ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 531 - (show annotations)
Tue Jun 1 13:53:26 2010 UTC (11 years, 4 months ago) by ph10
File size: 92360 byte(s)
Check for running out of memory when using heap for recursion.
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 int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29 .ti +5n
30 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31 .ti +5n
32 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33 .PP
34 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35 .ti +5n
36 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37 .ti +5n
38 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39 .ti +5n
40 .B int *\fIworkspace\fP, int \fIwscount\fP);
41 .PP
42 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43 .ti +5n
44 .B const char *\fIsubject\fP, int *\fIovector\fP,
45 .ti +5n
46 .B int \fIstringcount\fP, const char *\fIstringname\fP,
47 .ti +5n
48 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49 .PP
50 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
51 .ti +5n
52 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53 .ti +5n
54 .B int \fIbuffersize\fP);
55 .PP
56 .B int pcre_get_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 const char **\fIstringptr\fP);
63 .PP
64 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65 .ti +5n
66 .B const char *\fIname\fP);
67 .PP
68 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
69 .ti +5n
70 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71 .PP
72 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73 .ti +5n
74 .B int \fIstringcount\fP, int \fIstringnumber\fP,
75 .ti +5n
76 .B const char **\fIstringptr\fP);
77 .PP
78 .B int pcre_get_substring_list(const char *\fIsubject\fP,
79 .ti +5n
80 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81 .PP
82 .B void pcre_free_substring(const char *\fIstringptr\fP);
83 .PP
84 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
85 .PP
86 .B const unsigned char *pcre_maketables(void);
87 .PP
88 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
89 .ti +5n
90 .B int \fIwhat\fP, void *\fIwhere\fP);
91 .PP
92 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93 .B *\fIfirstcharptr\fP);
94 .PP
95 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96 .PP
97 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
98 .PP
99 .B char *pcre_version(void);
100 .PP
101 .B void *(*pcre_malloc)(size_t);
102 .PP
103 .B void (*pcre_free)(void *);
104 .PP
105 .B void *(*pcre_stack_malloc)(size_t);
106 .PP
107 .B void (*pcre_stack_free)(void *);
108 .PP
109 .B int (*pcre_callout)(pcre_callout_block *);
110 .
111 .
113 .rs
114 .sp
115 PCRE has its own native API, which is described in this document. There are
116 also some wrapper functions that correspond to the POSIX regular expression
117 API. These are described in the
118 .\" HREF
119 \fBpcreposix\fP
120 .\"
121 documentation. Both of these APIs define a set of C function calls. A C++
122 wrapper is distributed with PCRE. It is documented in the
123 .\" HREF
124 \fBpcrecpp\fP
125 .\"
126 page.
127 .P
128 The native API C function prototypes are defined in the header file
129 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132 and PCRE_MINOR to contain the major and minor release numbers for the library.
133 Applications can use these to include support for different releases of PCRE.
134 .P
135 In a Windows environment, if you want to statically link an application program
136 against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
137 including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
138 \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
139 \fB__declspec(dllimport)\fP, with unwanted results.
140 .P
141 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
142 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
143 in a Perl-compatible manner. A sample program that demonstrates the simplest
144 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
145 source distribution. A listing of this program is given in the
146 .\" HREF
147 \fBpcredemo\fP
148 .\"
149 documentation, and the
150 .\" HREF
151 \fBpcresample\fP
152 .\"
153 documentation describes how to compile and run it.
154 .P
155 A second matching function, \fBpcre_dfa_exec()\fP, which is not
156 Perl-compatible, is also provided. This uses a different algorithm for the
157 matching. The alternative algorithm finds all possible matches (at a given
158 point in the subject), and scans the subject just once (unless there are
159 lookbehind assertions). However, this algorithm does not return captured
160 substrings. A description of the two matching algorithms and their advantages
161 and disadvantages is given in the
162 .\" HREF
163 \fBpcrematching\fP
164 .\"
165 documentation.
166 .P
167 In addition to the main compiling and matching functions, there are convenience
168 functions for extracting captured substrings from a subject string that is
169 matched by \fBpcre_exec()\fP. They are:
170 .sp
171 \fBpcre_copy_substring()\fP
172 \fBpcre_copy_named_substring()\fP
173 \fBpcre_get_substring()\fP
174 \fBpcre_get_named_substring()\fP
175 \fBpcre_get_substring_list()\fP
176 \fBpcre_get_stringnumber()\fP
177 \fBpcre_get_stringtable_entries()\fP
178 .sp
179 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
180 provided, to free the memory used for extracted strings.
181 .P
182 The function \fBpcre_maketables()\fP is used to build a set of character tables
183 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
184 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
185 specialist use. Most commonly, no special tables are passed, in which case
186 internal tables that are generated when PCRE is built are used.
187 .P
188 The function \fBpcre_fullinfo()\fP is used to find out information about a
189 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
190 some of the available information, but is retained for backwards compatibility.
191 The function \fBpcre_version()\fP returns a pointer to a string containing the
192 version of PCRE and its date of release.
193 .P
194 The function \fBpcre_refcount()\fP maintains a reference count in a data block
195 containing a compiled pattern. This is provided for the benefit of
196 object-oriented applications.
197 .P
198 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
199 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
200 respectively. PCRE calls the memory management functions via these variables,
201 so a calling program can replace them if it wishes to intercept the calls. This
202 should be done before calling any PCRE functions.
203 .P
204 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
205 indirections to memory management functions. These special functions are used
206 only when PCRE is compiled to use the heap for remembering data, instead of
207 recursive function calls, when running the \fBpcre_exec()\fP function. See the
208 .\" HREF
209 \fBpcrebuild\fP
210 .\"
211 documentation for details of how to do this. It is a non-standard way of
212 building PCRE, for use in environments that have limited stacks. Because of the
213 greater use of memory management, it runs more slowly. Separate functions are
214 provided so that special-purpose external code can be used for this case. When
215 used, these functions are always called in a stack-like manner (last obtained,
216 first freed), and always for memory blocks of the same size. There is a
217 discussion about PCRE's stack usage in the
218 .\" HREF
219 \fBpcrestack\fP
220 .\"
221 documentation.
222 .P
223 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
224 by the caller to a "callout" function, which PCRE will then call at specified
225 points during a matching operation. Details are given in the
226 .\" HREF
227 \fBpcrecallout\fP
228 .\"
229 documentation.
230 .
231 .
232 .\" HTML <a name="newlines"></a>
234 .rs
235 .sp
236 PCRE supports five different conventions for indicating line breaks in
237 strings: a single CR (carriage return) character, a single LF (linefeed)
238 character, the two-character sequence CRLF, any of the three preceding, or any
239 Unicode newline sequence. The Unicode newline sequences are the three just
240 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
241 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
242 (paragraph separator, U+2029).
243 .P
244 Each of the first three conventions is used by at least one operating system as
245 its standard newline sequence. When PCRE is built, a default can be specified.
246 The default default is LF, which is the Unix standard. When PCRE is run, the
247 default can be overridden, either when a pattern is compiled, or when it is
248 matched.
249 .P
250 At compile time, the newline convention can be specified by the \fIoptions\fP
251 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
252 start of the pattern itself; this overrides any other settings. See the
253 .\" HREF
254 \fBpcrepattern\fP
255 .\"
256 page for details of the special character sequences.
257 .P
258 In the PCRE documentation the word "newline" is used to mean "the character or
259 pair of characters that indicate a line break". The choice of newline
260 convention affects the handling of the dot, circumflex, and dollar
261 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
262 recognized line ending sequence, the match position advancement for a
263 non-anchored pattern. There is more detail about this in the
264 .\" HTML <a href="#execoptions">
265 .\" </a>
266 section on \fBpcre_exec()\fP options
267 .\"
268 below.
269 .P
270 The choice of newline convention does not affect the interpretation of
271 the \en or \er escape sequences, nor does it affect what \eR matches, which is
272 controlled in a similar way, but by separate options.
273 .
274 .
276 .rs
277 .sp
278 The PCRE functions can be used in multi-threading applications, with the
279 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
280 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
281 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
282 .P
283 The compiled form of a regular expression is not altered during matching, so
284 the same compiled pattern can safely be used by several threads at once.
285 .
286 .
288 .rs
289 .sp
290 The compiled form of a regular expression can be saved and re-used at a later
291 time, possibly by a different program, and even on a host other than the one on
292 which it was compiled. Details are given in the
293 .\" HREF
294 \fBpcreprecompile\fP
295 .\"
296 documentation. However, compiling a regular expression with one version of PCRE
297 for use with a different version is not guaranteed to work and may cause
298 crashes.
