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1400 vast amount of resources when running patterns that are not going to match,
1401 but which have a very large number of possibilities in their search trees. The
1402 classic example is a pattern that uses nested unlimited repeats.
1403 .P
1404 Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1405 calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1406 imposed on the number of times this function is called during a match, which
1407 has the effect of limiting the amount of backtracking that can take place. For
1408 patterns that are not anchored, the count restarts from zero for each position
1409 in the subject string.
1410 .P
1411 When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1412 with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1413 is entirely different. However, there is still the possibility of runaway
1414 matching that goes on for a very long time, and so the \fImatch_limit\fP value
1415 is also used in this case (but in a different way) to limit how long the
1416 matching can continue.
1417 .P
1418 The default value for the limit can be set when PCRE is built; the default
1419 default is 10 million, which handles all but the most extreme cases. You can
1420 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1421 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1422 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1424 .P
1425 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1426 instead of limiting the total number of times that \fBmatch()\fP is called, it
1427 limits the depth of recursion. The recursion depth is a smaller number than the
1428 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1429 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1430 .P
1431 Limiting the recursion depth limits the amount of machine stack that can be
1432 used, or, when PCRE has been compiled to use memory on the heap instead of the
1433 stack, the amount of heap memory that can be used. This limit is relevant, and
1434 is ignored, when the pattern was successfully studied with
1436 .P
1437 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1438 built; the default default is the same value as the default for
1439 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1440 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1441 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1442 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1443 .P
1444 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1445 and is described in the
1446 .\" HREF
1447 \fBpcrecallout\fP
1448 .\"
1449 documentation.
1450 .P
1451 The \fItables\fP field is used to pass a character tables pointer to
1452 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1453 pattern. A non-NULL value is stored with the compiled pattern only if custom
1454 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1455 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1456 internal tables to be used. This facility is helpful when re-using patterns
1457 that have been saved after compiling with an external set of tables, because
1458 the external tables might be at a different address when \fBpcre_exec()\fP is
1459 called. See the
1460 .\" HREF
1461 \fBpcreprecompile\fP
1462 .\"
1463 documentation for a discussion of saving compiled patterns for later use.
1464 .P
1465 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1466 be set to point to a \fBchar *\fP variable. If the pattern contains any
1467 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1468 a name to pass back, a pointer to the name string (zero terminated) is placed
1469 in the variable pointed to by the \fImark\fP field. The names are within the
1470 compiled pattern; if you wish to retain such a name you must copy it before
1471 freeing the memory of a compiled pattern. If there is no name to pass back, the
1472 variable pointed to by the \fImark\fP field set to NULL. For details of the
1473 backtracking control verbs, see the section entitled
1474 .\" HTML <a href="pcrepattern#backtrackcontrol">
1475 .\" </a>
1476 "Backtracking control"
1477 .\"
1478 in the
1479 .\" HREF
1480 \fBpcrepattern\fP
1481 .\"
1482 documentation.
1483 .
1484 .
1485 .\" HTML <a name="execoptions"></a>
1486 .SS "Option bits for \fBpcre_exec()\fP"
1487 .rs
1488 .sp
1489 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1490 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1494 .sp
1496 .sp
1497 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1498 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1499 to be anchored by virtue of its contents, it cannot be made unachored at
1500 matching time.
1501 .sp
1504 .sp
1505 These options (which are mutually exclusive) control what the \eR escape
1506 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1507 match any Unicode newline sequence. These options override the choice that was
1508 made or defaulted when the pattern was compiled.
1509 .sp
1515 .sp
1516 These options override the newline definition that was chosen or defaulted when
1517 the pattern was compiled. For details, see the description of
1518 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1519 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1520 the way the match position is advanced after a match failure for an unanchored
1521 pattern.
1522 .P
1524 match attempt for an unanchored pattern fails when the current position is at a
1525 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1526 characters, the match position is advanced by two characters instead of one, in
1527 other words, to after the CRLF.
1528 .P
1529 The above rule is a compromise that makes the most common cases work as
1530 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1531 set), it does not match the string "\er\enA" because, after failing at the
1532 start, it skips both the CR and the LF before retrying. However, the pattern
1533 [\er\en]A does match that string, because it contains an explicit CR or LF
1534 reference, and so advances only by one character after the first failure.
