ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 859 - (show annotations)
Mon Jan 9 17:43:54 2012 UTC (8 years, 2 months ago) by ph10
File size: 113412 byte(s)
3 PCRE - Perl-compatible regular expressions
4 .sp
5 .B #include <pcre.h>
6 .
7 .
9 .rs
10 .sp
11 .SM
12 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
13 .ti +5n
14 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
15 .ti +5n
16 .B const unsigned char *\fItableptr\fP);
17 .PP
18 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19 .ti +5n
20 .B int *\fIerrorcodeptr\fP,
21 .ti +5n
22 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
23 .ti +5n
24 .B const unsigned char *\fItableptr\fP);
25 .PP
26 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
27 .ti +5n
28 .B const char **\fIerrptr\fP);
29 .PP
30 .B void pcre_free_study(pcre_extra *\fIextra\fP);
31 .PP
32 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
33 .ti +5n
34 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
35 .ti +5n
36 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37 .PP
38 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
39 .ti +5n
40 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
41 .ti +5n
42 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
43 .ti +5n
44 .B int *\fIworkspace\fP, int \fIwscount\fP);
45 .
46 .
48 .rs
49 .sp
50 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
51 .ti +5n
52 .B const char *\fIsubject\fP, int *\fIovector\fP,
53 .ti +5n
54 .B int \fIstringcount\fP, const char *\fIstringname\fP,
55 .ti +5n
56 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
57 .PP
58 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
59 .ti +5n
60 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
61 .ti +5n
62 .B int \fIbuffersize\fP);
63 .PP
64 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
65 .ti +5n
66 .B const char *\fIsubject\fP, int *\fIovector\fP,
67 .ti +5n
68 .B int \fIstringcount\fP, const char *\fIstringname\fP,
69 .ti +5n
70 .B const char **\fIstringptr\fP);
71 .PP
72 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
73 .ti +5n
74 .B const char *\fIname\fP);
75 .PP
76 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
77 .ti +5n
78 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
79 .PP
80 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
81 .ti +5n
82 .B int \fIstringcount\fP, int \fIstringnumber\fP,
83 .ti +5n
84 .B const char **\fIstringptr\fP);
85 .PP
86 .B int pcre_get_substring_list(const char *\fIsubject\fP,
87 .ti +5n
88 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
89 .PP
90 .B void pcre_free_substring(const char *\fIstringptr\fP);
91 .PP
92 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
93 .
94 .
96 .rs
97 .sp
98 .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
99 .PP
100 .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
101 .PP
102 .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
103 .ti +5n
104 .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
105 .PP
106 .B const unsigned char *pcre_maketables(void);
107 .PP
108 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
109 .ti +5n
110 .B int \fIwhat\fP, void *\fIwhere\fP);
111 .PP
112 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
113 .PP
114 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
115 .PP
116 .B const char *pcre_version(void);
117 .PP
118 .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
119 .ti +5n
120 .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
121 .
122 .
124 .rs
125 .sp
126 .B void *(*pcre_malloc)(size_t);
127 .PP
128 .B void (*pcre_free)(void *);
129 .PP
130 .B void *(*pcre_stack_malloc)(size_t);
131 .PP
132 .B void (*pcre_stack_free)(void *);
133 .PP
134 .B int (*pcre_callout)(pcre_callout_block *);
135 .
136 .
138 .rs
139 .sp
140 From release 8.30, PCRE can be compiled as a library for handling 16-bit
141 character strings as well as, or instead of, the original library that handles
142 8-bit character strings. To avoid too much complication, this document
143 describes the 8-bit versions of the functions, with only occasional references
144 to the 16-bit library.
145 .P
146 The 16-bit functions operate in the same way as their 8-bit counterparts; they
147 just use different data types for their arguments and results, and their names
148 start with \fBpcre16_\fP instead of \fBpcre_\fP. For every option that has UTF8
149 in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
150 UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
151 option names define the same bit values.
152 .P
153 References to bytes and UTF-8 in this document should be read as references to
154 16-bit data quantities and UTF-16 when using the 16-bit library, unless
155 specified otherwise. More details of the specific differences for the 16-bit
156 library are given in the
157 .\" HREF
158 \fBpcre16\fP
159 .\"
160 page.
161 .
162 .
164 .rs
165 .sp
166 PCRE has its own native API, which is described in this document. There are
167 also some wrapper functions (for the 8-bit library only) that correspond to the
168 POSIX regular expression API, but they do not give access to all the
169 functionality. They are described in the
170 .\" HREF
171 \fBpcreposix\fP
172 .\"
173 documentation. Both of these APIs define a set of C function calls. A C++
174 wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
175 documented in the
176 .\" HREF
177 \fBpcrecpp\fP
178 .\"
179 page.
180 .P
181 The native API C function prototypes are defined in the header file
182 \fBpcre.h\fP, and on Unix-like systems the (8-bit) library itself is called
183 \fBlibpcre\fP. It can normally be accessed by adding \fB-lpcre\fP to the
184 command for linking an application that uses PCRE. The header file defines the
185 macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
186 for the library. Applications can use these to include support for different
187 releases of PCRE.
188 .P
189 In a Windows environment, if you want to statically link an application program
190 against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
191 including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
192 \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
193 \fB__declspec(dllimport)\fP, with unwanted results.
194 .P
195 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
196 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
197 in a Perl-compatible manner. A sample program that demonstrates the simplest
198 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
199 source distribution. A listing of this program is given in the
200 .\" HREF
201 \fBpcredemo\fP
202 .\"
203 documentation, and the
204 .\" HREF
205 \fBpcresample\fP
206 .\"
207 documentation describes how to compile and run it.
208 .P
209 Just-in-time compiler support is an optional feature of PCRE that can be built
210 in appropriate hardware environments. It greatly speeds up the matching
211 performance of many patterns. Simple programs can easily request that it be
212 used if available, by setting an option that is ignored when it is not
213 relevant. More complicated programs might need to make use of the functions
214 \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
215 \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
216 These functions are discussed in the
217 .\" HREF
218 \fBpcrejit\fP
219 .\"
220 documentation.
221 .P
222 A second matching function, \fBpcre_dfa_exec()\fP, which is not
223 Perl-compatible, is also provided. This uses a different algorithm for the
224 matching. The alternative algorithm finds all possible matches (at a given
225 point in the subject), and scans the subject just once (unless there are
226 lookbehind assertions). However, this algorithm does not return captured
227 substrings. A description of the two matching algorithms and their advantages
228 and disadvantages is given in the
229 .\" HREF
230 \fBpcrematching\fP
231 .\"
232 documentation.
233 .P
234 In addition to the main compiling and matching functions, there are convenience
235 functions for extracting captured substrings from a subject string that is
236 matched by \fBpcre_exec()\fP. They are:
237 .sp
238 \fBpcre_copy_substring()\fP
239 \fBpcre_copy_named_substring()\fP
240 \fBpcre_get_substring()\fP
241 \fBpcre_get_named_substring()\fP
242 \fBpcre_get_substring_list()\fP
243 \fBpcre_get_stringnumber()\fP
244 \fBpcre_get_stringtable_entries()\fP
245 .sp
246 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
247 provided, to free the memory used for extracted strings.
248 .P
249 The function \fBpcre_maketables()\fP is used to build a set of character tables
250 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
251 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
252 specialist use. Most commonly, no special tables are passed, in which case
253 internal tables that are generated when PCRE is built are used.
254 .P
255 The function \fBpcre_fullinfo()\fP is used to find out information about a
256 compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
257 string containing the version of PCRE and its date of release.
258 .P
259 The function \fBpcre_refcount()\fP maintains a reference count in a data block
260 containing a compiled pattern. This is provided for the benefit of
261 object-oriented applications.
262 .P
263 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
264 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
265 respectively. PCRE calls the memory management functions via these variables,
266 so a calling program can replace them if it wishes to intercept the calls. This
267 should be done before calling any PCRE functions.
268 .P
269 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
270 indirections to memory management functions. These special functions are used
271 only when PCRE is compiled to use the heap for remembering data, instead of
272 recursive function calls, when running the \fBpcre_exec()\fP function. See the
273 .\" HREF
274 \fBpcrebuild\fP
275 .\"
276 documentation for details of how to do this. It is a non-standard way of
277 building PCRE, for use in environments that have limited stacks. Because of the
278 greater use of memory management, it runs more slowly. Separate functions are
279 provided so that special-purpose external code can be used for this case. When
280 used, these functions are always called in a stack-like manner (last obtained,
281 first freed), and always for memory blocks of the same size. There is a
282 discussion about PCRE's stack usage in the
283 .\" HREF
284 \fBpcrestack\fP
285 .\"
286 documentation.
287 .P
288 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
289 by the caller to a "callout" function, which PCRE will then call at specified
290 points during a matching operation. Details are given in the
291 .\" HREF
292 \fBpcrecallout\fP
293 .\"
294 documentation.
295 .
296 .
297 .\" HTML <a name="newlines"></a>
299 .rs
300 .sp
301 PCRE supports five different conventions for indicating line breaks in
302 strings: a single CR (carriage return) character, a single LF (linefeed)
303 character, the two-character sequence CRLF, any of the three preceding, or any
304 Unicode newline sequence. The Unicode newline sequences are the three just
305 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
306 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
307 (paragraph separator, U+2029).
308 .P
309 Each of the first three conventions is used by at least one operating system as
310 its standard newline sequence. When PCRE is built, a default can be specified.
311 The default default is LF, which is the Unix standard. When PCRE is run, the
312 default can be overridden, either when a pattern is compiled, or when it is
313 matched.
314 .P
315 At compile time, the newline convention can be specified by the \fIoptions\fP
316 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
317 start of the pattern itself; this overrides any other settings. See the
318 .\" HREF
319 \fBpcrepattern\fP
320 .\"
321 page for details of the special character sequences.
322 .P
323 In the PCRE documentation the word "newline" is used to mean "the character or
324 pair of characters that indicate a line break". The choice of newline
325 convention affects the handling of the dot, circumflex, and dollar
326 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
327 recognized line ending sequence, the match position advancement for a
328 non-anchored pattern. There is more detail about this in the
329 .\" HTML <a href="#execoptions">
330 .\" </a>
331 section on \fBpcre_exec()\fP options
332 .\"
333 below.
334 .P
335 The choice of newline convention does not affect the interpretation of
336 the \en or \er escape sequences, nor does it affect what \eR matches, which is
337 controlled in a similar way, but by separate options.
338 .
339 .
341 .rs
342 .sp
343 The PCRE functions can be used in multi-threading applications, with the
344 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
345 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
346 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
347 .P
348 The compiled form of a regular expression is not altered during matching, so
349 the same compiled pattern can safely be used by several threads at once.