299 .
300 .
302 .rs
303 .sp
304 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
305 .PP
306 The function \fBpcre_config()\fP makes it possible for a PCRE client to
307 discover which optional features have been compiled into the PCRE library. The
308 .\" HREF
309 \fBpcrebuild\fP
310 .\"
311 documentation has more details about these optional features.
312 .P
313 The first argument for \fBpcre_config()\fP is an integer, specifying which
314 information is required; the second argument is a pointer to a variable into
315 which the information is placed. The following information is available:
316 .sp
318 .sp
319 The output is an integer that is set to one if UTF-8 support is available;
320 otherwise it is set to zero.
321 .sp
323 .sp
324 The output is an integer that is set to one if support for Unicode character
325 properties is available; otherwise it is set to zero.
326 .sp
328 .sp
329 The output is an integer whose value specifies the default character sequence
330 that is recognized as meaning "newline". The four values that are supported
331 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
332 Though they are derived from ASCII, the same values are returned in EBCDIC
333 environments. The default should normally correspond to the standard sequence
334 for your operating system.
335 .sp
337 .sp
338 The output is an integer whose value indicates what character sequences the \eR
339 escape sequence matches by default. A value of 0 means that \eR matches any
340 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
341 or CRLF. The default can be overridden when a pattern is compiled or matched.
342 .sp
344 .sp
345 The output is an integer that contains the number of bytes used for internal
346 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
347 allow larger regular expressions to be compiled, at the expense of slower
348 matching. The default value of 2 is sufficient for all but the most massive
349 patterns, since it allows the compiled pattern to be up to 64K in size.
350 .sp
352 .sp
353 The output is an integer that contains the threshold above which the POSIX
354 interface uses \fBmalloc()\fP for output vectors. Further details are given in
355 the
356 .\" HREF
357 \fBpcreposix\fP
358 .\"
359 documentation.
360 .sp
362 .sp
363 The output is a long integer that gives the default limit for the number of
364 internal matching function calls in a \fBpcre_exec()\fP execution. Further
365 details are given with \fBpcre_exec()\fP below.
366 .sp
368 .sp
369 The output is a long integer that gives the default limit for the depth of
370 recursion when calling the internal matching function in a \fBpcre_exec()\fP
371 execution. Further details are given with \fBpcre_exec()\fP below.
372 .sp
374 .sp
375 The output is an integer that is set to one if internal recursion when running
376 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
377 to remember their state. This is the usual way that PCRE is compiled. The
378 output is zero if PCRE was compiled to use blocks of data on the heap instead
379 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
380 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
381 avoiding the use of the stack.
382 .
383 .
385 .rs
386 .sp
387 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
388 .ti +5n
389 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
390 .ti +5n
391 .B const unsigned char *\fItableptr\fP);
392 .sp
393 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
394 .ti +5n
395 .B int *\fIerrorcodeptr\fP,
396 .ti +5n
397 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
398 .ti +5n
399 .B const unsigned char *\fItableptr\fP);
400 .P
401 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
402 called to compile a pattern into an internal form. The only difference between
403 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
404 \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
405 too much repetition, we refer just to \fBpcre_compile()\fP below, but the
406 information applies equally to \fBpcre_compile2()\fP.
407 .P
408 The pattern is a C string terminated by a binary zero, and is passed in the
409 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
410 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
411 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
412 for a structure whose contents are not externally defined. It is up to the
413 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
414 .P
415 Although the compiled code of a PCRE regex is relocatable, that is, it does not
416 depend on memory location, the complete \fBpcre\fP data block is not
417 fully relocatable, because it may contain a copy of the \fItableptr\fP
418 argument, which is an address (see below).
419 .P
420 The \fIoptions\fP argument contains various bit settings that affect the
421 compilation. It should be zero if no options are required. The available
422 options are described below. Some of them (in particular, those that are
423 compatible with Perl, but some others as well) can also be set and unset from
424 within the pattern (see the detailed description in the
425 .\" HREF
426 \fBpcrepattern\fP
427 .\"
428 documentation). For those options that can be different in different parts of
429 the pattern, the contents of the \fIoptions\fP argument specifies their
430 settings at the start of compilation and execution. The PCRE_ANCHORED,
431 PCRE_BSR_\fIxxx\fP, and PCRE_NEWLINE_\fIxxx\fP options can be set at the time
432 of matching as well as at compile time.
433 .P
434 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
435 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
436 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
437 error message. This is a static string that is part of the library. You must
438 not try to free it. The byte offset from the start of the pattern to the
439 character that was being processed when the error was discovered is placed in
440 the variable pointed to by \fIerroffset\fP, which must not be NULL. If it is,
441 an immediate error is given. Some errors are not detected until checks are
442 carried out when the whole pattern has been scanned; in this case the offset is
443 set to the end of the pattern.
444 .P
445 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
446 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
447 returned via this argument in the event of an error. This is in addition to the
448 textual error message. Error codes and messages are listed below.
449 .P
450 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
451 character tables that are built when PCRE is compiled, using the default C
452 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
453 call to \fBpcre_maketables()\fP. This value is stored with the compiled
454 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
455 passed to it. For more discussion, see the section on locale support below.
456 .P
457 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
458 .sp
459 pcre *re;
460 const char *error;
461 int erroffset;
462 re = pcre_compile(
463 "^A.*Z", /* the pattern */
464 0, /* default options */
465 &error, /* for error message */
466 &erroffset, /* for error offset */
467 NULL); /* use default character tables */
468 .sp
469 The following names for option bits are defined in the \fBpcre.h\fP header
470 file:
471 .sp
473 .sp
474 If this bit is set, the pattern is forced to be "anchored", that is, it is
475 constrained to match only at the first matching point in the string that is
476 being searched (the "subject string"). This effect can also be achieved by
477 appropriate constructs in the pattern itself, which is the only way to do it in
478 Perl.
479 .sp
481 .sp
482 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
483 all with number 255, before each pattern item. For discussion of the callout
484 facility, see the
485 .\" HREF
486 \fBpcrecallout\fP
487 .\"
488 documentation.
489 .sp
492 .sp
493 These options (which are mutually exclusive) control what the \eR escape
494 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
495 match any Unicode newline sequence. The default is specified when PCRE is
496 built. It can be overridden from within the pattern, or by setting an option
497 when a compiled pattern is matched.
498 .sp
500 .sp
501 If this bit is set, letters in the pattern match both upper and lower case
502 letters. It is equivalent to Perl's /i option, and it can be changed within a
503 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
504 concept of case for characters whose values are less than 128, so caseless
505 matching is always possible. For characters with higher values, the concept of
506 case is supported if PCRE is compiled with Unicode property support, but not
507 otherwise. If you want to use caseless matching for characters 128 and above,
508 you must ensure that PCRE is compiled with Unicode property support as well as
509 with UTF-8 support.
510 .sp
512 .sp
513 If this bit is set, a dollar metacharacter in the pattern matches only at the
514 end of the subject string. Without this option, a dollar also matches
515 immediately before a newline at the end of the string (but not before any other
516 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
517 There is no equivalent to this option in Perl, and no way to set it within a
518 pattern.
519 .sp
521 .sp
522 If this bit is set, a dot metacharater in the pattern matches all characters,
523 including those that indicate newline. Without it, a dot does not match when
524 the current position is at a newline. This option is equivalent to Perl's /s
525 option, and it can be changed within a pattern by a (?s) option setting. A
526 negative class such as [^a] always matches newline characters, independent of
527 the setting of this option.
528 .sp
530 .sp
531 If this bit is set, names used to identify capturing subpatterns need not be
532 unique. This can be helpful for certain types of pattern when it is known that
533 only one instance of the named subpattern can ever be matched. There are more
534 details of named subpatterns below; see also the
535 .\" HREF
536 \fBpcrepattern\fP
537 .\"
538 documentation.
539 .sp
541 .sp
542 If this bit is set, whitespace data characters in the pattern are totally
543 ignored except when escaped or inside a character class. Whitespace does not
544 include the VT character (code 11). In addition, characters between an
545 unescaped # outside a character class and the next newline, inclusive, are also
546 ignored. This is equivalent to Perl's /x option, and it can be changed within a
547 pattern by a (?x) option setting.
548 .P
549 This option makes it possible to include comments inside complicated patterns.
550 Note, however, that this applies only to data characters. Whitespace characters
551 may never appear within special character sequences in a pattern, for example
552 within the sequence (?( which introduces a conditional subpattern.