1535 .P
1536 An explicit match for CR of LF is either a literal appearance of one of those
1537 characters, or one of the \er or \en escape sequences. Implicit matches such as
1538 [^X] do not count, nor does \es (which includes CR and LF in the characters
1539 that it matches).
1540 .P
1541 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1542 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1543 .sp
1545 .sp
1546 This option specifies that first character of the subject string is not the
1547 beginning of a line, so the circumflex metacharacter should not match before
1548 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1549 never to match. This option affects only the behaviour of the circumflex
1550 metacharacter. It does not affect \eA.
1551 .sp
1553 .sp
1554 This option specifies that the end of the subject string is not the end of a
1555 line, so the dollar metacharacter should not match it nor (except in multiline
1556 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1557 compile time) causes dollar never to match. This option affects only the
1558 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1559 .sp
1561 .sp
1562 An empty string is not considered to be a valid match if this option is set. If
1563 there are alternatives in the pattern, they are tried. If all the alternatives
1564 match the empty string, the entire match fails. For example, if the pattern
1565 .sp
1566 a?b?
1567 .sp
1568 is applied to a string not beginning with "a" or "b", it matches an empty
1569 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1570 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1571 .sp
1573 .sp
1574 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1575 the start of the subject is permitted. If the pattern is anchored, such a match
1576 can occur only if the pattern contains \eK.
1577 .P
1578 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1579 does make a special case of a pattern match of the empty string within its
1580 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1581 emulate Perl's behaviour after matching a null string by first trying the match
1582 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1583 if that fails, by advancing the starting offset (see below) and trying an
1584 ordinary match again. There is some code that demonstrates how to do this in
1585 the
1586 .\" HREF
1587 \fBpcredemo\fP
1588 .\"
1589 sample program. In the most general case, you have to check to see if the
1590 newline convention recognizes CRLF as a newline, and if so, and the current
1591 character is CR followed by LF, advance the starting offset by two characters
1592 instead of one.
1593 .sp
1595 .sp
1596 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1597 a match, in order to speed up the process. For example, if it is known that an
1598 unanchored match must start with a specific character, it searches the subject
1599 for that character, and fails immediately if it cannot find it, without
1600 actually running the main matching function. This means that a special item
1601 such as (*COMMIT) at the start of a pattern is not considered until after a
1602 suitable starting point for the match has been found. When callouts or (*MARK)
1603 items are in use, these "start-up" optimizations can cause them to be skipped
1604 if the pattern is never actually used. The start-up optimizations are in effect
1605 a pre-scan of the subject that takes place before the pattern is run.
1606 .P
1607 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1608 causing performance to suffer, but ensuring that in cases where the result is
1609 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1610 are considered at every possible starting position in the subject string. If
1611 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1612 time.
1613 .P
1614 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1615 Consider the pattern
1616 .sp
1618 .sp
1619 When this is compiled, PCRE records the fact that a match must start with the
1620 character "A". Suppose the subject string is "DEFABC". The start-up
1621 optimization scans along the subject, finds "A" and runs the first match
1622 attempt from there. The (*COMMIT) item means that the pattern must match the
1623 current starting position, which in this case, it does. However, if the same
1624 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1625 subject string does not happen. The first match attempt is run starting from
1626 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1627 the overall result is "no match". If the pattern is studied, more start-up
1628 optimizations may be used. For example, a minimum length for the subject may be
1629 recorded. Consider the pattern
1630 .sp
1631 (*MARK:A)(X|Y)
1632 .sp
1633 The minimum length for a match is one character. If the subject is "ABC", there
1634 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1635 If the pattern is studied, the final attempt does not take place, because PCRE
1636 knows that the subject is too short, and so the (*MARK) is never encountered.
1637 In this case, studying the pattern does not affect the overall match result,
1638 which is still "no match", but it does affect the auxiliary information that is
1639 returned.
1640 .sp
1642 .sp
1643 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1644 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1645 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1646 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1647 strings in the
1648 .\" HTML <a href="pcre.html#utf8strings">
1649 .\" </a>
1650 section on UTF-8 support
1651 .\"
1652 in the main
1653 .\" HREF
1654 \fBpcre\fP
1655 .\"
1656 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1657 the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1658 a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1659 both cases, information about the precise nature of the error may also be
1660 returned (see the descriptions of these errors in the section entitled \fIError
1661 return values from\fP \fBpcre_exec()\fP
1662 .\" HTML <a href="#errorlist">
1663 .\" </a>
1664 below).