350 .P
351 If the just-in-time optimization feature is being used, it needs separate
352 memory stack areas for each thread. See the
353 .\" HREF
354 \fBpcrejit\fP
355 .\"
356 documentation for more details.
357 .
358 .
360 .rs
361 .sp
362 The compiled form of a regular expression can be saved and re-used at a later
363 time, possibly by a different program, and even on a host other than the one on
364 which it was compiled. Details are given in the
365 .\" HREF
366 \fBpcreprecompile\fP
367 .\"
368 documentation, which includes a description of the
369 \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
370 expression with one version of PCRE for use with a different version is not
371 guaranteed to work and may cause crashes.
372 .
373 .
375 .rs
376 .sp
377 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
378 .PP
379 The function \fBpcre_config()\fP makes it possible for a PCRE client to
380 discover which optional features have been compiled into the PCRE library. The
381 .\" HREF
382 \fBpcrebuild\fP
383 .\"
384 documentation has more details about these optional features.
385 .P
386 The first argument for \fBpcre_config()\fP is an integer, specifying which
387 information is required; the second argument is a pointer to a variable into
388 which the information is placed. The returned value is zero on success, or the
389 negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
390 not recognized. The following information is available:
391 .sp
393 .sp
394 The output is an integer that is set to one if UTF-8 support is available;
395 otherwise it is set to zero. If this option is given to the 16-bit version of
396 this function, \fBpcre16_config()\fP, the result is PCRE_ERROR_BADOPTION.
397 .sp
399 .sp
400 The output is an integer that is set to one if UTF-16 support is available;
401 otherwise it is set to zero. This value should normally be given to the 16-bit
402 version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
403 version of this function, the result is PCRE_ERROR_BADOPTION.
404 .sp
406 .sp
407 The output is an integer that is set to one if support for Unicode character
408 properties is available; otherwise it is set to zero.
409 .sp
411 .sp
412 The output is an integer that is set to one if support for just-in-time
413 compiling is available; otherwise it is set to zero.
414 .sp
416 .sp
417 The output is an integer whose value specifies the default character sequence
418 that is recognized as meaning "newline". The four values that are supported
419 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
420 Though they are derived from ASCII, the same values are returned in EBCDIC
421 environments. The default should normally correspond to the standard sequence
422 for your operating system.
423 .sp
425 .sp
426 The output is an integer whose value indicates what character sequences the \eR
427 escape sequence matches by default. A value of 0 means that \eR matches any
428 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
429 or CRLF. The default can be overridden when a pattern is compiled or matched.
430 .sp
432 .sp
433 The output is an integer that contains the number of bytes used for internal
434 linkage in compiled regular expressions. For the 8-bit library, the value can
435 be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
436 a number of bytes. The default value of 2 is sufficient for all but the most
437 massive patterns, since it allows the compiled pattern to be up to 64K in size.
438 Larger values allow larger regular expressions to be compiled, at the expense
439 of slower matching.
440 .sp
442 .sp
443 The output is an integer that contains the threshold above which the POSIX
444 interface uses \fBmalloc()\fP for output vectors. Further details are given in
445 the
446 .\" HREF
447 \fBpcreposix\fP
448 .\"
449 documentation.
450 .sp
452 .sp
453 The output is a long integer that gives the default limit for the number of
454 internal matching function calls in a \fBpcre_exec()\fP execution. Further
455 details are given with \fBpcre_exec()\fP below.
456 .sp
458 .sp
459 The output is a long integer that gives the default limit for the depth of
460 recursion when calling the internal matching function in a \fBpcre_exec()\fP
461 execution. Further details are given with \fBpcre_exec()\fP below.
462 .sp
464 .sp
465 The output is an integer that is set to one if internal recursion when running
466 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
467 to remember their state. This is the usual way that PCRE is compiled. The
468 output is zero if PCRE was compiled to use blocks of data on the heap instead
469 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
470 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
471 avoiding the use of the stack.
472 .
473 .
475 .rs
476 .sp
477 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
478 .ti +5n
479 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
480 .ti +5n
481 .B const unsigned char *\fItableptr\fP);
482 .sp
483 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
484 .ti +5n
485 .B int *\fIerrorcodeptr\fP,
486 .ti +5n
487 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
488 .ti +5n
489 .B const unsigned char *\fItableptr\fP);
490 .P
491 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
492 called to compile a pattern into an internal form. The only difference between
493 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
494 \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
495 too much repetition, we refer just to \fBpcre_compile()\fP below, but the
496 information applies equally to \fBpcre_compile2()\fP.
497 .P
498 The pattern is a C string terminated by a binary zero, and is passed in the
499 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
500 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
501 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
502 for a structure whose contents are not externally defined. It is up to the
503 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
504 .P
505 Although the compiled code of a PCRE regex is relocatable, that is, it does not
506 depend on memory location, the complete \fBpcre\fP data block is not
507 fully relocatable, because it may contain a copy of the \fItableptr\fP
508 argument, which is an address (see below).
509 .P
510 The \fIoptions\fP argument contains various bit settings that affect the
511 compilation. It should be zero if no options are required. The available
512 options are described below. Some of them (in particular, those that are
513 compatible with Perl, but some others as well) can also be set and unset from
514 within the pattern (see the detailed description in the
515 .\" HREF
516 \fBpcrepattern\fP
517 .\"
518 documentation). For those options that can be different in different parts of
519 the pattern, the contents of the \fIoptions\fP argument specifies their
520 settings at the start of compilation and execution. The PCRE_ANCHORED,
522 PCRE_NO_START_OPT options can be set at the time of matching as well as at
523 compile time.
524 .P
525 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
526 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
527 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
528 error message. This is a static string that is part of the library. You must
529 not try to free it. Normally, the offset from the start of the pattern to the
530 byte that was being processed when the error was discovered is placed in the
531 variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
532 immediate error is given). However, for an invalid UTF-8 string, the offset is
533 that of the first byte of the failing character.
534 .P
535 Some errors are not detected until the whole pattern has been scanned; in these
536 cases, the offset passed back is the length of the pattern. Note that the
537 offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
538 into the middle of a UTF-8 character.
539 .P
540 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
541 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
542 returned via this argument in the event of an error. This is in addition to the
543 textual error message. Error codes and messages are listed below.
544 .P
545 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
546 character tables that are built when PCRE is compiled, using the default C
547 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
548 call to \fBpcre_maketables()\fP. This value is stored with the compiled
549 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
550 passed to it. For more discussion, see the section on locale support below.
551 .P
552 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
553 .sp
554 pcre *re;
555 const char *error;
556 int erroffset;
557 re = pcre_compile(
558 "^A.*Z", /* the pattern */
559 0, /* default options */
560 &error, /* for error message */
561 &erroffset, /* for error offset */
562 NULL); /* use default character tables */
563 .sp
564 The following names for option bits are defined in the \fBpcre.h\fP header
565 file:
566 .sp
568 .sp
569 If this bit is set, the pattern is forced to be "anchored", that is, it is
570 constrained to match only at the first matching point in the string that is
571 being searched (the "subject string"). This effect can also be achieved by
572 appropriate constructs in the pattern itself, which is the only way to do it in
573 Perl.
574 .sp
576 .sp
577 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
578 all with number 255, before each pattern item. For discussion of the callout
579 facility, see the
580 .\" HREF
581 \fBpcrecallout\fP
582 .\"
583 documentation.
584 .sp
587 .sp
588 These options (which are mutually exclusive) control what the \eR escape
589 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
590 match any Unicode newline sequence. The default is specified when PCRE is
591 built. It can be overridden from within the pattern, or by setting an option
592 when a compiled pattern is matched.
593 .sp
595 .sp
596 If this bit is set, letters in the pattern match both upper and lower case
597 letters. It is equivalent to Perl's /i option, and it can be changed within a
598 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
599 concept of case for characters whose values are less than 128, so caseless
600 matching is always possible. For characters with higher values, the concept of
601 case is supported if PCRE is compiled with Unicode property support, but not
602 otherwise. If you want to use caseless matching for characters 128 and above,
603 you must ensure that PCRE is compiled with Unicode property support as well as
604 with UTF-8 support.
605 .sp
607 .sp
608 If this bit is set, a dollar metacharacter in the pattern matches only at the
609 end of the subject string. Without this option, a dollar also matches
610 immediately before a newline at the end of the string (but not before any other
611 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
612 There is no equivalent to this option in Perl, and no way to set it within a
613 pattern.
614 .sp
616 .sp
617 If this bit is set, a dot metacharacter in the pattern matches a character of
618 any value, including one that indicates a newline. However, it only ever
619 matches one character, even if newlines are coded as CRLF. Without this option,
620 a dot does not match when the current position is at a newline. This option is
621 equivalent to Perl's /s option, and it can be changed within a pattern by a
622 (?s) option setting. A negative class such as [^a] always matches newline
623 characters, independent of the setting of this option.
624 .sp
626 .sp
627 If this bit is set, names used to identify capturing subpatterns need not be
628 unique. This can be helpful for certain types of pattern when it is known that
629 only one instance of the named subpattern can ever be matched. There are more
630 details of named subpatterns below; see also the
631 .\" HREF
632 \fBpcrepattern\fP
633 .\"
634 documentation.
635 .sp
637 .sp
638 If this bit is set, whitespace data characters in the pattern are totally
639 ignored except when escaped or inside a character class. Whitespace does not
640 include the VT character (code 11). In addition, characters between an
641 unescaped # outside a character class and the next newline, inclusive, are also
642 ignored. This is equivalent to Perl's /x option, and it can be changed within a
643 pattern by a (?x) option setting.
644 .P
645 Which characters are interpreted as newlines is controlled by the options
646 passed to \fBpcre_compile()\fP or by a special sequence at the start of the
647 pattern, as described in the section entitled
648 .\" HTML <a href="pcrepattern.html#newlines">
649 .\" </a>
650 "Newline conventions"
651 .\"
652 in the \fBpcrepattern\fP documentation. Note that the end of this type of
653 comment is a literal newline sequence in the pattern; escape sequences that
654 happen to represent a newline do not count.
655 .P
656 This option makes it possible to include comments inside complicated patterns.
657 Note, however, that this applies only to data characters. Whitespace characters
658 may never appear within special character sequences in a pattern, for example
659 within the sequence (?( that introduces a conditional subpattern.
660 .sp
662 .sp
663 This option was invented in order to turn on additional functionality of PCRE
664 that is incompatible with Perl, but it is currently of very little use. When
665 set, any backslash in a pattern that is followed by a letter that has no
666 special meaning causes an error, thus reserving these combinations for future
667 expansion. By default, as in Perl, a backslash followed by a letter with no
668 special meaning is treated as a literal. (Perl can, however, be persuaded to
669 give an error for this, by running it with the -w option.) There are at present
670 no other features controlled by this option. It can also be set by a (?X)
671 option setting within a pattern.