553 .sp
555 .sp
556 This option was invented in order to turn on additional functionality of PCRE
557 that is incompatible with Perl, but it is currently of very little use. When
558 set, any backslash in a pattern that is followed by a letter that has no
559 special meaning causes an error, thus reserving these combinations for future
560 expansion. By default, as in Perl, a backslash followed by a letter with no
561 special meaning is treated as a literal. (Perl can, however, be persuaded to
562 give an error for this, by running it with the -w option.) There are at present
563 no other features controlled by this option. It can also be set by a (?X)
564 option setting within a pattern.
565 .sp
567 .sp
568 If this option is set, an unanchored pattern is required to match before or at
569 the first newline in the subject string, though the matched text may continue
570 over the newline.
571 .sp
573 .sp
574 If this option is set, PCRE's behaviour is changed in some ways so that it is
575 compatible with JavaScript rather than Perl. The changes are as follows:
576 .P
577 (1) A lone closing square bracket in a pattern causes a compile-time error,
578 because this is illegal in JavaScript (by default it is treated as a data
579 character). Thus, the pattern AB]CD becomes illegal when this option is set.
580 .P
581 (2) At run time, a back reference to an unset subpattern group matches an empty
582 string (by default this causes the current matching alternative to fail). A
583 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
584 an "a" in the subject), whereas it fails by default, for Perl compatibility.
585 .sp
587 .sp
588 By default, PCRE treats the subject string as consisting of a single line of
589 characters (even if it actually contains newlines). The "start of line"
590 metacharacter (^) matches only at the start of the string, while the "end of
591 line" metacharacter ($) matches only at the end of the string, or before a
592 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
593 Perl.
594 .P
595 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
596 match immediately following or immediately before internal newlines in the
597 subject string, respectively, as well as at the very start and end. This is
598 equivalent to Perl's /m option, and it can be changed within a pattern by a
599 (?m) option setting. If there are no newlines in a subject string, or no
600 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
601 .sp
607 .sp
608 These options override the default newline definition that was chosen when PCRE
609 was built. Setting the first or the second specifies that a newline is
610 indicated by a single character (CR or LF, respectively). Setting
611 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
612 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
613 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
614 that any Unicode newline sequence should be recognized. The Unicode newline
615 sequences are the three just mentioned, plus the single characters VT (vertical
616 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
617 separator, U+2028), and PS (paragraph separator, U+2029). The last two are
618 recognized only in UTF-8 mode.
619 .P
620 The newline setting in the options word uses three bits that are treated
621 as a number, giving eight possibilities. Currently only six are used (default
622 plus the five values above). This means that if you set more than one newline
623 option, the combination may or may not be sensible. For example,
625 other combinations may yield unused numbers and cause an error.
626 .P
627 The only time that a line break is specially recognized when compiling a
628 pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
629 class is encountered. This indicates a comment that lasts until after the next
630 line break sequence. In other circumstances, line break sequences are treated
631 as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
632 as whitespace characters and are therefore ignored.
633 .P
634 The newline option that is set at compile time becomes the default that is used
635 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
636 .sp
638 .sp
639 If this option is set, it disables the use of numbered capturing parentheses in
640 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
641 were followed by ?: but named parentheses can still be used for capturing (and
642 they acquire numbers in the usual way). There is no equivalent of this option
643 in Perl.
644 .sp
646 .sp
647 This option changes the way PCRE processes \eb, \ed, \es, \ew, and some of the
648 POSIX character classes. By default, only ASCII characters are recognized, but
649 if PCRE_UCP is set, Unicode properties are used instead to classify characters.
650 More details are given in the section on
651 .\" HTML <a href="pcre.html#genericchartypes">
652 .\" </a>
653 generic character types
654 .\"
655 in the
656 .\" HREF
657 \fBpcrepattern\fP
658 .\"
659 page. If you set PCRE_UCP, matching one of the items it affects takes much
660 longer. The option is available only if PCRE has been compiled with Unicode
661 property support.
662 .sp
664 .sp
665 This option inverts the "greediness" of the quantifiers so that they are not
666 greedy by default, but become greedy if followed by "?". It is not compatible
667 with Perl. It can also be set by a (?U) option setting within the pattern.
668 .sp
670 .sp
671 This option causes PCRE to regard both the pattern and the subject as strings
672 of UTF-8 characters instead of single-byte character strings. However, it is
673 available only when PCRE is built to include UTF-8 support. If not, the use
674 of this option provokes an error. Details of how this option changes the
675 behaviour of PCRE are given in the
676 .\" HTML <a href="pcre.html#utf8support">
677 .\" </a>
678 section on UTF-8 support
679 .\"
680 in the main
681 .\" HREF
682 \fBpcre\fP
683 .\"
684 page.
685 .sp
687 .sp
688 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
689 automatically checked. There is a discussion about the
690 .\" HTML <a href="pcre.html#utf8strings">
691 .\" </a>
692 validity of UTF-8 strings
693 .\"
694 in the main
695 .\" HREF
696 \fBpcre\fP
697 .\"
698 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
699 returns an error. If you already know that your pattern is valid, and you want
700 to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
701 option. When it is set, the effect of passing an invalid UTF-8 string as a
702 pattern is undefined. It may cause your program to crash. Note that this option
703 can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
704 the UTF-8 validity checking of subject strings.
705 .
706 .
708 .rs
709 .sp
710 The following table lists the error codes than may be returned by
711 \fBpcre_compile2()\fP, along with the error messages that may be returned by
712 both compiling functions. As PCRE has developed, some error codes have fallen
713 out of use. To avoid confusion, they have not been re-used.
714 .sp
715 0 no error
716 1 \e at end of pattern
717 2 \ec at end of pattern
718 3 unrecognized character follows \e
719 4 numbers out of order in {} quantifier
720 5 number too big in {} quantifier
721 6 missing terminating ] for character class
722 7 invalid escape sequence in character class
723 8 range out of order in character class
724 9 nothing to repeat
725 10 [this code is not in use]
726 11 internal error: unexpected repeat
727 12 unrecognized character after (? or (?-
728 13 POSIX named classes are supported only within a class
729 14 missing )
730 15 reference to non-existent subpattern
731 16 erroffset passed as NULL
732 17 unknown option bit(s) set
733 18 missing ) after comment
734 19 [this code is not in use]
735 20 regular expression is too large
736 21 failed to get memory
737 22 unmatched parentheses
738 23 internal error: code overflow
739 24 unrecognized character after (?<
740 25 lookbehind assertion is not fixed length
741 26 malformed number or name after (?(
742 27 conditional group contains more than two branches
743 28 assertion expected after (?(
744 29 (?R or (?[+-]digits must be followed by )
745 30 unknown POSIX class name
746 31 POSIX collating elements are not supported
747 32 this version of PCRE is not compiled with PCRE_UTF8 support
748 33 [this code is not in use]
749 34 character value in \ex{...} sequence is too large
750 35 invalid condition (?(0)
751 36 \eC not allowed in lookbehind assertion
752 37 PCRE does not support \eL, \el, \eN, \eU, or \eu
753 38 number after (?C is > 255
754 39 closing ) for (?C expected
755 40 recursive call could loop indefinitely
756 41 unrecognized character after (?P
757 42 syntax error in subpattern name (missing terminator)
758 43 two named subpatterns have the same name
759 44 invalid UTF-8 string
760 45 support for \eP, \ep, and \eX has not been compiled
761 46 malformed \eP or \ep sequence
762 47 unknown property name after \eP or \ep
763 48 subpattern name is too long (maximum 32 characters)
764 49 too many named subpatterns (maximum 10000)
765 50 [this code is not in use]
766 51 octal value is greater than \e377 (not in UTF-8 mode)
767 52 internal error: overran compiling workspace
768 53 internal error: previously-checked referenced subpattern not found
769 54 DEFINE group contains more than one branch
770 55 repeating a DEFINE group is not allowed
771 56 inconsistent NEWLINE options
772 57 \eg is not followed by a braced, angle-bracketed, or quoted
773 name/number or by a plain number
774 58 a numbered reference must not be zero
775 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
776 60 (*VERB) not recognized
777 61 number is too big
778 62 subpattern name expected
779 63 digit expected after (?+
780 64 ] is an invalid data character in JavaScript compatibility mode
781 65 different names for subpatterns of the same number are not allowed
782 66 (*MARK) must have an argument
783 67 this version of PCRE is not compiled with PCRE_UCP support
784 .sp
785 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
786 be used if the limits were changed when PCRE was built.
787 .
788 .