1665 .\"
1666 If \fIstartoffset\fP contains a value that does not point to the start of a
1667 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1668 returned.
1669 .P
1670 If you already know that your subject is valid, and you want to skip these
1671 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1672 calling \fBpcre_exec()\fP. You might want to do this for the second and
1673 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1674 all the matches in a single subject string. However, you should be sure that
1675 the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1676 end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1677 invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1678 undefined. Your program may crash.
1679 .sp
1682 .sp
1683 These options turn on the partial matching feature. For backwards
1684 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1685 occurs if the end of the subject string is reached successfully, but there are
1686 not enough subject characters to complete the match. If this happens when
1687 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1688 testing any remaining alternatives. Only if no complete match can be found is
1689 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1690 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1691 but only if no complete match can be found.
1692 .P
1693 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1694 partial match is found, \fBpcre_exec()\fP immediately returns
1695 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1696 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1697 important that an alternative complete match.
1698 .P
1699 In both cases, the portion of the string that was inspected when the partial
1700 match was found is set as the first matching string. There is a more detailed
1701 discussion of partial and multi-segment matching, with examples, in the
1702 .\" HREF
1703 \fBpcrepartial\fP
1704 .\"
1705 documentation.
1706 .
1707 .
1708 .SS "The string to be matched by \fBpcre_exec()\fP"
1709 .rs
1710 .sp
1711 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1712 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1713 in \fIstartoffset\fP. If this is negative or greater than the length of the
1714 subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1715 offset is zero, the search for a match starts at the beginning of the subject,
1716 and this is by far the most common case. In UTF-8 mode, the byte offset must
1717 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1718 pattern string, the subject may contain binary zero bytes.
1719 .P
1720 A non-zero starting offset is useful when searching for another match in the
1721 same subject by calling \fBpcre_exec()\fP again after a previous success.
1722 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1723 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1724 lookbehind. For example, consider the pattern
1725 .sp
1726 \eBiss\eB
1727 .sp
1728 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1729 the current position in the subject is not a word boundary.) When applied to
1730 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1731 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1732 subject, namely "issipi", it does not match, because \eB is always false at the
1733 start of the subject, which is deemed to be a word boundary. However, if
1734 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1735 set to 4, it finds the second occurrence of "iss" because it is able to look
1736 behind the starting point to discover that it is preceded by a letter.
1737 .P
1738 Finding all the matches in a subject is tricky when the pattern can match an
1739 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1740 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1741 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1742 and trying an ordinary match again. There is some code that demonstrates how to
1743 do this in the
1744 .\" HREF
1745 \fBpcredemo\fP
1746 .\"
1747 sample program. In the most general case, you have to check to see if the
1748 newline convention recognizes CRLF as a newline, and if so, and the current
1749 character is CR followed by LF, advance the starting offset by two characters
1750 instead of one.
1751 .P
1752 If a non-zero starting offset is passed when the pattern is anchored, one
1753 attempt to match at the given offset is made. This can only succeed if the
1754 pattern does not require the match to be at the start of the subject.
1755 .
1756 .
1757 .SS "How \fBpcre_exec()\fP returns captured substrings"
1758 .rs
1759 .sp
1760 In general, a pattern matches a certain portion of the subject, and in
1761 addition, further substrings from the subject may be picked out by parts of the
1762 pattern. Following the usage in Jeffrey Friedl's book, this is called
1763 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1764 a fragment of a pattern that picks out a substring. PCRE supports several other
1765 kinds of parenthesized subpattern that do not cause substrings to be captured.
1766 .P
1767 Captured substrings are returned to the caller via a vector of integers whose
1768 address is passed in \fIovector\fP. The number of elements in the vector is
1769 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1770 argument is NOT the size of \fIovector\fP in bytes.
1771 .P
1772 The first two-thirds of the vector is used to pass back captured substrings,
1773 each substring using a pair of integers. The remaining third of the vector is
1774 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1775 and is not available for passing back information. The number passed in
1776 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1777 rounded down.
1778 .P
1779 When a match is successful, information about captured substrings is returned
1780 in pairs of integers, starting at the beginning of \fIovector\fP, and
1781 continuing up to two-thirds of its length at the most. The first element of
1782 each pair is set to the byte offset of the first character in a substring, and
1783 the second is set to the byte offset of the first character after the end of a
1784 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1785 mode. They are not character counts.