672 .sp
674 .sp
675 If this option is set, an unanchored pattern is required to match before or at
676 the first newline in the subject string, though the matched text may continue
677 over the newline.
678 .sp
680 .sp
681 If this option is set, PCRE's behaviour is changed in some ways so that it is
682 compatible with JavaScript rather than Perl. The changes are as follows:
683 .P
684 (1) A lone closing square bracket in a pattern causes a compile-time error,
685 because this is illegal in JavaScript (by default it is treated as a data
686 character). Thus, the pattern AB]CD becomes illegal when this option is set.
687 .P
688 (2) At run time, a back reference to an unset subpattern group matches an empty
689 string (by default this causes the current matching alternative to fail). A
690 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
691 an "a" in the subject), whereas it fails by default, for Perl compatibility.
692 .P
693 (3) \eU matches an upper case "U" character; by default \eU causes a compile
694 time error (Perl uses \eU to upper case subsequent characters).
695 .P
696 (4) \eu matches a lower case "u" character unless it is followed by four
697 hexadecimal digits, in which case the hexadecimal number defines the code point
698 to match. By default, \eu causes a compile time error (Perl uses it to upper
699 case the following character).
700 .P
701 (5) \ex matches a lower case "x" character unless it is followed by two
702 hexadecimal digits, in which case the hexadecimal number defines the code point
703 to match. By default, as in Perl, a hexadecimal number is always expected after
704 \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
705 binary zero character followed by z).
706 .sp
708 .sp
709 By default, PCRE treats the subject string as consisting of a single line of
710 characters (even if it actually contains newlines). The "start of line"
711 metacharacter (^) matches only at the start of the string, while the "end of
712 line" metacharacter ($) matches only at the end of the string, or before a
713 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
714 Perl.
715 .P
716 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
717 match immediately following or immediately before internal newlines in the
718 subject string, respectively, as well as at the very start and end. This is
719 equivalent to Perl's /m option, and it can be changed within a pattern by a
720 (?m) option setting. If there are no newlines in a subject string, or no
721 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
722 .sp
728 .sp
729 These options override the default newline definition that was chosen when PCRE
730 was built. Setting the first or the second specifies that a newline is
731 indicated by a single character (CR or LF, respectively). Setting
732 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
733 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
734 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
735 that any Unicode newline sequence should be recognized. The Unicode newline
736 sequences are the three just mentioned, plus the single characters VT (vertical
737 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
738 separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
739 library, the last two are recognized only in UTF-8 mode.
740 .P
741 The newline setting in the options word uses three bits that are treated
742 as a number, giving eight possibilities. Currently only six are used (default
743 plus the five values above). This means that if you set more than one newline
744 option, the combination may or may not be sensible. For example,
746 other combinations may yield unused numbers and cause an error.
747 .P
748 The only time that a line break in a pattern is specially recognized when
749 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
750 and so are ignored in this mode. Also, an unescaped # outside a character class
751 indicates a comment that lasts until after the next line break sequence. In
752 other circumstances, line break sequences in patterns are treated as literal
753 data.
754 .P
755 The newline option that is set at compile time becomes the default that is used
756 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
757 .sp
759 .sp
760 If this option is set, it disables the use of numbered capturing parentheses in
761 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
762 were followed by ?: but named parentheses can still be used for capturing (and
763 they acquire numbers in the usual way). There is no equivalent of this option
764 in Perl.
765 .sp
767 .sp
768 This is an option that acts at matching time; that is, it is really an option
769 for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
770 it is remembered with the compiled pattern and assumed at matching time. For
771 details see the discussion of PCRE_NO_START_OPTIMIZE
772 .\" HTML <a href="#execoptions">
773 .\" </a>
774 below.
775 .\"
776 .sp
778 .sp
779 This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
780 \ew, and some of the POSIX character classes. By default, only ASCII characters
781 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
782 classify characters. More details are given in the section on
783 .\" HTML <a href="pcre.html#genericchartypes">
784 .\" </a>
785 generic character types
786 .\"
787 in the
788 .\" HREF
789 \fBpcrepattern\fP
790 .\"
791 page. If you set PCRE_UCP, matching one of the items it affects takes much
792 longer. The option is available only if PCRE has been compiled with Unicode
793 property support.
794 .sp
796 .sp
797 This option inverts the "greediness" of the quantifiers so that they are not
798 greedy by default, but become greedy if followed by "?". It is not compatible
799 with Perl. It can also be set by a (?U) option setting within the pattern.
800 .sp
802 .sp
803 This option causes PCRE to regard both the pattern and the subject as strings
804 of UTF-8 characters instead of single-byte strings. However, it is available
805 only when PCRE is built to include UTF support. If not, the use of this option
806 provokes an error. Details of how this option changes the behaviour of PCRE are
807 given in the
808 .\" HREF
809 \fBpcreunicode\fP
810 .\"
811 page.
812 .sp
814 .sp
815 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8
816 string is automatically checked. There is a discussion about the
817 .\" HTML <a href="pcreunicode.html#utf8strings">
818 .\" </a>
819 validity of UTF-8 strings
820 .\"
821 in the
822 .\" HREF
823 \fBpcreunicode\fP
824 .\"
825 page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
826 error. If you already know that your pattern is valid, and you want to skip
827 this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
828 When it is set, the effect of passing an invalid UTF-8 string as a pattern is
829 undefined. It may cause your program to crash. Note that this option can also
830 be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
831 validity checking of subject strings.
832 .
833 .
835 .rs
836 .sp
837 The following table lists the error codes than may be returned by
838 \fBpcre_compile2()\fP, along with the error messages that may be returned by
839 both compiling functions. Note that error messages are always 8-bit ASCII
840 strings, even in 16-bit mode. As PCRE has developed, some error codes have
841 fallen out of use. To avoid confusion, they have not been re-used.
842 .sp
843 0 no error
844 1 \e at end of pattern
845 2 \ec at end of pattern
846 3 unrecognized character follows \e
847 4 numbers out of order in {} quantifier
848 5 number too big in {} quantifier
849 6 missing terminating ] for character class
850 7 invalid escape sequence in character class
851 8 range out of order in character class
852 9 nothing to repeat
853 10 [this code is not in use]
854 11 internal error: unexpected repeat
855 12 unrecognized character after (? or (?-
856 13 POSIX named classes are supported only within a class
857 14 missing )
858 15 reference to non-existent subpattern
859 16 erroffset passed as NULL
860 17 unknown option bit(s) set
861 18 missing ) after comment
862 19 [this code is not in use]
863 20 regular expression is too large
864 21 failed to get memory
865 22 unmatched parentheses
866 23 internal error: code overflow
867 24 unrecognized character after (?<
868 25 lookbehind assertion is not fixed length
869 26 malformed number or name after (?(
870 27 conditional group contains more than two branches
871 28 assertion expected after (?(
872 29 (?R or (?[+-]digits must be followed by )
873 30 unknown POSIX class name
874 31 POSIX collating elements are not supported
875 32 this version of PCRE is compiled without UTF support
876 33 [this code is not in use]
877 34 character value in \ex{...} sequence is too large
878 35 invalid condition (?(0)
879 36 \eC not allowed in lookbehind assertion
880 37 PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
881 38 number after (?C is > 255
882 39 closing ) for (?C expected
883 40 recursive call could loop indefinitely
884 41 unrecognized character after (?P
885 42 syntax error in subpattern name (missing terminator)
886 43 two named subpatterns have the same name
887 44 invalid UTF-8 string (specifically UTF-8)
888 45 support for \eP, \ep, and \eX has not been compiled
889 46 malformed \eP or \ep sequence
890 47 unknown property name after \eP or \ep
891 48 subpattern name is too long (maximum 32 characters)
892 49 too many named subpatterns (maximum 10000)
893 50 [this code is not in use]
894 51 octal value is greater than \e377 in 8-bit non-UTF-8 mode
895 52 internal error: overran compiling workspace
896 53 internal error: previously-checked referenced subpattern
897 not found
898 54 DEFINE group contains more than one branch
899 55 repeating a DEFINE group is not allowed
900 56 inconsistent NEWLINE options
901 57 \eg is not followed by a braced, angle-bracketed, or quoted
902 name/number or by a plain number
903 58 a numbered reference must not be zero
904 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
905 60 (*VERB) not recognized
906 61 number is too big
907 62 subpattern name expected
908 63 digit expected after (?+
909 64 ] is an invalid data character in JavaScript compatibility mode
910 65 different names for subpatterns of the same number are
911 not allowed
912 66 (*MARK) must have an argument
913 67 this version of PCRE is not compiled with Unicode property
914 support
915 68 \ec must be followed by an ASCII character
916 69 \ek is not followed by a braced, angle-bracketed, or quoted name
917 70 internal error: unknown opcode in find_fixedlength()
918 71 \eN is not supported in a class
919 72 too many forward references
920 73 disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
921 74 invalid UTF-16 string (specifically UTF-16)
922 .sp
923 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
924 be used if the limits were changed when PCRE was built.
925 .
926 .
927 .\" HTML <a name="studyingapattern"></a>
929 .rs
930 .sp
931 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
932 .ti +5n
933 .B const char **\fIerrptr\fP);
934 .PP
935 If a compiled pattern is going to be used several times, it is worth spending
936 more time analyzing it in order to speed up the time taken for matching. The
937 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
938 argument. If studying the pattern produces additional information that will
939 help speed up matching, \fBpcre_study()\fP returns a pointer to a
940 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
941 results of the study.
942 .P
943 The returned value from \fBpcre_study()\fP can be passed directly to
944 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
945 also contains other fields that can be set by the caller before the block is
946 passed; these are described
947 .\" HTML <a href="#extradata">
948 .\" </a>
949 below
950 .\"
951 in the section on matching a pattern.
952 .P
953 If studying the pattern does not produce any useful information,
954 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
955 wants to pass any of the other fields to \fBpcre_exec()\fP or
956 \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
957 .P
958 The second argument of \fBpcre_study()\fP contains option bits. There is only
959 one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
960 compiler is available, the pattern is further compiled into machine code that
961 executes much faster than the \fBpcre_exec()\fP matching function. If
962 the just-in-time compiler is not available, this option is ignored. All other
963 bits in the \fIoptions\fP argument must be zero.
964 .P
965 JIT compilation is a heavyweight optimization. It can take some time for
966 patterns to be analyzed, and for one-off matches and simple patterns the
967 benefit of faster execution might be offset by a much slower study time.
968 Not all patterns can be optimized by the JIT compiler. For those that cannot be
969 handled, matching automatically falls back to the \fBpcre_exec()\fP
970 interpreter. For more details, see the
971 .\" HREF
972 \fBpcrejit\fP
973 .\"
974 documentation.