790 .rs
791 .sp
792 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
793 .ti +5n
794 .B const char **\fIerrptr\fP);
795 .PP
796 If a compiled pattern is going to be used several times, it is worth spending
797 more time analyzing it in order to speed up the time taken for matching. The
798 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
799 argument. If studying the pattern produces additional information that will
800 help speed up matching, \fBpcre_study()\fP returns a pointer to a
801 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
802 results of the study.
803 .P
804 The returned value from \fBpcre_study()\fP can be passed directly to
805 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
806 also contains other fields that can be set by the caller before the block is
807 passed; these are described
808 .\" HTML <a href="#extradata">
809 .\" </a>
810 below
811 .\"
812 in the section on matching a pattern.
813 .P
814 If studying the pattern does not produce any useful information,
815 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
816 wants to pass any of the other fields to \fBpcre_exec()\fP or
817 \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
818 .P
819 The second argument of \fBpcre_study()\fP contains option bits. At present, no
820 options are defined, and this argument should always be zero.
821 .P
822 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
823 studying succeeds (even if no data is returned), the variable it points to is
824 set to NULL. Otherwise it is set to point to a textual error message. This is a
825 static string that is part of the library. You must not try to free it. You
826 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
827 sure that it has run successfully.
828 .P
829 This is a typical call to \fBpcre_study\fP():
830 .sp
831 pcre_extra *pe;
832 pe = pcre_study(
833 re, /* result of pcre_compile() */
834 0, /* no options exist */
835 &error); /* set to NULL or points to a message */
836 .sp
837 Studying a pattern does two things: first, a lower bound for the length of
838 subject string that is needed to match the pattern is computed. This does not
839 mean that there are any strings of that length that match, but it does
840 guarantee that no shorter strings match. The value is used by
841 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
842 match strings that are shorter than the lower bound. You can find out the value
843 in a calling program via the \fBpcre_fullinfo()\fP function.
844 .P
845 Studying a pattern is also useful for non-anchored patterns that do not have a
846 single fixed starting character. A bitmap of possible starting bytes is
847 created. This speeds up finding a position in the subject at which to start
848 matching.
849 .
850 .
851 .\" HTML <a name="localesupport"></a>
853 .rs
854 .sp
855 PCRE handles caseless matching, and determines whether characters are letters,
856 digits, or whatever, by reference to a set of tables, indexed by character
857 value. When running in UTF-8 mode, this applies only to characters with codes
858 less than 128. By default, higher-valued codes never match escapes such as \ew
859 or \ed, but they can be tested with \ep if PCRE is built with Unicode character
860 property support. Alternatively, the PCRE_UCP option can be set at compile
861 time; this causes \ew and friends to use Unicode property support instead of
862 built-in tables. The use of locales with Unicode is discouraged. If you are
863 handling characters with codes greater than 128, you should either use UTF-8
864 and Unicode, or use locales, but not try to mix the two.
865 .P
866 PCRE contains an internal set of tables that are used when the final argument
867 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
868 Normally, the internal tables recognize only ASCII characters. However, when
869 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
870 default "C" locale of the local system, which may cause them to be different.
871 .P
872 The internal tables can always be overridden by tables supplied by the
873 application that calls PCRE. These may be created in a different locale from
874 the default. As more and more applications change to using Unicode, the need
875 for this locale support is expected to die away.
876 .P
877 External tables are built by calling the \fBpcre_maketables()\fP function,
878 which has no arguments, in the relevant locale. The result can then be passed
879 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
880 example, to build and use tables that are appropriate for the French locale
881 (where accented characters with values greater than 128 are treated as letters),
882 the following code could be used:
883 .sp
884 setlocale(LC_CTYPE, "fr_FR");
885 tables = pcre_maketables();
886 re = pcre_compile(..., tables);
887 .sp
888 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
889 are using Windows, the name for the French locale is "french".
890 .P
891 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
892 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
893 that the memory containing the tables remains available for as long as it is
894 needed.
895 .P
896 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
897 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
898 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
899 pattern, compilation, studying and matching all happen in the same locale, but
900 different patterns can be compiled in different locales.
901 .P
902 It is possible to pass a table pointer or NULL (indicating the use of the
903 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
904 this facility could be used to match a pattern in a different locale from the
905 one in which it was compiled. Passing table pointers at run time is discussed
906 below in the section on matching a pattern.
907 .
908 .
910 .rs
911 .sp
912 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
913 .ti +5n
914 .B int \fIwhat\fP, void *\fIwhere\fP);
915 .PP
916 The \fBpcre_fullinfo()\fP function returns information about a compiled
917 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
918 nevertheless retained for backwards compability (and is documented below).
919 .P
920 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
921 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
922 the pattern was not studied. The third argument specifies which piece of
923 information is required, and the fourth argument is a pointer to a variable
924 to receive the data. The yield of the function is zero for success, or one of
925 the following negative numbers:
926 .sp
927 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
928 the argument \fIwhere\fP was NULL
929 PCRE_ERROR_BADMAGIC the "magic number" was not found
930 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
931 .sp
932 The "magic number" is placed at the start of each compiled pattern as an simple
933 check against passing an arbitrary memory pointer. Here is a typical call of
934 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
935 .sp
936 int rc;
937 size_t length;
938 rc = pcre_fullinfo(
939 re, /* result of pcre_compile() */
940 pe, /* result of pcre_study(), or NULL */
941 PCRE_INFO_SIZE, /* what is required */
942 &length); /* where to put the data */
943 .sp
944 The possible values for the third argument are defined in \fBpcre.h\fP, and are
945 as follows:
946 .sp
948 .sp
949 Return the number of the highest back reference in the pattern. The fourth
950 argument should point to an \fBint\fP variable. Zero is returned if there are
951 no back references.
952 .sp
954 .sp
955 Return the number of capturing subpatterns in the pattern. The fourth argument
956 should point to an \fBint\fP variable.
957 .sp
959 .sp
960 Return a pointer to the internal default character tables within PCRE. The
961 fourth argument should point to an \fBunsigned char *\fP variable. This
962 information call is provided for internal use by the \fBpcre_study()\fP
963 function. External callers can cause PCRE to use its internal tables by passing
964 a NULL table pointer.
965 .sp
967 .sp
968 Return information about the first byte of any matched string, for a
969 non-anchored pattern. The fourth argument should point to an \fBint\fP
970 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
971 still recognized for backwards compatibility.)
972 .P
973 If there is a fixed first byte, for example, from a pattern such as
974 (cat|cow|coyote), its value is returned. Otherwise, if either
975 .sp
976 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
977 starts with "^", or
978 .sp
979 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
980 (if it were set, the pattern would be anchored),
981 .sp
982 -1 is returned, indicating that the pattern matches only at the start of a
983 subject string or after any newline within the string. Otherwise -2 is
984 returned. For anchored patterns, -2 is returned.
985 .sp
987 .sp
988 If the pattern was studied, and this resulted in the construction of a 256-bit
989 table indicating a fixed set of bytes for the first byte in any matching
990 string, a pointer to the table is returned. Otherwise NULL is returned. The
991 fourth argument should point to an \fBunsigned char *\fP variable.
992 .sp
994 .sp
995 Return 1 if the pattern contains any explicit matches for CR or LF characters,
996 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
997 explicit match is either a literal CR or LF character, or \er or \en.
998 .sp
1000 .sp
1001 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1002 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1003 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1004 .sp
1006 .sp
1007 Return the value of the rightmost literal byte that must exist in any matched
1008 string, other than at its start, if such a byte has been recorded. The fourth
1009 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1010 returned. For anchored patterns, a last literal byte is recorded only if it
1011 follows something of variable length. For example, for the pattern
1012 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1013 is -1.
1014 .sp
1016 .sp
1017 If the pattern was studied and a minimum length for matching subject strings
1018 was computed, its value is returned. Otherwise the returned value is -1. The
1019 value is a number of characters, not bytes (this may be relevant in UTF-8
1020 mode). The fourth argument should point to an \fBint\fP variable. A
1021 non-negative value is a lower bound to the length of any matching string. There
1022 may not be any strings of that length that do actually match, but every string
1023 that does match is at least that long.
1024 .sp
1028 .sp
1029 PCRE supports the use of named as well as numbered capturing parentheses. The
1030 names are just an additional way of identifying the parentheses, which still
1031 acquire numbers. Several convenience functions such as
1032 \fBpcre_get_named_substring()\fP are provided for extracting captured
1033 substrings by name. It is also possible to extract the data directly, by first
1034 converting the name to a number in order to access the correct pointers in the
1035 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1036 you need to use the name-to-number map, which is described by these three
1037 values.