1786 .P
1787 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1788 portion of the subject string matched by the entire pattern. The next pair is
1789 used for the first capturing subpattern, and so on. The value returned by
1790 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1791 For example, if two substrings have been captured, the returned value is 3. If
1792 there are no capturing subpatterns, the return value from a successful match is
1793 1, indicating that just the first pair of offsets has been set.
1794 .P
1795 If a capturing subpattern is matched repeatedly, it is the last portion of the
1796 string that it matched that is returned.
1797 .P
1798 If the vector is too small to hold all the captured substring offsets, it is
1799 used as far as possible (up to two-thirds of its length), and the function
1800 returns a value of zero. If the substring offsets are not of interest,
1801 \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1802 \fIovecsize\fP as zero. However, if the pattern contains back references and
1803 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1804 has to get additional memory for use during matching. Thus it is usually
1805 advisable to supply an \fIovector\fP.
1806 .P
1807 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1808 subpatterns there are in a compiled pattern. The smallest size for
1809 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1810 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1811 .P
1812 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1813 the subject when subpattern \fIn\fP has not been used at all. For example, if
1814 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1815 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1816 happens, both values in the offset pairs corresponding to unused subpatterns
1817 are set to -1.
1818 .P
1819 Offset values that correspond to unused subpatterns at the end of the
1820 expression are also set to -1. For example, if the string "abc" is matched
1821 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1822 return from the function is 2, because the highest used capturing subpattern
1823 number is 1, and the offsets for for the second and third capturing subpatterns
1824 (assuming the vector is large enough, of course) are set to -1.
1825 .P
1826 \fBNote\fP: Elements in the first two-thids of \fIovector\fP that do not
1827 correspond to capturing parentheses in the pattern are never changed. That is,
1828 if a pattern contains \fIn\fP capturing parentheses, no more than
1829 \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1830 elements (in the first two-thirds) retain whatever values they previously had.
1831 .P
1832 Some convenience functions are provided for extracting the captured substrings
1833 as separate strings. These are described below.
1834 .
1835 .
1836 .\" HTML <a name="errorlist"></a>
1837 .SS "Error return values from \fBpcre_exec()\fP"
1838 .rs
1839 .sp
1840 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1841 defined in the header file:
1842 .sp
1844 .sp
1845 The subject string did not match the pattern.
1846 .sp
1848 .sp
1849 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1850 NULL and \fIovecsize\fP was not zero.
1851 .sp
1853 .sp
1854 An unrecognized bit was set in the \fIoptions\fP argument.
1855 .sp
1857 .sp
1858 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1859 the case when it is passed a junk pointer and to detect when a pattern that was
1860 compiled in an environment of one endianness is run in an environment with the
1861 other endianness. This is the error that PCRE gives when the magic number is
1862 not present.
1863 .sp
1865 .sp
1866 While running the pattern match, an unknown item was encountered in the
1867 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1868 of the compiled pattern.
1869 .sp
1871 .sp
1872 If a pattern contains back references, but the \fIovector\fP that is passed to
1873 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1874 gets a block of memory at the start of matching to use for this purpose. If the
1875 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1876 automatically freed at the end of matching.
1877 .P
1878 This error is also given if \fBpcre_stack_malloc()\fP fails in
1879 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1880 \fB--disable-stack-for-recursion\fP.
1881 .sp
1883 .sp
1884 This error is used by the \fBpcre_copy_substring()\fP,
1885 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1886 below). It is never returned by \fBpcre_exec()\fP.
1887 .sp
1889 .sp
1890 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1891 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1892 above.
1893 .sp
1895 .sp
1896 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1897 use by callout functions that want to yield a distinctive error code. See the
1898 .\" HREF
1899 \fBpcrecallout\fP
1900 .\"
1901 documentation for details.
1902 .sp
1904 .sp
1905 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1906 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1907 (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1908 UTF-8 character is placed in the first element, and a reason code is placed in
1909 the second element. The reason codes are listed in the
1910 .\" HTML <a href="#badutf8reasons">
1911 .\" </a>
1912 following section.
1913 .\"
1914 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1915 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1916 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1917 .sp
1919 .sp
1920 The UTF-8 byte sequence that was passed as a subject was checked and found to
1921 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
1922 \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1923 end of the subject.
1924 .sp
1926 .sp
1927 The subject string did not match, but it did match partially. See the
1928 .\" HREF
1929 \fBpcrepartial\fP
1930 .\"
1931 documentation for details of partial matching.