975 .P
976 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
977 studying succeeds (even if no data is returned), the variable it points to is
978 set to NULL. Otherwise it is set to point to a textual error message. This is a
979 static string that is part of the library. You must not try to free it. You
980 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
981 sure that it has run successfully.
982 .P
983 When you are finished with a pattern, you can free the memory used for the
984 study data by calling \fBpcre_free_study()\fP. This function was added to the
985 API for release 8.20. For earlier versions, the memory could be freed with
986 \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
987 where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
988 new function when convenient.
989 .P
990 This is a typical way in which \fBpcre_study\fP() is used (except that in a
991 real application there should be tests for errors):
992 .sp
993 int rc;
994 pcre *re;
995 pcre_extra *sd;
996 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
997 sd = pcre_study(
998 re, /* result of pcre_compile() */
999 0, /* no options */
1000 &error); /* set to NULL or points to a message */
1001 rc = pcre_exec( /* see below for details of pcre_exec() options */
1002 re, sd, "subject", 7, 0, 0, ovector, 30);
1003 ...
1004 pcre_free_study(sd);
1005 pcre_free(re);
1006 .sp
1007 Studying a pattern does two things: first, a lower bound for the length of
1008 subject string that is needed to match the pattern is computed. This does not
1009 mean that there are any strings of that length that match, but it does
1010 guarantee that no shorter strings match. The value is used by
1011 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
1012 match strings that are shorter than the lower bound. You can find out the value
1013 in a calling program via the \fBpcre_fullinfo()\fP function.
1014 .P
1015 Studying a pattern is also useful for non-anchored patterns that do not have a
1016 single fixed starting character. A bitmap of possible starting bytes is
1017 created. This speeds up finding a position in the subject at which to start
1018 matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1019 .P
1020 These two optimizations apply to both \fBpcre_exec()\fP and
1021 \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
1022 \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
1023 just-in-time compiling is successful. The optimizations can be disabled by
1024 setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
1025 \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
1026 callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
1027 to make use of these facilities in cases where matching fails. See the
1028 discussion of PCRE_NO_START_OPTIMIZE
1029 .\" HTML <a href="#execoptions">
1030 .\" </a>
1031 below.
1032 .\"
1033 .
1034 .
1035 .\" HTML <a name="localesupport"></a>
1037 .rs
1038 .sp
1039 PCRE handles caseless matching, and determines whether characters are letters,
1040 digits, or whatever, by reference to a set of tables, indexed by character
1041 value. When running in UTF-8 mode, this applies only to characters
1042 with codes less than 128. By default, higher-valued codes never match escapes
1043 such as \ew or \ed, but they can be tested with \ep if PCRE is built with
1044 Unicode character property support. Alternatively, the PCRE_UCP option can be
1045 set at compile time; this causes \ew and friends to use Unicode property
1046 support instead of built-in tables. The use of locales with Unicode is
1047 discouraged. If you are handling characters with codes greater than 128, you
1048 should either use UTF-8 and Unicode, or use locales, but not try to mix the
1049 two.
1050 .P
1051 PCRE contains an internal set of tables that are used when the final argument
1052 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
1053 Normally, the internal tables recognize only ASCII characters. However, when
1054 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1055 default "C" locale of the local system, which may cause them to be different.
1056 .P
1057 The internal tables can always be overridden by tables supplied by the
1058 application that calls PCRE. These may be created in a different locale from
1059 the default. As more and more applications change to using Unicode, the need
1060 for this locale support is expected to die away.
1061 .P
1062 External tables are built by calling the \fBpcre_maketables()\fP function,
1063 which has no arguments, in the relevant locale. The result can then be passed
1064 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
1065 example, to build and use tables that are appropriate for the French locale
1066 (where accented characters with values greater than 128 are treated as letters),
1067 the following code could be used:
1068 .sp
1069 setlocale(LC_CTYPE, "fr_FR");
1070 tables = pcre_maketables();
1071 re = pcre_compile(..., tables);
1072 .sp
1073 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1074 are using Windows, the name for the French locale is "french".
1075 .P
1076 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1077 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1078 that the memory containing the tables remains available for as long as it is
1079 needed.
1080 .P
1081 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
1082 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
1083 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
1084 pattern, compilation, studying and matching all happen in the same locale, but
1085 different patterns can be compiled in different locales.
1086 .P
1087 It is possible to pass a table pointer or NULL (indicating the use of the
1088 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
1089 this facility could be used to match a pattern in a different locale from the
1090 one in which it was compiled. Passing table pointers at run time is discussed
1091 below in the section on matching a pattern.
1092 .
1093 .
1094 .\" HTML <a name="infoaboutpattern"></a>
1096 .rs
1097 .sp
1098 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1099 .ti +5n
1100 .B int \fIwhat\fP, void *\fIwhere\fP);
1101 .PP
1102 The \fBpcre_fullinfo()\fP function returns information about a compiled
1103 pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1104 library at version 8.30, after more than 10 years of obsolescence.
1105 .P
1106 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1107 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
1108 the pattern was not studied. The third argument specifies which piece of
1109 information is required, and the fourth argument is a pointer to a variable
1110 to receive the data. The yield of the function is zero for success, or one of
1111 the following negative numbers:
1112 .sp
1113 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1114 the argument \fIwhere\fP was NULL
1115 PCRE_ERROR_BADMAGIC the "magic number" was not found
1116 PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
1117 endianness
1118 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
1119 .sp
1120 The "magic number" is placed at the start of each compiled pattern as an simple
1121 check against passing an arbitrary memory pointer. The endianness error can
1122 occur if a compiled pattern is saved and reloaded on a different host. Here is
1123 a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1124 pattern:
1125 .sp
1126 int rc;
1127 size_t length;
1128 rc = pcre_fullinfo(
1129 re, /* result of pcre_compile() */
1130 sd, /* result of pcre_study(), or NULL */
1131 PCRE_INFO_SIZE, /* what is required */
1132 &length); /* where to put the data */
1133 .sp
1134 The possible values for the third argument are defined in \fBpcre.h\fP, and are
1135 as follows:
1136 .sp
1138 .sp
1139 Return the number of the highest back reference in the pattern. The fourth
1140 argument should point to an \fBint\fP variable. Zero is returned if there are
1141 no back references.
1142 .sp
1144 .sp
1145 Return the number of capturing subpatterns in the pattern. The fourth argument
1146 should point to an \fBint\fP variable.
1147 .sp
1149 .sp
1150 Return a pointer to the internal default character tables within PCRE. The
1151 fourth argument should point to an \fBunsigned char *\fP variable. This
1152 information call is provided for internal use by the \fBpcre_study()\fP
1153 function. External callers can cause PCRE to use its internal tables by passing
1154 a NULL table pointer.
1155 .sp
1157 .sp
1158 Return information about the first data unit of any matched string, for a
1159 non-anchored pattern. (The name of this option refers to the 8-bit library,
1160 where data units are bytes.) The fourth argument should point to an \fBint\fP
1161 variable.
1162 .P
1163 If there is a fixed first value, for example, the letter "c" from a pattern
1164 such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1165 value is always less than 256; in the 16-bit library the value can be up to
1166 0xffff.
1167 .P
1168 If there is no fixed first value, and if either
1169 .sp
1170 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1171 starts with "^", or
1172 .sp
1173 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1174 (if it were set, the pattern would be anchored),
1175 .sp
1176 -1 is returned, indicating that the pattern matches only at the start of a
1177 subject string or after any newline within the string. Otherwise -2 is
1178 returned. For anchored patterns, -2 is returned.
1179 .sp
1181 .sp
1182 If the pattern was studied, and this resulted in the construction of a 256-bit
1183 table indicating a fixed set of values for the first data unit in any matching
1184 string, a pointer to the table is returned. Otherwise NULL is returned. The
1185 fourth argument should point to an \fBunsigned char *\fP variable.
1186 .sp
1188 .sp
1189 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1190 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1191 explicit match is either a literal CR or LF character, or \er or \en.
1192 .sp
1194 .sp
1195 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1196 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1197 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1198 .sp
1200 .sp
1201 Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1202 just-in-time compiling was successful. The fourth argument should point to an
1203 \fBint\fP variable. A return value of 0 means that JIT support is not available
1204 in this version of PCRE, or that the pattern was not studied with the
1205 PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1206 particular pattern. See the
1207 .\" HREF
1208 \fBpcrejit\fP
1209 .\"
1210 documentation for details of what can and cannot be handled.
1211 .sp
1213 .sp
1214 If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1215 return the size of the JIT compiled code, otherwise return zero. The fourth
1216 argument should point to a \fBsize_t\fP variable.
1217 .sp
1219 .sp
1220 Return the value of the rightmost literal data unit that must exist in any
1221 matched string, other than at its start, if such a value has been recorded. The
1222 fourth argument should point to an \fBint\fP variable. If there is no such
1223 value, -1 is returned. For anchored patterns, a last literal value is recorded
1224 only if it follows something of variable length. For example, for the pattern
1225 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1226 is -1.
1227 .sp
1229 .sp
1230 If the pattern was studied and a minimum length for matching subject strings
1231 was computed, its value is returned. Otherwise the returned value is -1. The
1232 value is a number of characters, which in UTF-8 mode may be different from the
1233 number of bytes. The fourth argument should point to an \fBint\fP variable. A
1234 non-negative value is a lower bound to the length of any matching string. There
1235 may not be any strings of that length that do actually match, but every string
1236 that does match is at least that long.
1237 .sp
1241 .sp
1242 PCRE supports the use of named as well as numbered capturing parentheses. The
1243 names are just an additional way of identifying the parentheses, which still
1244 acquire numbers. Several convenience functions such as
1245 \fBpcre_get_named_substring()\fP are provided for extracting captured
1246 substrings by name. It is also possible to extract the data directly, by first
1247 converting the name to a number in order to access the correct pointers in the
1248 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1249 you need to use the name-to-number map, which is described by these three
1250 values.
1251 .P
1252 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1253 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1254 entry; both of these return an \fBint\fP value. The entry size depends on the
1255 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1256 entry of the table. This is a pointer to \fBchar\fP in the 8-bit library, where
1257 the first two bytes of each entry are the number of the capturing parenthesis,
1258 most significant byte first. In the 16-bit library, the pointer points to
1259 16-bit data units, the first of which contains the parenthesis number. The rest
1260 of the entry is the corresponding name, zero terminated.