1038 .P
1039 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1040 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1041 entry; both of these return an \fBint\fP value. The entry size depends on the
1042 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1043 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1044 are the number of the capturing parenthesis, most significant byte first. The
1045 rest of the entry is the corresponding name, zero terminated.
1046 .P
1047 The names are in alphabetical order. Duplicate names may appear if (?| is used
1048 to create multiple groups with the same number, as described in the
1049 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1050 .\" </a>
1051 section on duplicate subpattern numbers
1052 .\"
1053 in the
1054 .\" HREF
1055 \fBpcrepattern\fP
1056 .\"
1057 page. Duplicate names for subpatterns with different numbers are permitted only
1058 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1059 table in the order in which they were found in the pattern. In the absence of
1060 (?| this is the order of increasing number; when (?| is used this is not
1061 necessarily the case because later subpatterns may have lower numbers.
1062 .P
1063 As a simple example of the name/number table, consider the following pattern
1064 (assume PCRE_EXTENDED is set, so white space - including newlines - is
1065 ignored):
1066 .sp
1067 .\" JOIN
1068 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1069 (?<month>\ed\ed) - (?<day>\ed\ed) )
1070 .sp
1071 There are four named subpatterns, so the table has four entries, and each entry
1072 in the table is eight bytes long. The table is as follows, with non-printing
1073 bytes shows in hexadecimal, and undefined bytes shown as ??:
1074 .sp
1075 00 01 d a t e 00 ??
1076 00 05 d a y 00 ?? ??
1077 00 04 m o n t h 00
1078 00 02 y e a r 00 ??
1079 .sp
1080 When writing code to extract data from named subpatterns using the
1081 name-to-number map, remember that the length of the entries is likely to be
1082 different for each compiled pattern.
1083 .sp
1085 .sp
1086 Return 1 if the pattern can be used for partial matching with
1087 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1088 \fBint\fP variable. From release 8.00, this always returns 1, because the
1089 restrictions that previously applied to partial matching have been lifted. The
1090 .\" HREF
1091 \fBpcrepartial\fP
1092 .\"
1093 documentation gives details of partial matching.
1094 .sp
1096 .sp
1097 Return a copy of the options with which the pattern was compiled. The fourth
1098 argument should point to an \fBunsigned long int\fP variable. These option bits
1099 are those specified in the call to \fBpcre_compile()\fP, modified by any
1100 top-level option settings at the start of the pattern itself. In other words,
1101 they are the options that will be in force when matching starts. For example,
1102 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1104 .P
1105 A pattern is automatically anchored by PCRE if all of its top-level
1106 alternatives begin with one of the following:
1107 .sp
1108 ^ unless PCRE_MULTILINE is set
1109 \eA always
1110 \eG always
1111 .\" JOIN
1112 .* if PCRE_DOTALL is set and there are no back
1113 references to the subpattern in which .* appears
1114 .sp
1115 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1116 \fBpcre_fullinfo()\fP.
1117 .sp
1119 .sp
1120 Return the size of the compiled pattern, that is, the value that was passed as
1121 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1122 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1123 variable.
1124 .sp
1126 .sp
1127 Return the size of the data block pointed to by the \fIstudy_data\fP field in
1128 a \fBpcre_extra\fP block. That is, it is the value that was passed to
1129 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1130 created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no
1131 study data, zero is returned. The fourth argument should point to a
1132 \fBsize_t\fP variable.
1133 .
1134 .
1136 .rs
1137 .sp
1138 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1139 .B *\fIfirstcharptr\fP);
1140 .PP
1141 The \fBpcre_info()\fP function is now obsolete because its interface is too
1142 restrictive to return all the available data about a compiled pattern. New
1143 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1144 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1145 following negative numbers:
1146 .sp
1147 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1148 PCRE_ERROR_BADMAGIC the "magic number" was not found
1149 .sp
1150 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1151 pattern was compiled is placed in the integer it points to (see
1152 PCRE_INFO_OPTIONS above).
1153 .P
1154 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1155 it is used to pass back information about the first character of any matched
1156 string (see PCRE_INFO_FIRSTBYTE above).
1157 .
1158 .
1160 .rs
1161 .sp
1162 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1163 .PP
1164 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1165 data block that contains a compiled pattern. It is provided for the benefit of
1166 applications that operate in an object-oriented manner, where different parts
1167 of the application may be using the same compiled pattern, but you want to free
1168 the block when they are all done.
1169 .P
1170 When a pattern is compiled, the reference count field is initialized to zero.
1171 It is changed only by calling this function, whose action is to add the
1172 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1173 function is the new value. However, the value of the count is constrained to
1174 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1175 it is forced to the appropriate limit value.
1176 .P
1177 Except when it is zero, the reference count is not correctly preserved if a
1178 pattern is compiled on one host and then transferred to a host whose byte-order
1179 is different. (This seems a highly unlikely scenario.)
1180 .
1181 .
1183 .rs
1184 .sp
1185 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1186 .ti +5n
1187 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1188 .ti +5n
1189 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1190 .P
1191 The function \fBpcre_exec()\fP is called to match a subject string against a
1192 compiled pattern, which is passed in the \fIcode\fP argument. If the
1193 pattern was studied, the result of the study should be passed in the
1194 \fIextra\fP argument. This function is the main matching facility of the
1195 library, and it operates in a Perl-like manner. For specialist use there is
1196 also an alternative matching function, which is described
1197 .\" HTML <a href="#dfamatch">
1198 .\" </a>
1199 below
1200 .\"
1201 in the section about the \fBpcre_dfa_exec()\fP function.
1202 .P
1203 In most applications, the pattern will have been compiled (and optionally
1204 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1205 possible to save compiled patterns and study data, and then use them later
1206 in different processes, possibly even on different hosts. For a discussion
1207 about this, see the
1208 .\" HREF
1209 \fBpcreprecompile\fP
1210 .\"
1211 documentation.
1212 .P
1213 Here is an example of a simple call to \fBpcre_exec()\fP:
1214 .sp
1215 int rc;
1216 int ovector[30];
1217 rc = pcre_exec(
1218 re, /* result of pcre_compile() */
1219 NULL, /* we didn't study the pattern */
1220 "some string", /* the subject string */
1221 11, /* the length of the subject string */
1222 0, /* start at offset 0 in the subject */
1223 0, /* default options */
1224 ovector, /* vector of integers for substring information */
1225 30); /* number of elements (NOT size in bytes) */
1226 .
1227 .\" HTML <a name="extradata"></a>
1228 .SS "Extra data for \fBpcre_exec()\fR"
1229 .rs
1230 .sp
1231 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1232 data block. The \fBpcre_study()\fP function returns such a block (when it
1233 doesn't return NULL), but you can also create one for yourself, and pass
1234 additional information in it. The \fBpcre_extra\fP block contains the following
1235 fields (not necessarily in this order):
1236 .sp
1237 unsigned long int \fIflags\fP;
1238 void *\fIstudy_data\fP;
1239 unsigned long int \fImatch_limit\fP;
1240 unsigned long int \fImatch_limit_recursion\fP;
1241 void *\fIcallout_data\fP;
1242 const unsigned char *\fItables\fP;
1243 unsigned char **\fImark\fP;
1244 .sp
1245 The \fIflags\fP field is a bitmap that specifies which of the other fields
1246 are set. The flag bits are:
1247 .sp
1254 .sp
1255 Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1256 \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1257 the appropriate flag bit. You should not set this yourself, but you may add to
1258 the block by setting the other fields and their corresponding flag bits.
1259 .P
1260 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1261 vast amount of resources when running patterns that are not going to match,
1262 but which have a very large number of possibilities in their search trees. The
1263 classic example is a pattern that uses nested unlimited repeats.
1264 .P
1265 Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1266 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1267 number of times this function is called during a match, which has the effect of
1268 limiting the amount of backtracking that can take place. For patterns that are
1269 not anchored, the count restarts from zero for each position in the subject
1270 string.
1271 .P
1272 The default value for the limit can be set when PCRE is built; the default
1273 default is 10 million, which handles all but the most extreme cases. You can
1274 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1275 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1276 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1278 .P
1279 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1280 instead of limiting the total number of times that \fBmatch()\fP is called, it
1281 limits the depth of recursion. The recursion depth is a smaller number than the
1282 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1283 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1284 .P
1285 Limiting the recursion depth limits the amount of stack that can be used, or,
1286 when PCRE has been compiled to use memory on the heap instead of the stack, the
1287 amount of heap memory that can be used.
1288 .P
1289 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1290 built; the default default is the same value as the default for
1291 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1292 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1293 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1294 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1295 .P
1296 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1297 and is described in the
1298 .\" HREF
1299 \fBpcrecallout\fP
1300 .\"
1301 documentation.