1932 .sp
1934 .sp
1935 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1936 option was used with a compiled pattern containing items that were not
1937 supported for partial matching. From release 8.00 onwards, there are no
1938 restrictions on partial matching.
1939 .sp
1941 .sp
1942 An unexpected internal error has occurred. This error could be caused by a bug
1943 in PCRE or by overwriting of the compiled pattern.
1944 .sp
1946 .sp
1947 This error is given if the value of the \fIovecsize\fP argument is negative.
1948 .sp
1950 .sp
1951 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1952 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1953 description above.
1954 .sp
1956 .sp
1957 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1958 .sp
1960 .sp
1961 The value of \fIstartoffset\fP was negative or greater than the length of the
1962 subject, that is, the value in \fIlength\fP.
1963 .sp
1965 .sp
1966 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
1967 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
1968 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
1969 fact sufficient to detect this case, but this special error code for
1970 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
1971 retained for backwards compatibility.
1972 .sp
1974 .sp
1975 This error is returned when \fBpcre_exec()\fP detects a recursion loop within
1976 the pattern. Specifically, it means that either the whole pattern or a
1977 subpattern has been called recursively for the second time at the same position
1978 in the subject string. Some simple patterns that might do this are detected and
1979 faulted at compile time, but more complicated cases, in particular mutual
1980 recursions between two different subpatterns, cannot be detected until run
1981 time.
1982 .sp
1984 .sp
1985 This error is returned when a pattern that was successfully studied using the
1986 PCRE_STUDY_JIT_COMPILE option is matched, but the memory available for the
1987 just-in-time processing stack is not large enough. See the
1988 .\" HREF
1989 \fBpcrejit\fP
1990 .\"
1991 documentation for more details.
1992 .P
1993 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1994 .
1995 .
1996 .\" HTML <a name="badutf8reasons"></a>
1997 .SS "Reason codes for invalid UTF-8 strings"
1998 .rs
1999 .sp
2000 When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2001 PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2002 least 2, the offset of the start of the invalid UTF-8 character is placed in
2003 the first output vector element (\fIovector[0]\fP) and a reason code is placed
2004 in the second element (\fIovector[1]\fP). The reason codes are given names in
2005 the \fBpcre.h\fP header file:
2006 .sp
2012 .sp
2013 The string ends with a truncated UTF-8 character; the code specifies how many
2014 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2015 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2016 allows for up to 6 bytes, and this is checked first; hence the possibility of
2017 4 or 5 missing bytes.
2018 .sp
2023 PCRE_UTF8_ERR10
2024 .sp
2025 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2026 character do not have the binary value 0b10 (that is, either the most
2027 significant bit is 0, or the next bit is 1).
2028 .sp
2029 PCRE_UTF8_ERR11
2030 PCRE_UTF8_ERR12
2031 .sp
2032 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2033 these code points are excluded by RFC 3629.
2034 .sp
2035 PCRE_UTF8_ERR13
2036 .sp
2037 A 4-byte character has a value greater than 0x10fff; these code points are
2038 excluded by RFC 3629.
2039 .sp
2040 PCRE_UTF8_ERR14
2041 .sp
2042 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2043 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2044 from UTF-8.
2045 .sp
2046 PCRE_UTF8_ERR15
2047 PCRE_UTF8_ERR16
2048 PCRE_UTF8_ERR17
2049 PCRE_UTF8_ERR18
2050 PCRE_UTF8_ERR19
2051 .sp
2052 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2053 value that can be represented by fewer bytes, which is invalid. For example,
2054 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2055 one byte.
2056 .sp
2057 PCRE_UTF8_ERR20
2058 .sp
2059 The two most significant bits of the first byte of a character have the binary
2060 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2061 byte can only validly occur as the second or subsequent byte of a multi-byte
2062 character.
2063 .sp
2064 PCRE_UTF8_ERR21
2065 .sp
2066 The first byte of a character has the value 0xfe or 0xff. These values can
2067 never occur in a valid UTF-8 string.
2068 .
2069 .