1261 .P
1262 The names are in alphabetical order. Duplicate names may appear if (?| is used
1263 to create multiple groups with the same number, as described in the
1264 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1265 .\" </a>
1266 section on duplicate subpattern numbers
1267 .\"
1268 in the
1269 .\" HREF
1270 \fBpcrepattern\fP
1271 .\"
1272 page. Duplicate names for subpatterns with different numbers are permitted only
1273 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1274 table in the order in which they were found in the pattern. In the absence of
1275 (?| this is the order of increasing number; when (?| is used this is not
1276 necessarily the case because later subpatterns may have lower numbers.
1277 .P
1278 As a simple example of the name/number table, consider the following pattern
1279 after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1280 space - including newlines - is ignored):
1281 .sp
1282 .\" JOIN
1283 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1284 (?<month>\ed\ed) - (?<day>\ed\ed) )
1285 .sp
1286 There are four named subpatterns, so the table has four entries, and each entry
1287 in the table is eight bytes long. The table is as follows, with non-printing
1288 bytes shows in hexadecimal, and undefined bytes shown as ??:
1289 .sp
1290 00 01 d a t e 00 ??
1291 00 05 d a y 00 ?? ??
1292 00 04 m o n t h 00
1293 00 02 y e a r 00 ??
1294 .sp
1295 When writing code to extract data from named subpatterns using the
1296 name-to-number map, remember that the length of the entries is likely to be
1297 different for each compiled pattern.
1298 .sp
1300 .sp
1301 Return 1 if the pattern can be used for partial matching with
1302 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1303 \fBint\fP variable. From release 8.00, this always returns 1, because the
1304 restrictions that previously applied to partial matching have been lifted. The
1305 .\" HREF
1306 \fBpcrepartial\fP
1307 .\"
1308 documentation gives details of partial matching.
1309 .sp
1311 .sp
1312 Return a copy of the options with which the pattern was compiled. The fourth
1313 argument should point to an \fBunsigned long int\fP variable. These option bits
1314 are those specified in the call to \fBpcre_compile()\fP, modified by any
1315 top-level option settings at the start of the pattern itself. In other words,
1316 they are the options that will be in force when matching starts. For example,
1317 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1319 .P
1320 A pattern is automatically anchored by PCRE if all of its top-level
1321 alternatives begin with one of the following:
1322 .sp
1323 ^ unless PCRE_MULTILINE is set
1324 \eA always
1325 \eG always
1326 .\" JOIN
1327 .* if PCRE_DOTALL is set and there are no back
1328 references to the subpattern in which .* appears
1329 .sp
1330 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1331 \fBpcre_fullinfo()\fP.
1332 .sp
1334 .sp
1335 Return the size of the compiled pattern in bytes (for both libraries). The
1336 fourth argument should point to a \fBsize_t\fP variable. This value does not
1337 include the size of the \fBpcre\fP structure that is returned by
1338 \fBpcre_compile()\fP. The value that is passed as the argument to
1339 \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1340 place the compiled data is the value returned by this option plus the size of
1341 the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1342 does not alter the value returned by this option.
1343 .sp
1345 .sp
1346 Return the size in bytes of the data block pointed to by the \fIstudy_data\fP
1347 field in a \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no
1348 study data, zero is returned. The fourth argument should point to a
1349 \fBsize_t\fP variable. The \fIstudy_data\fP field is set by \fBpcre_study()\fP
1350 to record information that will speed up matching (see the section entitled
1351 .\" HTML <a href="#studyingapattern">
1352 .\" </a>
1353 "Studying a pattern"
1354 .\"
1355 above). The format of the \fIstudy_data\fP block is private, but its length
1356 is made available via this option so that it can be saved and restored (see the
1357 .\" HREF
1358 \fBpcreprecompile\fP
1359 .\"
1360 documentation for details).
1361 .
1362 .
1364 .rs
1365 .sp
1366 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1367 .PP
1368 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1369 data block that contains a compiled pattern. It is provided for the benefit of
1370 applications that operate in an object-oriented manner, where different parts
1371 of the application may be using the same compiled pattern, but you want to free
1372 the block when they are all done.
1373 .P
1374 When a pattern is compiled, the reference count field is initialized to zero.
1375 It is changed only by calling this function, whose action is to add the
1376 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1377 function is the new value. However, the value of the count is constrained to
1378 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1379 it is forced to the appropriate limit value.
1380 .P
1381 Except when it is zero, the reference count is not correctly preserved if a
1382 pattern is compiled on one host and then transferred to a host whose byte-order
1383 is different. (This seems a highly unlikely scenario.)
1384 .
1385 .
1387 .rs
1388 .sp
1389 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1390 .ti +5n
1391 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1392 .ti +5n
1393 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1394 .P
1395 The function \fBpcre_exec()\fP is called to match a subject string against a
1396 compiled pattern, which is passed in the \fIcode\fP argument. If the
1397 pattern was studied, the result of the study should be passed in the
1398 \fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
1399 and \fIextra\fP arguments as many times as you like, in order to match
1400 different subject strings with the same pattern.
1401 .P
1402 This function is the main matching facility of the library, and it operates in
1403 a Perl-like manner. For specialist use there is also an alternative matching
1404 function, which is described
1405 .\" HTML <a href="#dfamatch">
1406 .\" </a>
1407 below
1408 .\"
1409 in the section about the \fBpcre_dfa_exec()\fP function.
1410 .P
1411 In most applications, the pattern will have been compiled (and optionally
1412 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1413 possible to save compiled patterns and study data, and then use them later
1414 in different processes, possibly even on different hosts. For a discussion
1415 about this, see the
1416 .\" HREF
1417 \fBpcreprecompile\fP
1418 .\"
1419 documentation.
1420 .P
1421 Here is an example of a simple call to \fBpcre_exec()\fP:
1422 .sp
1423 int rc;
1424 int ovector[30];
1425 rc = pcre_exec(
1426 re, /* result of pcre_compile() */
1427 NULL, /* we didn't study the pattern */
1428 "some string", /* the subject string */
1429 11, /* the length of the subject string */
1430 0, /* start at offset 0 in the subject */
1431 0, /* default options */
1432 ovector, /* vector of integers for substring information */
1433 30); /* number of elements (NOT size in bytes) */
1434 .
1435 .
1436 .\" HTML <a name="extradata"></a>
1437 .SS "Extra data for \fBpcre_exec()\fR"
1438 .rs
1439 .sp
1440 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1441 data block. The \fBpcre_study()\fP function returns such a block (when it
1442 doesn't return NULL), but you can also create one for yourself, and pass
1443 additional information in it. The \fBpcre_extra\fP block contains the following
1444 fields (not necessarily in this order):
1445 .sp
1446 unsigned long int \fIflags\fP;
1447 void *\fIstudy_data\fP;
1448 void *\fIexecutable_jit\fP;
1449 unsigned long int \fImatch_limit\fP;
1450 unsigned long int \fImatch_limit_recursion\fP;
1451 void *\fIcallout_data\fP;
1452 const unsigned char *\fItables\fP;
1453 unsigned char **\fImark\fP;
1454 .sp
1455 In the 16-bit version of this structure, the \fImark\fP field has type
1456 "PCRE_UCHAR16 **".
1457 .P
1458 The \fIflags\fP field is a bitmap that specifies which of the other fields
1459 are set. The flag bits are:
1460 .sp
1468 .sp
1469 Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1470 the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1471 returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1472 should not set these yourself, but you may add to the block by setting the
1473 other fields and their corresponding flag bits.
1474 .P
1475 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1476 vast amount of resources when running patterns that are not going to match,
1477 but which have a very large number of possibilities in their search trees. The
1478 classic example is a pattern that uses nested unlimited repeats.
1479 .P
1480 Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1481 calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1482 imposed on the number of times this function is called during a match, which
1483 has the effect of limiting the amount of backtracking that can take place. For
1484 patterns that are not anchored, the count restarts from zero for each position
1485 in the subject string.
1486 .P
1487 When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1488 with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1489 is entirely different. However, there is still the possibility of runaway
1490 matching that goes on for a very long time, and so the \fImatch_limit\fP value
1491 is also used in this case (but in a different way) to limit how long the
1492 matching can continue.
1493 .P
1494 The default value for the limit can be set when PCRE is built; the default
1495 default is 10 million, which handles all but the most extreme cases. You can
1496 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1497 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1498 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1500 .P
1501 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1502 instead of limiting the total number of times that \fBmatch()\fP is called, it
1503 limits the depth of recursion. The recursion depth is a smaller number than the
1504 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1505 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1506 .P
1507 Limiting the recursion depth limits the amount of machine stack that can be
1508 used, or, when PCRE has been compiled to use memory on the heap instead of the
1509 stack, the amount of heap memory that can be used. This limit is not relevant,
1510 and is ignored, if the pattern was successfully studied with
1512 .P
1513 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1514 built; the default default is the same value as the default for
1515 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1516 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1517 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1518 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1519 .P
1520 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1521 and is described in the
1522 .\" HREF
1523 \fBpcrecallout\fP
1524 .\"
1525 documentation.
1526 .P
1527 The \fItables\fP field is used to pass a character tables pointer to
1528 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1529 pattern. A non-NULL value is stored with the compiled pattern only if custom
1530 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1531 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1532 internal tables to be used. This facility is helpful when re-using patterns
1533 that have been saved after compiling with an external set of tables, because
1534 the external tables might be at a different address when \fBpcre_exec()\fP is
1535 called. See the
1536 .\" HREF
1537 \fBpcreprecompile\fP
1538 .\"
1539 documentation for a discussion of saving compiled patterns for later use.
1540 .P
1541 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1542 be set to point to a suitable variable. If the pattern contains any
1543 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1544 a name to pass back, a pointer to the name string (zero terminated) is placed
1545 in the variable pointed to by the \fImark\fP field. The names are within the
1546 compiled pattern; if you wish to retain such a name you must copy it before
1547 freeing the memory of a compiled pattern. If there is no name to pass back, the
1548 variable pointed to by the \fImark\fP field is set to NULL. For details of the
1549 backtracking control verbs, see the section entitled
1550 .\" HTML <a href="pcrepattern#backtrackcontrol">
1551 .\" </a>
1552 "Backtracking control"
1553 .\"
1554 in the
1555 .\" HREF
1556 \fBpcrepattern\fP
1557 .\"
1558 documentation.
1559 .
1560 .
1561 .\" HTML <a name="execoptions"></a>
1562 .SS "Option bits for \fBpcre_exec()\fP"
1563 .rs
1564 .sp
1565 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1566 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1570 .P
1571 If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1572 the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1574 particular that partial matching is not supported. If an unsupported option is
1575 used, JIT execution is disabled and the normal interpretive code in
1576 \fBpcre_exec()\fP is run.
1577 .sp
1579 .sp
1580 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1581 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1582 to be anchored by virtue of its contents, it cannot be made unachored at
1583 matching time.
1584 .sp
1587 .sp
1588 These options (which are mutually exclusive) control what the \eR escape
1589 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1590 match any Unicode newline sequence. These options override the choice that was
1591 made or defaulted when the pattern was compiled.