1302 .P
1303 The \fItables\fP field is used to pass a character tables pointer to
1304 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1305 pattern. A non-NULL value is stored with the compiled pattern only if custom
1306 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1307 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1308 internal tables to be used. This facility is helpful when re-using patterns
1309 that have been saved after compiling with an external set of tables, because
1310 the external tables might be at a different address when \fBpcre_exec()\fP is
1311 called. See the
1312 .\" HREF
1313 \fBpcreprecompile\fP
1314 .\"
1315 documentation for a discussion of saving compiled patterns for later use.
1316 .P
1317 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1318 be set to point to a \fBchar *\fP variable. If the pattern contains any
1319 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1320 a name to pass back, a pointer to the name string (zero terminated) is placed
1321 in the variable pointed to by the \fImark\fP field. The names are within the
1322 compiled pattern; if you wish to retain such a name you must copy it before
1323 freeing the memory of a compiled pattern. If there is no name to pass back, the
1324 variable pointed to by the \fImark\fP field set to NULL. For details of the
1325 backtracking control verbs, see the section entitled
1326 .\" HTML <a href="pcrepattern#backtrackcontrol">
1327 .\" </a>
1328 "Backtracking control"
1329 .\"
1330 in the
1331 .\" HREF
1332 \fBpcrepattern\fP
1333 .\"
1334 documentation.
1335 .
1336 .
1337 .\" HTML <a name="execoptions"></a>
1338 .SS "Option bits for \fBpcre_exec()\fP"
1339 .rs
1340 .sp
1341 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1342 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1346 .sp
1348 .sp
1349 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1350 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1351 to be anchored by virtue of its contents, it cannot be made unachored at
1352 matching time.
1353 .sp
1356 .sp
1357 These options (which are mutually exclusive) control what the \eR escape
1358 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1359 match any Unicode newline sequence. These options override the choice that was
1360 made or defaulted when the pattern was compiled.
1361 .sp
1367 .sp
1368 These options override the newline definition that was chosen or defaulted when
1369 the pattern was compiled. For details, see the description of
1370 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1371 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1372 the way the match position is advanced after a match failure for an unanchored
1373 pattern.
1374 .P
1376 match attempt for an unanchored pattern fails when the current position is at a
1377 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1378 characters, the match position is advanced by two characters instead of one, in
1379 other words, to after the CRLF.
1380 .P
1381 The above rule is a compromise that makes the most common cases work as
1382 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1383 set), it does not match the string "\er\enA" because, after failing at the
1384 start, it skips both the CR and the LF before retrying. However, the pattern
1385 [\er\en]A does match that string, because it contains an explicit CR or LF
1386 reference, and so advances only by one character after the first failure.
1387 .P
1388 An explicit match for CR of LF is either a literal appearance of one of those
1389 characters, or one of the \er or \en escape sequences. Implicit matches such as
1390 [^X] do not count, nor does \es (which includes CR and LF in the characters
1391 that it matches).
1392 .P
1393 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1394 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1395 .sp
1397 .sp
1398 This option specifies that first character of the subject string is not the
1399 beginning of a line, so the circumflex metacharacter should not match before
1400 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1401 never to match. This option affects only the behaviour of the circumflex
1402 metacharacter. It does not affect \eA.
1403 .sp
1405 .sp
1406 This option specifies that the end of the subject string is not the end of a
1407 line, so the dollar metacharacter should not match it nor (except in multiline
1408 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1409 compile time) causes dollar never to match. This option affects only the
1410 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1411 .sp
1413 .sp
1414 An empty string is not considered to be a valid match if this option is set. If
1415 there are alternatives in the pattern, they are tried. If all the alternatives
1416 match the empty string, the entire match fails. For example, if the pattern
1417 .sp
1418 a?b?
1419 .sp
1420 is applied to a string not beginning with "a" or "b", it matches an empty
1421 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1422 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1423 .sp
1425 .sp
1426 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1427 the start of the subject is permitted. If the pattern is anchored, such a match
1428 can occur only if the pattern contains \eK.
1429 .P
1430 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1431 does make a special case of a pattern match of the empty string within its
1432 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1433 emulate Perl's behaviour after matching a null string by first trying the match
1434 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1435 if that fails, by advancing the starting offset (see below) and trying an
1436 ordinary match again. There is some code that demonstrates how to do this in
1437 the
1438 .\" HREF
1439 \fBpcredemo\fP
1440 .\"
1441 sample program.
1442 .sp
1444 .sp
1445 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1446 a match, in order to speed up the process. For example, if it is known that a
1447 match must start with a specific character, it searches the subject for that
1448 character, and fails immediately if it cannot find it, without actually running
1449 the main matching function. When callouts are in use, these optimizations can
1450 cause them to be skipped. This option disables the "start-up" optimizations,
1451 causing performance to suffer, but ensuring that the callouts do occur.
1452 .sp
1454 .sp
1455 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1456 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1457 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1458 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1459 strings in the
1460 .\" HTML <a href="pcre.html#utf8strings">
1461 .\" </a>
1462 section on UTF-8 support
1463 .\"
1464 in the main
1465 .\" HREF
1466 \fBpcre\fP
1467 .\"
1468 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1469 the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1470 PCRE_ERROR_BADUTF8_OFFSET is returned.
1471 .P
1472 If you already know that your subject is valid, and you want to skip these
1473 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1474 calling \fBpcre_exec()\fP. You might want to do this for the second and
1475 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1476 all the matches in a single subject string. However, you should be sure that
1477 the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1478 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1479 subject, or a value of \fIstartoffset\fP that does not point to the start of a
1480 UTF-8 character, is undefined. Your program may crash.
1481 .sp
1484 .sp
1485 These options turn on the partial matching feature. For backwards
1486 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1487 occurs if the end of the subject string is reached successfully, but there are
1488 not enough subject characters to complete the match. If this happens when
1489 PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns
1490 PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues
1491 by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL
1492 returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that
1493 was inspected when the partial match was found is set as the first matching
1494 string. There is a more detailed discussion in the
1495 .\" HREF
1496 \fBpcrepartial\fP
1497 .\"
1498 documentation.
1499 .
1500 .SS "The string to be matched by \fBpcre_exec()\fP"
1501 .rs
1502 .sp
1503 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1504 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1505 in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1506 a UTF-8 character. Unlike the pattern string, the subject may contain binary
1507 zero bytes. When the starting offset is zero, the search for a match starts at
1508 the beginning of the subject, and this is by far the most common case.
1509 .P
1510 A non-zero starting offset is useful when searching for another match in the
1511 same subject by calling \fBpcre_exec()\fP again after a previous success.
1512 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1513 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1514 lookbehind. For example, consider the pattern
1515 .sp
1516 \eBiss\eB
1517 .sp
1518 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1519 the current position in the subject is not a word boundary.) When applied to
1520 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1521 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1522 subject, namely "issipi", it does not match, because \eB is always false at the
1523 start of the subject, which is deemed to be a word boundary. However, if
1524 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1525 set to 4, it finds the second occurrence of "iss" because it is able to look
1526 behind the starting point to discover that it is preceded by a letter.
1527 .P
1528 If a non-zero starting offset is passed when the pattern is anchored, one
1529 attempt to match at the given offset is made. This can only succeed if the
1530 pattern does not require the match to be at the start of the subject.
1531 .
1532 .SS "How \fBpcre_exec()\fP returns captured substrings"
1533 .rs
1534 .sp
1535 In general, a pattern matches a certain portion of the subject, and in
1536 addition, further substrings from the subject may be picked out by parts of the
1537 pattern. Following the usage in Jeffrey Friedl's book, this is called
1538 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1539 a fragment of a pattern that picks out a substring. PCRE supports several other
1540 kinds of parenthesized subpattern that do not cause substrings to be captured.
1541 .P
1542 Captured substrings are returned to the caller via a vector of integers whose
1543 address is passed in \fIovector\fP. The number of elements in the vector is
1544 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1545 argument is NOT the size of \fIovector\fP in bytes.
1546 .P
1547 The first two-thirds of the vector is used to pass back captured substrings,
1548 each substring using a pair of integers. The remaining third of the vector is
1549 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1550 and is not available for passing back information. The number passed in
1551 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1552 rounded down.
1553 .P
1554 When a match is successful, information about captured substrings is returned
1555 in pairs of integers, starting at the beginning of \fIovector\fP, and
1556 continuing up to two-thirds of its length at the most. The first element of
1557 each pair is set to the byte offset of the first character in a substring, and
1558 the second is set to the byte offset of the first character after the end of a
1559 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1560 mode. They are not character counts.