2071 .rs
2072 .sp
2073 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2074 .ti +5n
2075 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
2076 .ti +5n
2077 .B int \fIbuffersize\fP);
2078 .PP
2079 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2080 .ti +5n
2081 .B int \fIstringcount\fP, int \fIstringnumber\fP,
2082 .ti +5n
2083 .B const char **\fIstringptr\fP);
2084 .PP
2085 .B int pcre_get_substring_list(const char *\fIsubject\fP,
2086 .ti +5n
2087 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
2088 .PP
2089 Captured substrings can be accessed directly by using the offsets returned by
2090 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
2091 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
2092 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2093 as new, separate, zero-terminated strings. These functions identify substrings
2094 by number. The next section describes functions for extracting named
2095 substrings.
2096 .P
2097 A substring that contains a binary zero is correctly extracted and has a
2098 further zero added on the end, but the result is not, of course, a C string.
2099 However, you can process such a string by referring to the length that is
2100 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2101 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2102 for handling strings containing binary zeros, because the end of the final
2103 string is not independently indicated.
2104 .P
2105 The first three arguments are the same for all three of these functions:
2106 \fIsubject\fP is the subject string that has just been successfully matched,
2107 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
2108 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
2109 captured by the match, including the substring that matched the entire regular
2110 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
2111 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
2112 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
2113 number of elements in the vector divided by three.
2114 .P
2115 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
2116 extract a single substring, whose number is given as \fIstringnumber\fP. A
2117 value of zero extracts the substring that matched the entire pattern, whereas
2118 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2119 the string is placed in \fIbuffer\fP, whose length is given by
2120 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2121 obtained via \fBpcre_malloc\fP, and its address is returned via
2122 \fIstringptr\fP. The yield of the function is the length of the string, not
2123 including the terminating zero, or one of these error codes:
2124 .sp
2126 .sp
2127 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2128 memory failed for \fBpcre_get_substring()\fP.
2129 .sp
2131 .sp
2132 There is no substring whose number is \fIstringnumber\fP.
2133 .P
2134 The \fBpcre_get_substring_list()\fP function extracts all available substrings
2135 and builds a list of pointers to them. All this is done in a single block of
2136 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2137 is returned via \fIlistptr\fP, which is also the start of the list of string
2138 pointers. The end of the list is marked by a NULL pointer. The yield of the
2139 function is zero if all went well, or the error code
2140 .sp
2142 .sp
2143 if the attempt to get the memory block failed.
2144 .P
2145 When any of these functions encounter a substring that is unset, which can
2146 happen when capturing subpattern number \fIn+1\fP matches some part of the
2147 subject, but subpattern \fIn\fP has not been used at all, they return an empty
2148 string. This can be distinguished from a genuine zero-length substring by
2149 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2150 substrings.
2151 .P
2152 The two convenience functions \fBpcre_free_substring()\fP and
2153 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2154 a previous call of \fBpcre_get_substring()\fP or
2155 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2156 the function pointed to by \fBpcre_free\fP, which of course could be called
2157 directly from a C program. However, PCRE is used in some situations where it is
2158 linked via a special interface to another programming language that cannot use
2159 \fBpcre_free\fP directly; it is for these cases that the functions are
2160 provided.
2161 .
2162 .
2164 .rs
2165 .sp
2166 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2167 .ti +5n
2168 .B const char *\fIname\fP);
2169 .PP
2170 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2171 .ti +5n
2172 .B const char *\fIsubject\fP, int *\fIovector\fP,
2173 .ti +5n
2174 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2175 .ti +5n
2176 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2177 .PP
2178 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2179 .ti +5n
2180 .B const char *\fIsubject\fP, int *\fIovector\fP,
2181 .ti +5n
2182 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2183 .ti +5n
2184 .B const char **\fIstringptr\fP);
2185 .PP
2186 To extract a substring by name, you first have to find associated number.
2187 For example, for this pattern
2188 .sp
2189 (a+)b(?<xxx>\ed+)...
2190 .sp
2191 the number of the subpattern called "xxx" is 2. If the name is known to be
2192 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2193 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2194 pattern, and the second is the name. The yield of the function is the
2195 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2196 that name.
2197 .P
2198 Given the number, you can extract the substring directly, or use one of the
2199 functions described in the previous section. For convenience, there are also
2200 two functions that do the whole job.
2201 .P
2202 Most of the arguments of \fBpcre_copy_named_substring()\fP and
2203 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2204 functions that extract by number. As these are described in the previous
2205 section, they are not re-described here. There are just two differences:
2206 .P
2207 First, instead of a substring number, a substring name is given. Second, there
2208 is an extra argument, given at the start, which is a pointer to the compiled
2209 pattern. This is needed in order to gain access to the name-to-number
2210 translation table.