1592 .sp
1598 .sp
1599 These options override the newline definition that was chosen or defaulted when
1600 the pattern was compiled. For details, see the description of
1601 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1602 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1603 the way the match position is advanced after a match failure for an unanchored
1604 pattern.
1605 .P
1607 match attempt for an unanchored pattern fails when the current position is at a
1608 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1609 characters, the match position is advanced by two characters instead of one, in
1610 other words, to after the CRLF.
1611 .P
1612 The above rule is a compromise that makes the most common cases work as
1613 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1614 set), it does not match the string "\er\enA" because, after failing at the
1615 start, it skips both the CR and the LF before retrying. However, the pattern
1616 [\er\en]A does match that string, because it contains an explicit CR or LF
1617 reference, and so advances only by one character after the first failure.
1618 .P
1619 An explicit match for CR of LF is either a literal appearance of one of those
1620 characters, or one of the \er or \en escape sequences. Implicit matches such as
1621 [^X] do not count, nor does \es (which includes CR and LF in the characters
1622 that it matches).
1623 .P
1624 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1625 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1626 .sp
1628 .sp
1629 This option specifies that first character of the subject string is not the
1630 beginning of a line, so the circumflex metacharacter should not match before
1631 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1632 never to match. This option affects only the behaviour of the circumflex
1633 metacharacter. It does not affect \eA.
1634 .sp
1636 .sp
1637 This option specifies that the end of the subject string is not the end of a
1638 line, so the dollar metacharacter should not match it nor (except in multiline
1639 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1640 compile time) causes dollar never to match. This option affects only the
1641 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1642 .sp
1644 .sp
1645 An empty string is not considered to be a valid match if this option is set. If
1646 there are alternatives in the pattern, they are tried. If all the alternatives
1647 match the empty string, the entire match fails. For example, if the pattern
1648 .sp
1649 a?b?
1650 .sp
1651 is applied to a string not beginning with "a" or "b", it matches an empty
1652 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1653 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1654 .sp
1656 .sp
1657 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1658 the start of the subject is permitted. If the pattern is anchored, such a match
1659 can occur only if the pattern contains \eK.
1660 .P
1661 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1662 does make a special case of a pattern match of the empty string within its
1663 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1664 emulate Perl's behaviour after matching a null string by first trying the match
1665 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1666 if that fails, by advancing the starting offset (see below) and trying an
1667 ordinary match again. There is some code that demonstrates how to do this in
1668 the
1669 .\" HREF
1670 \fBpcredemo\fP
1671 .\"
1672 sample program. In the most general case, you have to check to see if the
1673 newline convention recognizes CRLF as a newline, and if so, and the current
1674 character is CR followed by LF, advance the starting offset by two characters
1675 instead of one.
1676 .sp
1678 .sp
1679 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1680 a match, in order to speed up the process. For example, if it is known that an
1681 unanchored match must start with a specific character, it searches the subject
1682 for that character, and fails immediately if it cannot find it, without
1683 actually running the main matching function. This means that a special item
1684 such as (*COMMIT) at the start of a pattern is not considered until after a
1685 suitable starting point for the match has been found. When callouts or (*MARK)
1686 items are in use, these "start-up" optimizations can cause them to be skipped
1687 if the pattern is never actually used. The start-up optimizations are in effect
1688 a pre-scan of the subject that takes place before the pattern is run.
1689 .P
1690 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1691 causing performance to suffer, but ensuring that in cases where the result is
1692 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1693 are considered at every possible starting position in the subject string. If
1694 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1695 time.
1696 .P
1697 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1698 Consider the pattern
1699 .sp
1701 .sp
1702 When this is compiled, PCRE records the fact that a match must start with the
1703 character "A". Suppose the subject string is "DEFABC". The start-up
1704 optimization scans along the subject, finds "A" and runs the first match
1705 attempt from there. The (*COMMIT) item means that the pattern must match the
1706 current starting position, which in this case, it does. However, if the same
1707 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1708 subject string does not happen. The first match attempt is run starting from
1709 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1710 the overall result is "no match". If the pattern is studied, more start-up
1711 optimizations may be used. For example, a minimum length for the subject may be
1712 recorded. Consider the pattern
1713 .sp
1714 (*MARK:A)(X|Y)
1715 .sp
1716 The minimum length for a match is one character. If the subject is "ABC", there
1717 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1718 If the pattern is studied, the final attempt does not take place, because PCRE
1719 knows that the subject is too short, and so the (*MARK) is never encountered.
1720 In this case, studying the pattern does not affect the overall match result,
1721 which is still "no match", but it does affect the auxiliary information that is
1722 returned.
1723 .sp
1725 .sp
1726 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1727 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1728 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1729 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1730 strings in the
1731 .\" HREF
1732 \fBpcreunicode\fP
1733 .\"
1734 page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1735 error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1736 truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1737 cases, information about the precise nature of the error may also be returned
1738 (see the descriptions of these errors in the section entitled \fIError return
1739 values from\fP \fBpcre_exec()\fP
1740 .\" HTML <a href="#errorlist">
1741 .\" </a>
1742 below).
1743 .\"
1744 If \fIstartoffset\fP contains a value that does not point to the start of a
1745 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1746 returned.
1747 .P
1748 If you already know that your subject is valid, and you want to skip these
1749 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1750 calling \fBpcre_exec()\fP. You might want to do this for the second and
1751 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1752 all the matches in a single subject string. However, you should be sure that
1753 the value of \fIstartoffset\fP points to the start of a character (or the end
1754 of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1755 invalid string as a subject or an invalid value of \fIstartoffset\fP is
1756 undefined. Your program may crash.
1757 .sp
1760 .sp
1761 These options turn on the partial matching feature. For backwards
1762 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1763 occurs if the end of the subject string is reached successfully, but there are
1764 not enough subject characters to complete the match. If this happens when
1765 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1766 testing any remaining alternatives. Only if no complete match can be found is
1767 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1768 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1769 but only if no complete match can be found.
1770 .P
1771 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1772 partial match is found, \fBpcre_exec()\fP immediately returns
1773 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1774 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1775 important that an alternative complete match.
1776 .P
1777 In both cases, the portion of the string that was inspected when the partial
1778 match was found is set as the first matching string. There is a more detailed
1779 discussion of partial and multi-segment matching, with examples, in the
1780 .\" HREF
1781 \fBpcrepartial\fP
1782 .\"
1783 documentation.
1784 .
1785 .
1786 .SS "The string to be matched by \fBpcre_exec()\fP"
1787 .rs
1788 .sp
1789 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1790 \fIsubject\fP, a length in bytes in \fIlength\fP, and a starting byte offset
1791 in \fIstartoffset\fP. If this is negative or greater than the length of the
1792 subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1793 offset is zero, the search for a match starts at the beginning of the subject,
1794 and this is by far the most common case. In UTF-8 mode, the byte offset must
1795 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1796 pattern string, the subject may contain binary zero bytes.
1797 .P
1798 A non-zero starting offset is useful when searching for another match in the
1799 same subject by calling \fBpcre_exec()\fP again after a previous success.
1800 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1801 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1802 lookbehind. For example, consider the pattern
1803 .sp
1804 \eBiss\eB
1805 .sp
1806 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1807 the current position in the subject is not a word boundary.) When applied to
1808 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1809 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1810 subject, namely "issipi", it does not match, because \eB is always false at the
1811 start of the subject, which is deemed to be a word boundary. However, if
1812 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1813 set to 4, it finds the second occurrence of "iss" because it is able to look
1814 behind the starting point to discover that it is preceded by a letter.
1815 .P
1816 Finding all the matches in a subject is tricky when the pattern can match an
1817 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1818 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1819 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1820 and trying an ordinary match again. There is some code that demonstrates how to
1821 do this in the
1822 .\" HREF
1823 \fBpcredemo\fP
1824 .\"
1825 sample program. In the most general case, you have to check to see if the
1826 newline convention recognizes CRLF as a newline, and if so, and the current
1827 character is CR followed by LF, advance the starting offset by two characters
1828 instead of one.
1829 .P
1830 If a non-zero starting offset is passed when the pattern is anchored, one
1831 attempt to match at the given offset is made. This can only succeed if the
1832 pattern does not require the match to be at the start of the subject.
1833 .
1834 .
1835 .SS "How \fBpcre_exec()\fP returns captured substrings"
1836 .rs
1837 .sp
1838 In general, a pattern matches a certain portion of the subject, and in
1839 addition, further substrings from the subject may be picked out by parts of the
1840 pattern. Following the usage in Jeffrey Friedl's book, this is called
1841 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1842 a fragment of a pattern that picks out a substring. PCRE supports several other
1843 kinds of parenthesized subpattern that do not cause substrings to be captured.
1844 .P
1845 Captured substrings are returned to the caller via a vector of integers whose
1846 address is passed in \fIovector\fP. The number of elements in the vector is
1847 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1848 argument is NOT the size of \fIovector\fP in bytes.
1849 .P
1850 The first two-thirds of the vector is used to pass back captured substrings,
1851 each substring using a pair of integers. The remaining third of the vector is
1852 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1853 and is not available for passing back information. The number passed in
1854 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1855 rounded down.
1856 .P
1857 When a match is successful, information about captured substrings is returned
1858 in pairs of integers, starting at the beginning of \fIovector\fP, and
1859 continuing up to two-thirds of its length at the most. The first element of
1860 each pair is set to the byte offset of the first character in a substring, and
1861 the second is set to the byte offset of the first character after the end of a
1862 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1863 mode. They are not character counts.
1864 .P
1865 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1866 portion of the subject string matched by the entire pattern. The next pair is
1867 used for the first capturing subpattern, and so on. The value returned by
1868 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1869 For example, if two substrings have been captured, the returned value is 3. If
1870 there are no capturing subpatterns, the return value from a successful match is
1871 1, indicating that just the first pair of offsets has been set.
1872 .P
1873 If a capturing subpattern is matched repeatedly, it is the last portion of the
1874 string that it matched that is returned.
1875 .P
1876 If the vector is too small to hold all the captured substring offsets, it is
1877 used as far as possible (up to two-thirds of its length), and the function
1878 returns a value of zero. If neither the actual string matched not any captured
1879 substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
1880 passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
1881 back references and the \fIovector\fP is not big enough to remember the related
1882 substrings, PCRE has to get additional memory for use during matching. Thus it
1883 is usually advisable to supply an \fIovector\fP of reasonable size.