1561 .P
1562 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1563 portion of the subject string matched by the entire pattern. The next pair is
1564 used for the first capturing subpattern, and so on. The value returned by
1565 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1566 For example, if two substrings have been captured, the returned value is 3. If
1567 there are no capturing subpatterns, the return value from a successful match is
1568 1, indicating that just the first pair of offsets has been set.
1569 .P
1570 If a capturing subpattern is matched repeatedly, it is the last portion of the
1571 string that it matched that is returned.
1572 .P
1573 If the vector is too small to hold all the captured substring offsets, it is
1574 used as far as possible (up to two-thirds of its length), and the function
1575 returns a value of zero. If the substring offsets are not of interest,
1576 \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1577 \fIovecsize\fP as zero. However, if the pattern contains back references and
1578 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1579 has to get additional memory for use during matching. Thus it is usually
1580 advisable to supply an \fIovector\fP.
1581 .P
1582 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1583 subpatterns there are in a compiled pattern. The smallest size for
1584 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1585 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1586 .P
1587 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1588 the subject when subpattern \fIn\fP has not been used at all. For example, if
1589 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1590 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1591 happens, both values in the offset pairs corresponding to unused subpatterns
1592 are set to -1.
1593 .P
1594 Offset values that correspond to unused subpatterns at the end of the
1595 expression are also set to -1. For example, if the string "abc" is matched
1596 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1597 return from the function is 2, because the highest used capturing subpattern
1598 number is 1. However, you can refer to the offsets for the second and third
1599 capturing subpatterns if you wish (assuming the vector is large enough, of
1600 course).
1601 .P
1602 Some convenience functions are provided for extracting the captured substrings
1603 as separate strings. These are described below.
1604 .
1605 .\" HTML <a name="errorlist"></a>
1606 .SS "Error return values from \fBpcre_exec()\fP"
1607 .rs
1608 .sp
1609 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1610 defined in the header file:
1611 .sp
1613 .sp
1614 The subject string did not match the pattern.
1615 .sp
1617 .sp
1618 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1619 NULL and \fIovecsize\fP was not zero.
1620 .sp
1622 .sp
1623 An unrecognized bit was set in the \fIoptions\fP argument.
1624 .sp
1626 .sp
1627 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1628 the case when it is passed a junk pointer and to detect when a pattern that was
1629 compiled in an environment of one endianness is run in an environment with the
1630 other endianness. This is the error that PCRE gives when the magic number is
1631 not present.
1632 .sp
1634 .sp
1635 While running the pattern match, an unknown item was encountered in the
1636 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1637 of the compiled pattern.
1638 .sp
1640 .sp
1641 If a pattern contains back references, but the \fIovector\fP that is passed to
1642 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1643 gets a block of memory at the start of matching to use for this purpose. If the
1644 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1645 automatically freed at the end of matching.
1646 .P
1647 This error is also given if \fBpcre_stack_malloc()\fP fails in
1648 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1649 \fB--disable-stack-for-recursion\fP.
1650 .sp
1652 .sp
1653 This error is used by the \fBpcre_copy_substring()\fP,
1654 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1655 below). It is never returned by \fBpcre_exec()\fP.
1656 .sp
1658 .sp
1659 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1660 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1661 above.
1662 .sp
1664 .sp
1665 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1666 use by callout functions that want to yield a distinctive error code. See the
1667 .\" HREF
1668 \fBpcrecallout\fP
1669 .\"
1670 documentation for details.
1671 .sp
1673 .sp
1674 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1675 .sp
1677 .sp
1678 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1679 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1680 .sp
1682 .sp
1683 The subject string did not match, but it did match partially. See the
1684 .\" HREF
1685 \fBpcrepartial\fP
1686 .\"
1687 documentation for details of partial matching.
1688 .sp
1690 .sp
1691 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1692 option was used with a compiled pattern containing items that were not
1693 supported for partial matching. From release 8.00 onwards, there are no
1694 restrictions on partial matching.
1695 .sp
1697 .sp
1698 An unexpected internal error has occurred. This error could be caused by a bug
1699 in PCRE or by overwriting of the compiled pattern.
1700 .sp
1702 .sp
1703 This error is given if the value of the \fIovecsize\fP argument is negative.
1704 .sp
1706 .sp
1707 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1708 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1709 description above.
1710 .sp
1712 .sp
1713 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1714 .P
1715 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1716 .
1717 .
1719 .rs
1720 .sp
1721 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1722 .ti +5n
1723 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1724 .ti +5n
1725 .B int \fIbuffersize\fP);
1726 .PP
1727 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1728 .ti +5n
1729 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1730 .ti +5n
1731 .B const char **\fIstringptr\fP);
1732 .PP
1733 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1734 .ti +5n
1735 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1736 .PP
1737 Captured substrings can be accessed directly by using the offsets returned by
1738 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1739 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1740 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1741 as new, separate, zero-terminated strings. These functions identify substrings
1742 by number. The next section describes functions for extracting named
1743 substrings.
1744 .P
1745 A substring that contains a binary zero is correctly extracted and has a
1746 further zero added on the end, but the result is not, of course, a C string.
1747 However, you can process such a string by referring to the length that is
1748 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1749 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1750 for handling strings containing binary zeros, because the end of the final
1751 string is not independently indicated.
1752 .P
1753 The first three arguments are the same for all three of these functions:
1754 \fIsubject\fP is the subject string that has just been successfully matched,
1755 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1756 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1757 captured by the match, including the substring that matched the entire regular
1758 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1759 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1760 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1761 number of elements in the vector divided by three.
1762 .P
1763 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1764 extract a single substring, whose number is given as \fIstringnumber\fP. A
1765 value of zero extracts the substring that matched the entire pattern, whereas
1766 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1767 the string is placed in \fIbuffer\fP, whose length is given by
1768 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1769 obtained via \fBpcre_malloc\fP, and its address is returned via
1770 \fIstringptr\fP. The yield of the function is the length of the string, not
1771 including the terminating zero, or one of these error codes:
1772 .sp
1774 .sp
1775 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1776 memory failed for \fBpcre_get_substring()\fP.
1777 .sp
1779 .sp
1780 There is no substring whose number is \fIstringnumber\fP.
1781 .P
1782 The \fBpcre_get_substring_list()\fP function extracts all available substrings
1783 and builds a list of pointers to them. All this is done in a single block of
1784 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1785 is returned via \fIlistptr\fP, which is also the start of the list of string
1786 pointers. The end of the list is marked by a NULL pointer. The yield of the
1787 function is zero if all went well, or the error code
1788 .sp
1790 .sp
1791 if the attempt to get the memory block failed.
1792 .P
1793 When any of these functions encounter a substring that is unset, which can
1794 happen when capturing subpattern number \fIn+1\fP matches some part of the
1795 subject, but subpattern \fIn\fP has not been used at all, they return an empty
1796 string. This can be distinguished from a genuine zero-length substring by
1797 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1798 substrings.
1799 .P
1800 The two convenience functions \fBpcre_free_substring()\fP and
1801 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1802 a previous call of \fBpcre_get_substring()\fP or
1803 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1804 the function pointed to by \fBpcre_free\fP, which of course could be called
1805 directly from a C program. However, PCRE is used in some situations where it is
1806 linked via a special interface to another programming language that cannot use
1807 \fBpcre_free\fP directly; it is for these cases that the functions are
1808 provided.
1809 .
1810 .
1812 .rs
1813 .sp
1814 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1815 .ti +5n
1816 .B const char *\fIname\fP);
1817 .PP
1818 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1819 .ti +5n
1820 .B const char *\fIsubject\fP, int *\fIovector\fP,
1821 .ti +5n
1822 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1823 .ti +5n
1824 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1825 .PP
1826 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1827 .ti +5n
1828 .B const char *\fIsubject\fP, int *\fIovector\fP,
1829 .ti +5n
1830 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1831 .ti +5n
1832 .B const char **\fIstringptr\fP);
1833 .PP
1834 To extract a substring by name, you first have to find associated number.
1835 For example, for this pattern
1836 .sp
1837 (a+)b(?<xxx>\ed+)...
1838 .sp
1839 the number of the subpattern called "xxx" is 2. If the name is known to be
1840 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1841 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1842 pattern, and the second is the name. The yield of the function is the
1843 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1844 that name.
1845 .P
1846 Given the number, you can extract the substring directly, or use one of the
1847 functions described in the previous section. For convenience, there are also
1848 two functions that do the whole job.