2211 .P
2212 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2213 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2214 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2215 the behaviour may not be what you want (see the next section).
2216 .P
2217 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2218 subpatterns with the same number, as described in the
2219 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2220 .\" </a>
2221 section on duplicate subpattern numbers
2222 .\"
2223 in the
2224 .\" HREF
2225 \fBpcrepattern\fP
2226 .\"
2227 page, you cannot use names to distinguish the different subpatterns, because
2228 names are not included in the compiled code. The matching process uses only
2229 numbers. For this reason, the use of different names for subpatterns of the
2230 same number causes an error at compile time.
2231 .
2233 .rs
2234 .sp
2235 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2236 .ti +5n
2237 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2238 .PP
2239 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2240 are not required to be unique. (Duplicate names are always allowed for
2241 subpatterns with the same number, created by using the (?| feature. Indeed, if
2242 such subpatterns are named, they are required to use the same names.)
2243 .P
2244 Normally, patterns with duplicate names are such that in any one match, only
2245 one of the named subpatterns participates. An example is shown in the
2246 .\" HREF
2247 \fBpcrepattern\fP
2248 .\"
2249 documentation.
2250 .P
2251 When duplicates are present, \fBpcre_copy_named_substring()\fP and
2252 \fBpcre_get_named_substring()\fP return the first substring corresponding to
2253 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2254 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2255 returns one of the numbers that are associated with the name, but it is not
2256 defined which it is.
2257 .P
2258 If you want to get full details of all captured substrings for a given name,
2259 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2260 argument is the compiled pattern, and the second is the name. The third and
2261 fourth are pointers to variables which are updated by the function. After it
2262 has run, they point to the first and last entries in the name-to-number table
2263 for the given name. The function itself returns the length of each entry, or
2264 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2265 described above in the section entitled \fIInformation about a pattern\fP
2266 .\" HTML <a href="#infoaboutpattern">
2267 .\" </a>
2268 above.
2269 .\"
2270 Given all the relevant entries for the name, you can extract each of their
2271 numbers, and hence the captured data, if any.
2272 .
2273 .
2275 .rs
2276 .sp
2277 The traditional matching function uses a similar algorithm to Perl, which stops
2278 when it finds the first match, starting at a given point in the subject. If you
2279 want to find all possible matches, or the longest possible match, consider
2280 using the alternative matching function (see below) instead. If you cannot use
2281 the alternative function, but still need to find all possible matches, you
2282 can kludge it up by making use of the callout facility, which is described in
2283 the
2284 .\" HREF
2285 \fBpcrecallout\fP
2286 .\"
2287 documentation.
2288 .P
2289 What you have to do is to insert a callout right at the end of the pattern.
2290 When your callout function is called, extract and save the current matched
2291 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2292 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2293 will yield PCRE_ERROR_NOMATCH.
2294 .
2295 .
2296 .\" HTML <a name="dfamatch"></a>
2298 .rs
2299 .sp
2300 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2301 .ti +5n
2302 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2303 .ti +5n
2304 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2305 .ti +5n
2306 .B int *\fIworkspace\fP, int \fIwscount\fP);
2307 .P
2308 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2309 a compiled pattern, using a matching algorithm that scans the subject string
2310 just once, and does not backtrack. This has different characteristics to the
2311 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2312 patterns are not supported. Nevertheless, there are times when this kind of
2313 matching can be useful. For a discussion of the two matching algorithms, and a
2314 list of features that \fBpcre_dfa_exec()\fP does not support, see the
2315 .\" HREF
2316 \fBpcrematching\fP
2317 .\"
2318 documentation.
2319 .P
2320 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2321 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2322 different way, and this is described below. The other common arguments are used
2323 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2324 here.
2325 .P
2326 The two additional arguments provide workspace for the function. The workspace
2327 vector should contain at least 20 elements. It is used for keeping track of
2328 multiple paths through the pattern tree. More workspace will be needed for
2329 patterns and subjects where there are a lot of potential matches.