1884 .P
1885 There are some cases where zero is returned (indicating vector overflow) when
1886 in fact the vector is exactly the right size for the final match. For example,
1887 consider the pattern
1888 .sp
1889 (a)(?:(b)c|bd)
1890 .sp
1891 If a vector of 6 elements (allowing for only 1 captured substring) is given
1892 with subject string "abd", \fBpcre_exec()\fP will try to set the second
1893 captured string, thereby recording a vector overflow, before failing to match
1894 "c" and backing up to try the second alternative. The zero return, however,
1895 does correctly indicate that the maximum number of slots (namely 2) have been
1896 filled. In similar cases where there is temporary overflow, but the final
1897 number of used slots is actually less than the maximum, a non-zero value is
1898 returned.
1899 .P
1900 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1901 subpatterns there are in a compiled pattern. The smallest size for
1902 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1903 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1904 .P
1905 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1906 the subject when subpattern \fIn\fP has not been used at all. For example, if
1907 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1908 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1909 happens, both values in the offset pairs corresponding to unused subpatterns
1910 are set to -1.
1911 .P
1912 Offset values that correspond to unused subpatterns at the end of the
1913 expression are also set to -1. For example, if the string "abc" is matched
1914 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1915 return from the function is 2, because the highest used capturing subpattern
1916 number is 1, and the offsets for for the second and third capturing subpatterns
1917 (assuming the vector is large enough, of course) are set to -1.
1918 .P
1919 \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1920 correspond to capturing parentheses in the pattern are never changed. That is,
1921 if a pattern contains \fIn\fP capturing parentheses, no more than
1922 \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1923 elements (in the first two-thirds) retain whatever values they previously had.
1924 .P
1925 Some convenience functions are provided for extracting the captured substrings
1926 as separate strings. These are described below.
1927 .
1928 .
1929 .\" HTML <a name="errorlist"></a>
1930 .SS "Error return values from \fBpcre_exec()\fP"
1931 .rs
1932 .sp
1933 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1934 defined in the header file:
1935 .sp
1937 .sp
1938 The subject string did not match the pattern.
1939 .sp
1941 .sp
1942 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1943 NULL and \fIovecsize\fP was not zero.
1944 .sp
1946 .sp
1947 An unrecognized bit was set in the \fIoptions\fP argument.
1948 .sp
1950 .sp
1951 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1952 the case when it is passed a junk pointer and to detect when a pattern that was
1953 compiled in an environment of one endianness is run in an environment with the
1954 other endianness. This is the error that PCRE gives when the magic number is
1955 not present.
1956 .sp
1958 .sp
1959 While running the pattern match, an unknown item was encountered in the
1960 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1961 of the compiled pattern.
1962 .sp
1964 .sp
1965 If a pattern contains back references, but the \fIovector\fP that is passed to
1966 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1967 gets a block of memory at the start of matching to use for this purpose. If the
1968 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1969 automatically freed at the end of matching.
1970 .P
1971 This error is also given if \fBpcre_stack_malloc()\fP fails in
1972 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1973 \fB--disable-stack-for-recursion\fP.
1974 .sp
1976 .sp
1977 This error is used by the \fBpcre_copy_substring()\fP,
1978 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1979 below). It is never returned by \fBpcre_exec()\fP.
1980 .sp
1982 .sp
1983 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1984 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1985 above.
1986 .sp
1988 .sp
1989 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1990 use by callout functions that want to yield a distinctive error code. See the
1991 .\" HREF
1992 \fBpcrecallout\fP
1993 .\"
1994 documentation for details.
1995 .sp
1997 .sp
1998 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1999 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2000 (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2001 UTF-8 character is placed in the first element, and a reason code is placed in
2002 the second element. The reason codes are listed in the
2003 .\" HTML <a href="#badutf8reasons">
2004 .\" </a>
2005 following section.
2006 .\"
2007 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2008 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2009 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2010 .sp
2012 .sp
2013 The UTF-8 byte sequence that was passed as a subject was checked and found to
2014 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
2015 \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2016 end of the subject.
2017 .sp
2019 .sp
2020 The subject string did not match, but it did match partially. See the
2021 .\" HREF
2022 \fBpcrepartial\fP
2023 .\"
2024 documentation for details of partial matching.
2025 .sp
2027 .sp
2028 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2029 option was used with a compiled pattern containing items that were not
2030 supported for partial matching. From release 8.00 onwards, there are no
2031 restrictions on partial matching.
2032 .sp
2034 .sp
2035 An unexpected internal error has occurred. This error could be caused by a bug
2036 in PCRE or by overwriting of the compiled pattern.
2037 .sp
2039 .sp
2040 This error is given if the value of the \fIovecsize\fP argument is negative.
2041 .sp
2043 .sp
2044 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
2045 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
2046 description above.
2047 .sp
2049 .sp
2050 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2051 .sp
2053 .sp
2054 The value of \fIstartoffset\fP was negative or greater than the length of the
2055 subject, that is, the value in \fIlength\fP.
2056 .sp
2058 .sp
2059 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2060 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2061 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2062 fact sufficient to detect this case, but this special error code for
2063 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2064 retained for backwards compatibility.
2065 .sp
2067 .sp
2068 This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2069 the pattern. Specifically, it means that either the whole pattern or a
2070 subpattern has been called recursively for the second time at the same position
2071 in the subject string. Some simple patterns that might do this are detected and
2072 faulted at compile time, but more complicated cases, in particular mutual
2073 recursions between two different subpatterns, cannot be detected until run
2074 time.
2075 .sp
2077 .sp
2078 This error is returned when a pattern that was successfully studied using the
2079 PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2080 the just-in-time processing stack is not large enough. See the
2081 .\" HREF
2082 \fBpcrejit\fP
2083 .\"
2084 documentation for more details.
2085 .sp
2087 .sp
2088 This error is given if a pattern that was compiled by the 8-bit library is
2089 passed to a 16-bit library function, or vice versa.
2090 .sp
2092 .sp
2093 This error is given if a pattern that was compiled and saved is reloaded on a
2094 host with different endianness. The utility function
2095 \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2096 so that it runs on the new host.
2097 .P
2098 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
2099 .
2100 .
2101 .\" HTML <a name="badutf8reasons"></a>
2102 .SS "Reason codes for invalid UTF-8 strings"
2103 .rs
2104 .sp
2105 This section applies only to the 8-bit library. The corresponding information
2106 for the 16-bit library is given in the
2107 .\" HREF
2108 \fBpcre16\fP
2109 .\"
2110 page.
2111 .P
2112 When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2113 PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2114 least 2, the offset of the start of the invalid UTF-8 character is placed in
2115 the first output vector element (\fIovector[0]\fP) and a reason code is placed
2116 in the second element (\fIovector[1]\fP). The reason codes are given names in
2117 the \fBpcre.h\fP header file:
2118 .sp
2124 .sp
2125 The string ends with a truncated UTF-8 character; the code specifies how many
2126 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2127 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2128 allows for up to 6 bytes, and this is checked first; hence the possibility of
2129 4 or 5 missing bytes.
2130 .sp
2135 PCRE_UTF8_ERR10
2136 .sp
2137 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2138 character do not have the binary value 0b10 (that is, either the most
2139 significant bit is 0, or the next bit is 1).
2140 .sp
2141 PCRE_UTF8_ERR11
2142 PCRE_UTF8_ERR12
2143 .sp
2144 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2145 these code points are excluded by RFC 3629.
2146 .sp
2147 PCRE_UTF8_ERR13
2148 .sp
2149 A 4-byte character has a value greater than 0x10fff; these code points are
2150 excluded by RFC 3629.
2151 .sp
2152 PCRE_UTF8_ERR14
2153 .sp
2154 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2155 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2156 from UTF-8.
2157 .sp
2158 PCRE_UTF8_ERR15
2159 PCRE_UTF8_ERR16
2160 PCRE_UTF8_ERR17
2161 PCRE_UTF8_ERR18
2162 PCRE_UTF8_ERR19
2163 .sp
2164 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2165 value that can be represented by fewer bytes, which is invalid. For example,
2166 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2167 one byte.
2168 .sp
2169 PCRE_UTF8_ERR20
2170 .sp
2171 The two most significant bits of the first byte of a character have the binary
2172 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2173 byte can only validly occur as the second or subsequent byte of a multi-byte
2174 character.
2175 .sp
2176 PCRE_UTF8_ERR21
2177 .sp
2178 The first byte of a character has the value 0xfe or 0xff. These values can
2179 never occur in a valid UTF-8 string.
2180 .
2181 .
2183 .rs
2184 .sp
2185 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2186 .ti +5n
2187 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
2188 .ti +5n
2189 .B int \fIbuffersize\fP);
2190 .PP
2191 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2192 .ti +5n
2193 .B int \fIstringcount\fP, int \fIstringnumber\fP,
2194 .ti +5n
2195 .B const char **\fIstringptr\fP);
2196 .PP
2197 .B int pcre_get_substring_list(const char *\fIsubject\fP,
2198 .ti +5n
2199 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
2200 .PP
2201 Captured substrings can be accessed directly by using the offsets returned by
2202 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
2203 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
2204 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2205 as new, separate, zero-terminated strings. These functions identify substrings
2206 by number. The next section describes functions for extracting named
2207 substrings.
2208 .P
2209 A substring that contains a binary zero is correctly extracted and has a
2210 further zero added on the end, but the result is not, of course, a C string.
2211 However, you can process such a string by referring to the length that is
2212 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2213 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2214 for handling strings containing binary zeros, because the end of the final
2215 string is not independently indicated.
2216 .P
2217 The first three arguments are the same for all three of these functions:
2218 \fIsubject\fP is the subject string that has just been successfully matched,
2219 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
2220 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
2221 captured by the match, including the substring that matched the entire regular
2222 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
2223 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
2224 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
2225 number of elements in the vector divided by three.
2226 .P
2227 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
2228 extract a single substring, whose number is given as \fIstringnumber\fP. A
2229 value of zero extracts the substring that matched the entire pattern, whereas
2230 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2231 the string is placed in \fIbuffer\fP, whose length is given by
2232 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2233 obtained via \fBpcre_malloc\fP, and its address is returned via
2234 \fIstringptr\fP. The yield of the function is the length of the string, not
2235 including the terminating zero, or one of these error codes:
2236 .sp
2238 .sp
2239 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2240 memory failed for \fBpcre_get_substring()\fP.
2241 .sp
2243 .sp
2244 There is no substring whose number is \fIstringnumber\fP.
2245 .P
2246 The \fBpcre_get_substring_list()\fP function extracts all available substrings
2247 and builds a list of pointers to them. All this is done in a single block of
2248 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2249 is returned via \fIlistptr\fP, which is also the start of the list of string
2250 pointers. The end of the list is marked by a NULL pointer. The yield of the
2251 function is zero if all went well, or the error code
2252 .sp
2254 .sp
2255 if the attempt to get the memory block failed.