1849 .P
1850 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1851 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1852 functions that extract by number. As these are described in the previous
1853 section, they are not re-described here. There are just two differences:
1854 .P
1855 First, instead of a substring number, a substring name is given. Second, there
1856 is an extra argument, given at the start, which is a pointer to the compiled
1857 pattern. This is needed in order to gain access to the name-to-number
1858 translation table.
1859 .P
1860 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1861 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1862 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1863 the behaviour may not be what you want (see the next section).
1864 .P
1865 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
1866 subpatterns with the same number, as described in the
1867 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1868 .\" </a>
1869 section on duplicate subpattern numbers
1870 .\"
1871 in the
1872 .\" HREF
1873 \fBpcrepattern\fP
1874 .\"
1875 page, you cannot use names to distinguish the different subpatterns, because
1876 names are not included in the compiled code. The matching process uses only
1877 numbers. For this reason, the use of different names for subpatterns of the
1878 same number causes an error at compile time.
1879 .
1881 .rs
1882 .sp
1883 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1884 .ti +5n
1885 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1886 .PP
1887 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1888 are not required to be unique. (Duplicate names are always allowed for
1889 subpatterns with the same number, created by using the (?| feature. Indeed, if
1890 such subpatterns are named, they are required to use the same names.)
1891 .P
1892 Normally, patterns with duplicate names are such that in any one match, only
1893 one of the named subpatterns participates. An example is shown in the
1894 .\" HREF
1895 \fBpcrepattern\fP
1896 .\"
1897 documentation.
1898 .P
1899 When duplicates are present, \fBpcre_copy_named_substring()\fP and
1900 \fBpcre_get_named_substring()\fP return the first substring corresponding to
1901 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1902 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1903 returns one of the numbers that are associated with the name, but it is not
1904 defined which it is.
1905 .P
1906 If you want to get full details of all captured substrings for a given name,
1907 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1908 argument is the compiled pattern, and the second is the name. The third and
1909 fourth are pointers to variables which are updated by the function. After it
1910 has run, they point to the first and last entries in the name-to-number table
1911 for the given name. The function itself returns the length of each entry, or
1912 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1913 described above in the section entitled \fIInformation about a pattern\fP.
1914 Given all the relevant entries for the name, you can extract each of their
1915 numbers, and hence the captured data, if any.
1916 .
1917 .
1919 .rs
1920 .sp
1921 The traditional matching function uses a similar algorithm to Perl, which stops
1922 when it finds the first match, starting at a given point in the subject. If you
1923 want to find all possible matches, or the longest possible match, consider
1924 using the alternative matching function (see below) instead. If you cannot use
1925 the alternative function, but still need to find all possible matches, you
1926 can kludge it up by making use of the callout facility, which is described in
1927 the
1928 .\" HREF
1929 \fBpcrecallout\fP
1930 .\"
1931 documentation.
1932 .P
1933 What you have to do is to insert a callout right at the end of the pattern.
1934 When your callout function is called, extract and save the current matched
1935 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1936 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1937 will yield PCRE_ERROR_NOMATCH.
1938 .
1939 .
1940 .\" HTML <a name="dfamatch"></a>
1942 .rs
1943 .sp
1944 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1945 .ti +5n
1946 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1947 .ti +5n
1948 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1949 .ti +5n
1950 .B int *\fIworkspace\fP, int \fIwscount\fP);
1951 .P
1952 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1953 a compiled pattern, using a matching algorithm that scans the subject string
1954 just once, and does not backtrack. This has different characteristics to the
1955 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1956 patterns are not supported. Nevertheless, there are times when this kind of
1957 matching can be useful. For a discussion of the two matching algorithms, and a
1958 list of features that \fBpcre_dfa_exec()\fP does not support, see the
1959 .\" HREF
1960 \fBpcrematching\fP
1961 .\"
1962 documentation.
1963 .P
1964 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1965 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1966 different way, and this is described below. The other common arguments are used
1967 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1968 here.
1969 .P
1970 The two additional arguments provide workspace for the function. The workspace
1971 vector should contain at least 20 elements. It is used for keeping track of
1972 multiple paths through the pattern tree. More workspace will be needed for
1973 patterns and subjects where there are a lot of potential matches.
1974 .P
1975 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1976 .sp
1977 int rc;
1978 int ovector[10];
1979 int wspace[20];
1980 rc = pcre_dfa_exec(
1981 re, /* result of pcre_compile() */
1982 NULL, /* we didn't study the pattern */
1983 "some string", /* the subject string */
1984 11, /* the length of the subject string */
1985 0, /* start at offset 0 in the subject */
1986 0, /* default options */
1987 ovector, /* vector of integers for substring information */
1988 10, /* number of elements (NOT size in bytes) */
1989 wspace, /* working space vector */
1990 20); /* number of elements (NOT size in bytes) */
1991 .
1992 .SS "Option bits for \fBpcre_dfa_exec()\fP"
1993 .rs
1994 .sp
1995 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1996 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1999 and PCRE_DFA_RESTART. All but the last four of these are exactly the same as
2000 for \fBpcre_exec()\fP, so their description is not repeated here.
2001 .sp
2004 .sp
2005 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2006 details are slightly different. When PCRE_PARTIAL_HARD is set for
2007 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2008 is reached and there is still at least one matching possibility that requires
2009 additional characters. This happens even if some complete matches have also
2010 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2011 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2012 there have been no complete matches, but there is still at least one matching
2013 possibility. The portion of the string that was inspected when the longest
2014 partial match was found is set as the first matching string in both cases.
2015 .sp
2017 .sp
2018 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2019 soon as it has found one match. Because of the way the alternative algorithm
2020 works, this is necessarily the shortest possible match at the first possible
2021 matching point in the subject string.
2022 .sp
2024 .sp
2025 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2026 again, with additional subject characters, and have it continue with the same
2027 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2028 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2029 before because data about the match so far is left in them after a partial
2030 match. There is more discussion of this facility in the
2031 .\" HREF
2032 \fBpcrepartial\fP
2033 .\"
2034 documentation.
2035 .
2036 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2037 .rs
2038 .sp
2039 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2040 substring in the subject. Note, however, that all the matches from one run of
2041 the function start at the same point in the subject. The shorter matches are
2042 all initial substrings of the longer matches. For example, if the pattern
2043 .sp
2044 <.*>
2045 .sp
2046 is matched against the string
2047 .sp
2048 This is <something> <something else> <something further> no more
2049 .sp
2050 the three matched strings are
2051 .sp
2052 <something>
2053 <something> <something else>
2054 <something> <something else> <something further>
2055 .sp
2056 On success, the yield of the function is a number greater than zero, which is
2057 the number of matched substrings. The substrings themselves are returned in
2058 \fIovector\fP. Each string uses two elements; the first is the offset to the
2059 start, and the second is the offset to the end. In fact, all the strings have
2060 the same start offset. (Space could have been saved by giving this only once,
2061 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2062 returns data, even though the meaning of the strings is different.)
2063 .P
2064 The strings are returned in reverse order of length; that is, the longest
2065 matching string is given first. If there were too many matches to fit into
2066 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2067 the longest matches.
2068 .
2069 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2070 .rs
2071 .sp
2072 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2073 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2074 described
2075 .\" HTML <a href="#errorlist">
2076 .\" </a>
2077 above.
2078 .\"
2079 There are in addition the following errors that are specific to
2080 \fBpcre_dfa_exec()\fP:
2081 .sp
2083 .sp
2084 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2085 that it does not support, for instance, the use of \eC or a back reference.
2086 .sp
2088 .sp
2089 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2090 uses a back reference for the condition, or a test for recursion in a specific
2091 group. These are not supported.
2092 .sp
2094 .sp
2095 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2096 block that contains a setting of the \fImatch_limit\fP field. This is not
2097 supported (it is meaningless).
2098 .sp
2100 .sp
2101 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2102 \fIworkspace\fP vector.
2103 .sp
2105 .sp
2106 When a recursive subpattern is processed, the matching function calls itself
2107 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2108 error is given if the output vector is not large enough. This should be
2109 extremely rare, as a vector of size 1000 is used.
2110 .
2111 .
2112 .SH "SEE ALSO"
2113 .rs
2114 .sp
2115 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2116 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2117 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2118 .
2119 .
2121 .rs
2122 .sp
2123 .nf
2124 Philip Hazel
2125 University Computing Service
2126 Cambridge CB2 3QH, England.
2127 .fi
2128 .
2129 .
2131 .rs
2132 .sp
2133 .nf
2134 Last updated: 01 June 2010
2135 Copyright (c) 1997-2010 University of Cambridge.
2136 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5