2330 .P
2331 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2332 .sp
2333 int rc;
2334 int ovector[10];
2335 int wspace[20];
2336 rc = pcre_dfa_exec(
2337 re, /* result of pcre_compile() */
2338 NULL, /* we didn't study the pattern */
2339 "some string", /* the subject string */
2340 11, /* the length of the subject string */
2341 0, /* start at offset 0 in the subject */
2342 0, /* default options */
2343 ovector, /* vector of integers for substring information */
2344 10, /* number of elements (NOT size in bytes) */
2345 wspace, /* working space vector */
2346 20); /* number of elements (NOT size in bytes) */
2347 .
2348 .SS "Option bits for \fBpcre_dfa_exec()\fP"
2349 .rs
2350 .sp
2351 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2352 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2356 All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2357 so their description is not repeated here.
2358 .sp
2361 .sp
2362 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2363 details are slightly different. When PCRE_PARTIAL_HARD is set for
2364 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2365 is reached and there is still at least one matching possibility that requires
2366 additional characters. This happens even if some complete matches have also
2367 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2368 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2369 there have been no complete matches, but there is still at least one matching
2370 possibility. The portion of the string that was inspected when the longest
2371 partial match was found is set as the first matching string in both cases.
2372 There is a more detailed discussion of partial and multi-segment matching, with
2373 examples, in the
2374 .\" HREF
2375 \fBpcrepartial\fP
2376 .\"
2377 documentation.
2378 .sp
2380 .sp
2381 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2382 soon as it has found one match. Because of the way the alternative algorithm
2383 works, this is necessarily the shortest possible match at the first possible
2384 matching point in the subject string.
2385 .sp
2387 .sp
2388 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2389 again, with additional subject characters, and have it continue with the same
2390 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2391 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2392 before because data about the match so far is left in them after a partial
2393 match. There is more discussion of this facility in the
2394 .\" HREF
2395 \fBpcrepartial\fP
2396 .\"
2397 documentation.
2398 .
2399 .
2400 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2401 .rs
2402 .sp
2403 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2404 substring in the subject. Note, however, that all the matches from one run of
2405 the function start at the same point in the subject. The shorter matches are
2406 all initial substrings of the longer matches. For example, if the pattern
2407 .sp
2408 <.*>
2409 .sp
2410 is matched against the string
2411 .sp
2412 This is <something> <something else> <something further> no more
2413 .sp
2414 the three matched strings are
2415 .sp
2416 <something>
2417 <something> <something else>
2418 <something> <something else> <something further>
2419 .sp
2420 On success, the yield of the function is a number greater than zero, which is
2421 the number of matched substrings. The substrings themselves are returned in
2422 \fIovector\fP. Each string uses two elements; the first is the offset to the
2423 start, and the second is the offset to the end. In fact, all the strings have
2424 the same start offset. (Space could have been saved by giving this only once,
2425 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2426 returns data, even though the meaning of the strings is different.)
2427 .P
2428 The strings are returned in reverse order of length; that is, the longest
2429 matching string is given first. If there were too many matches to fit into
2430 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2431 the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2432 the entire \fIovector\fP for returning matched strings.
2433 .
2434 .
2435 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2436 .rs
2437 .sp
2438 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2439 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2440 described
2441 .\" HTML <a href="#errorlist">
2442 .\" </a>
2443 above.
2444 .\"
2445 There are in addition the following errors that are specific to
2446 \fBpcre_dfa_exec()\fP:
2447 .sp
2449 .sp
2450 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2451 that it does not support, for instance, the use of \eC or a back reference.
2452 .sp
2454 .sp
2455 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2456 uses a back reference for the condition, or a test for recursion in a specific
2457 group. These are not supported.
2458 .sp
2460 .sp
2461 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2462 block that contains a setting of the \fImatch_limit\fP or
2463 \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2464 meaningless for DFA matching).
2465 .sp
2467 .sp
2468 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2469 \fIworkspace\fP vector.
2470 .sp
2472 .sp
2473 When a recursive subpattern is processed, the matching function calls itself
2474 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2475 error is given if the output vector is not large enough. This should be
2476 extremely rare, as a vector of size 1000 is used.
2477 .
2478 .
2479 .SH "SEE ALSO"
2480 .rs
2481 .sp
2482 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2483 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2484 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2485 .
2486 .
2488 .rs
2489 .sp
2490 .nf
2491 Philip Hazel
2492 University Computing Service
2493 Cambridge CB2 3QH, England.
2494 .fi
2495 .
2496 .
2498 .rs
2499 .sp
2500 .nf
2501 Last updated: 29 August 2011
2502 Copyright (c) 1997-2011 University of Cambridge.
2503 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5