2256 .P
2257 When any of these functions encounter a substring that is unset, which can
2258 happen when capturing subpattern number \fIn+1\fP matches some part of the
2259 subject, but subpattern \fIn\fP has not been used at all, they return an empty
2260 string. This can be distinguished from a genuine zero-length substring by
2261 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2262 substrings.
2263 .P
2264 The two convenience functions \fBpcre_free_substring()\fP and
2265 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2266 a previous call of \fBpcre_get_substring()\fP or
2267 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2268 the function pointed to by \fBpcre_free\fP, which of course could be called
2269 directly from a C program. However, PCRE is used in some situations where it is
2270 linked via a special interface to another programming language that cannot use
2271 \fBpcre_free\fP directly; it is for these cases that the functions are
2272 provided.
2273 .
2274 .
2276 .rs
2277 .sp
2278 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2279 .ti +5n
2280 .B const char *\fIname\fP);
2281 .PP
2282 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2283 .ti +5n
2284 .B const char *\fIsubject\fP, int *\fIovector\fP,
2285 .ti +5n
2286 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2287 .ti +5n
2288 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2289 .PP
2290 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2291 .ti +5n
2292 .B const char *\fIsubject\fP, int *\fIovector\fP,
2293 .ti +5n
2294 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2295 .ti +5n
2296 .B const char **\fIstringptr\fP);
2297 .PP
2298 To extract a substring by name, you first have to find associated number.
2299 For example, for this pattern
2300 .sp
2301 (a+)b(?<xxx>\ed+)...
2302 .sp
2303 the number of the subpattern called "xxx" is 2. If the name is known to be
2304 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2305 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2306 pattern, and the second is the name. The yield of the function is the
2307 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2308 that name.
2309 .P
2310 Given the number, you can extract the substring directly, or use one of the
2311 functions described in the previous section. For convenience, there are also
2312 two functions that do the whole job.
2313 .P
2314 Most of the arguments of \fBpcre_copy_named_substring()\fP and
2315 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2316 functions that extract by number. As these are described in the previous
2317 section, they are not re-described here. There are just two differences:
2318 .P
2319 First, instead of a substring number, a substring name is given. Second, there
2320 is an extra argument, given at the start, which is a pointer to the compiled
2321 pattern. This is needed in order to gain access to the name-to-number
2322 translation table.
2323 .P
2324 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2325 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2326 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2327 the behaviour may not be what you want (see the next section).
2328 .P
2329 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2330 subpatterns with the same number, as described in the
2331 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2332 .\" </a>
2333 section on duplicate subpattern numbers
2334 .\"
2335 in the
2336 .\" HREF
2337 \fBpcrepattern\fP
2338 .\"
2339 page, you cannot use names to distinguish the different subpatterns, because
2340 names are not included in the compiled code. The matching process uses only
2341 numbers. For this reason, the use of different names for subpatterns of the
2342 same number causes an error at compile time.
2343 .
2344 .
2346 .rs
2347 .sp
2348 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2349 .ti +5n
2350 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2351 .PP
2352 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2353 are not required to be unique. (Duplicate names are always allowed for
2354 subpatterns with the same number, created by using the (?| feature. Indeed, if
2355 such subpatterns are named, they are required to use the same names.)
2356 .P
2357 Normally, patterns with duplicate names are such that in any one match, only
2358 one of the named subpatterns participates. An example is shown in the
2359 .\" HREF
2360 \fBpcrepattern\fP
2361 .\"
2362 documentation.
2363 .P
2364 When duplicates are present, \fBpcre_copy_named_substring()\fP and
2365 \fBpcre_get_named_substring()\fP return the first substring corresponding to
2366 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2367 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2368 returns one of the numbers that are associated with the name, but it is not
2369 defined which it is.
2370 .P
2371 If you want to get full details of all captured substrings for a given name,
2372 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2373 argument is the compiled pattern, and the second is the name. The third and
2374 fourth are pointers to variables which are updated by the function. After it
2375 has run, they point to the first and last entries in the name-to-number table
2376 for the given name. The function itself returns the length of each entry, or
2377 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2378 described above in the section entitled \fIInformation about a pattern\fP
2379 .\" HTML <a href="#infoaboutpattern">
2380 .\" </a>
2381 above.
2382 .\"
2383 Given all the relevant entries for the name, you can extract each of their
2384 numbers, and hence the captured data, if any.
2385 .
2386 .
2388 .rs
2389 .sp
2390 The traditional matching function uses a similar algorithm to Perl, which stops
2391 when it finds the first match, starting at a given point in the subject. If you
2392 want to find all possible matches, or the longest possible match, consider
2393 using the alternative matching function (see below) instead. If you cannot use
2394 the alternative function, but still need to find all possible matches, you
2395 can kludge it up by making use of the callout facility, which is described in
2396 the
2397 .\" HREF
2398 \fBpcrecallout\fP
2399 .\"
2400 documentation.
2401 .P
2402 What you have to do is to insert a callout right at the end of the pattern.
2403 When your callout function is called, extract and save the current matched
2404 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2405 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2406 will yield PCRE_ERROR_NOMATCH.
2407 .
2408 .
2409 .\" HTML <a name="dfamatch"></a>
2411 .rs
2412 .sp
2413 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2414 .ti +5n
2415 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2416 .ti +5n
2417 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2418 .ti +5n
2419 .B int *\fIworkspace\fP, int \fIwscount\fP);
2420 .P
2421 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2422 a compiled pattern, using a matching algorithm that scans the subject string
2423 just once, and does not backtrack. This has different characteristics to the
2424 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2425 patterns are not supported. Nevertheless, there are times when this kind of
2426 matching can be useful. For a discussion of the two matching algorithms, and a
2427 list of features that \fBpcre_dfa_exec()\fP does not support, see the
2428 .\" HREF
2429 \fBpcrematching\fP
2430 .\"
2431 documentation.
2432 .P
2433 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2434 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2435 different way, and this is described below. The other common arguments are used
2436 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2437 here.
2438 .P
2439 The two additional arguments provide workspace for the function. The workspace
2440 vector should contain at least 20 elements. It is used for keeping track of
2441 multiple paths through the pattern tree. More workspace will be needed for
2442 patterns and subjects where there are a lot of potential matches.
2443 .P
2444 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2445 .sp
2446 int rc;
2447 int ovector[10];
2448 int wspace[20];
2449 rc = pcre_dfa_exec(
2450 re, /* result of pcre_compile() */
2451 NULL, /* we didn't study the pattern */
2452 "some string", /* the subject string */
2453 11, /* the length of the subject string */
2454 0, /* start at offset 0 in the subject */
2455 0, /* default options */
2456 ovector, /* vector of integers for substring information */
2457 10, /* number of elements (NOT size in bytes) */
2458 wspace, /* working space vector */
2459 20); /* number of elements (NOT size in bytes) */
2460 .
2461 .SS "Option bits for \fBpcre_dfa_exec()\fP"
2462 .rs
2463 .sp
2464 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2465 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2469 All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2470 so their description is not repeated here.
2471 .sp
2474 .sp
2475 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2476 details are slightly different. When PCRE_PARTIAL_HARD is set for
2477 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2478 is reached and there is still at least one matching possibility that requires
2479 additional characters. This happens even if some complete matches have also
2480 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2481 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2482 there have been no complete matches, but there is still at least one matching
2483 possibility. The portion of the string that was inspected when the longest
2484 partial match was found is set as the first matching string in both cases.
2485 There is a more detailed discussion of partial and multi-segment matching, with
2486 examples, in the
2487 .\" HREF
2488 \fBpcrepartial\fP
2489 .\"
2490 documentation.
2491 .sp
2493 .sp
2494 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2495 soon as it has found one match. Because of the way the alternative algorithm
2496 works, this is necessarily the shortest possible match at the first possible
2497 matching point in the subject string.
2498 .sp
2500 .sp
2501 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2502 again, with additional subject characters, and have it continue with the same
2503 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2504 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2505 before because data about the match so far is left in them after a partial
2506 match. There is more discussion of this facility in the
2507 .\" HREF
2508 \fBpcrepartial\fP
2509 .\"
2510 documentation.
2511 .
2512 .
2513 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2514 .rs
2515 .sp
2516 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2517 substring in the subject. Note, however, that all the matches from one run of
2518 the function start at the same point in the subject. The shorter matches are
2519 all initial substrings of the longer matches. For example, if the pattern
2520 .sp
2521 <.*>
2522 .sp
2523 is matched against the string
2524 .sp
2525 This is <something> <something else> <something further> no more
2526 .sp
2527 the three matched strings are
2528 .sp
2529 <something>
2530 <something> <something else>
2531 <something> <something else> <something further>
2532 .sp
2533 On success, the yield of the function is a number greater than zero, which is
2534 the number of matched substrings. The substrings themselves are returned in
2535 \fIovector\fP. Each string uses two elements; the first is the offset to the
2536 start, and the second is the offset to the end. In fact, all the strings have
2537 the same start offset. (Space could have been saved by giving this only once,
2538 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2539 returns data, even though the meaning of the strings is different.)
2540 .P
2541 The strings are returned in reverse order of length; that is, the longest
2542 matching string is given first. If there were too many matches to fit into
2543 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2544 the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2545 the entire \fIovector\fP for returning matched strings.
2546 .
2547 .
2548 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2549 .rs
2550 .sp
2551 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2552 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2553 described
2554 .\" HTML <a href="#errorlist">
2555 .\" </a>
2556 above.
2557 .\"
2558 There are in addition the following errors that are specific to
2559 \fBpcre_dfa_exec()\fP:
2560 .sp
2562 .sp
2563 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2564 that it does not support, for instance, the use of \eC or a back reference.
2565 .sp
2567 .sp
2568 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2569 uses a back reference for the condition, or a test for recursion in a specific
2570 group. These are not supported.
2571 .sp
2573 .sp
2574 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2575 block that contains a setting of the \fImatch_limit\fP or
2576 \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2577 meaningless for DFA matching).
2578 .sp
2580 .sp
2581 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2582 \fIworkspace\fP vector.
2583 .sp
2585 .sp
2586 When a recursive subpattern is processed, the matching function calls itself
2587 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2588 error is given if the output vector is not large enough. This should be
2589 extremely rare, as a vector of size 1000 is used.
2590 .
2591 .
2592 .SH "SEE ALSO"
2593 .rs
2594 .sp
2595 \fBpcre16\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2596 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2597 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2598 .
2599 .
2601 .rs
2602 .sp
2603 .nf
2604 Philip Hazel
2605 University Computing Service
2606 Cambridge CB2 3QH, England.
2607 .fi
2608 .
2609 .
2611 .rs
2612 .sp
2613 .nf
2614 Last updated: 07 January 2012
2615 Copyright (c) 1997-2012 University of Cambridge.
2616 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5