ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 440 - (show annotations)
Wed Sep 9 10:14:40 2009 UTC (11 years, 7 months ago) by ph10
File size: 87031 byte(s)
Fix typo.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 .B #include <pcre.h>
8 .PP
9 .SM
10 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11 .ti +5n
12 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13 .ti +5n
14 .B const unsigned char *\fItableptr\fP);
15 .PP
16 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17 .ti +5n
18 .B int *\fIerrorcodeptr\fP,
19 .ti +5n
20 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21 .ti +5n
22 .B const unsigned char *\fItableptr\fP);
23 .PP
24 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25 .ti +5n
26 .B const char **\fIerrptr\fP);
27 .PP
28 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29 .ti +5n
30 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31 .ti +5n
32 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33 .PP
34 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35 .ti +5n
36 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37 .ti +5n
38 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39 .ti +5n
40 .B int *\fIworkspace\fP, int \fIwscount\fP);
41 .PP
42 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43 .ti +5n
44 .B const char *\fIsubject\fP, int *\fIovector\fP,
45 .ti +5n
46 .B int \fIstringcount\fP, const char *\fIstringname\fP,
47 .ti +5n
48 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49 .PP
50 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
51 .ti +5n
52 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53 .ti +5n
54 .B int \fIbuffersize\fP);
55 .PP
56 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57 .ti +5n
58 .B const char *\fIsubject\fP, int *\fIovector\fP,
59 .ti +5n
60 .B int \fIstringcount\fP, const char *\fIstringname\fP,
61 .ti +5n
62 .B const char **\fIstringptr\fP);
63 .PP
64 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65 .ti +5n
66 .B const char *\fIname\fP);
67 .PP
68 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
69 .ti +5n
70 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71 .PP
72 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73 .ti +5n
74 .B int \fIstringcount\fP, int \fIstringnumber\fP,
75 .ti +5n
76 .B const char **\fIstringptr\fP);
77 .PP
78 .B int pcre_get_substring_list(const char *\fIsubject\fP,
79 .ti +5n
80 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81 .PP
82 .B void pcre_free_substring(const char *\fIstringptr\fP);
83 .PP
84 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
85 .PP
86 .B const unsigned char *pcre_maketables(void);
87 .PP
88 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
89 .ti +5n
90 .B int \fIwhat\fP, void *\fIwhere\fP);
91 .PP
92 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93 .B *\fIfirstcharptr\fP);
94 .PP
95 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96 .PP
97 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
98 .PP
99 .B char *pcre_version(void);
100 .PP
101 .B void *(*pcre_malloc)(size_t);
102 .PP
103 .B void (*pcre_free)(void *);
104 .PP
105 .B void *(*pcre_stack_malloc)(size_t);
106 .PP
107 .B void (*pcre_stack_free)(void *);
108 .PP
109 .B int (*pcre_callout)(pcre_callout_block *);
110 .
111 .
113 .rs
114 .sp
115 PCRE has its own native API, which is described in this document. There are
116 also some wrapper functions that correspond to the POSIX regular expression
117 API. These are described in the
118 .\" HREF
119 \fBpcreposix\fP
120 .\"
121 documentation. Both of these APIs define a set of C function calls. A C++
122 wrapper is distributed with PCRE. It is documented in the
123 .\" HREF
124 \fBpcrecpp\fP
125 .\"
126 page.
127 .P
128 The native API C function prototypes are defined in the header file
129 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132 and PCRE_MINOR to contain the major and minor release numbers for the library.
133 Applications can use these to include support for different releases of PCRE.
134 .P
135 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137 in a Perl-compatible manner. A sample program that demonstrates the simplest
138 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
139 source distribution. A listing of this program is given in the
140 .\" HREF
141 \fBpcredemo\fP
142 .\"
143 documentation, and the
144 .\" HREF
145 \fBpcresample\fP
146 .\"
147 documentation describes how to compile and run it.
148 .P
149 A second matching function, \fBpcre_dfa_exec()\fP, which is not
150 Perl-compatible, is also provided. This uses a different algorithm for the
151 matching. The alternative algorithm finds all possible matches (at a given
152 point in the subject), and scans the subject just once (unless there are
153 lookbehind assertions). However, this algorithm does not return captured
154 substrings. A description of the two matching algorithms and their advantages
155 and disadvantages is given in the
156 .\" HREF
157 \fBpcrematching\fP
158 .\"
159 documentation.
160 .P
161 In addition to the main compiling and matching functions, there are convenience
162 functions for extracting captured substrings from a subject string that is
163 matched by \fBpcre_exec()\fP. They are:
164 .sp
165 \fBpcre_copy_substring()\fP
166 \fBpcre_copy_named_substring()\fP
167 \fBpcre_get_substring()\fP
168 \fBpcre_get_named_substring()\fP
169 \fBpcre_get_substring_list()\fP
170 \fBpcre_get_stringnumber()\fP
171 \fBpcre_get_stringtable_entries()\fP
172 .sp
173 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
174 provided, to free the memory used for extracted strings.
175 .P
176 The function \fBpcre_maketables()\fP is used to build a set of character tables
177 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
178 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
179 specialist use. Most commonly, no special tables are passed, in which case
180 internal tables that are generated when PCRE is built are used.
181 .P
182 The function \fBpcre_fullinfo()\fP is used to find out information about a
183 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
184 some of the available information, but is retained for backwards compatibility.
185 The function \fBpcre_version()\fP returns a pointer to a string containing the
186 version of PCRE and its date of release.
187 .P
188 The function \fBpcre_refcount()\fP maintains a reference count in a data block
189 containing a compiled pattern. This is provided for the benefit of
190 object-oriented applications.
191 .P
192 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
193 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
194 respectively. PCRE calls the memory management functions via these variables,
195 so a calling program can replace them if it wishes to intercept the calls. This
196 should be done before calling any PCRE functions.
197 .P
198 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
199 indirections to memory management functions. These special functions are used
200 only when PCRE is compiled to use the heap for remembering data, instead of
201 recursive function calls, when running the \fBpcre_exec()\fP function. See the
202 .\" HREF
203 \fBpcrebuild\fP
204 .\"
205 documentation for details of how to do this. It is a non-standard way of
206 building PCRE, for use in environments that have limited stacks. Because of the
207 greater use of memory management, it runs more slowly. Separate functions are
208 provided so that special-purpose external code can be used for this case. When
209 used, these functions are always called in a stack-like manner (last obtained,
210 first freed), and always for memory blocks of the same size. There is a
211 discussion about PCRE's stack usage in the
212 .\" HREF
213 \fBpcrestack\fP
214 .\"
215 documentation.
216 .P
217 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
218 by the caller to a "callout" function, which PCRE will then call at specified
219 points during a matching operation. Details are given in the
220 .\" HREF
221 \fBpcrecallout\fP
222 .\"
223 documentation.
224 .
225 .
226 .\" HTML <a name="newlines"></a>
228 .rs
229 .sp
230 PCRE supports five different conventions for indicating line breaks in
231 strings: a single CR (carriage return) character, a single LF (linefeed)
232 character, the two-character sequence CRLF, any of the three preceding, or any
233 Unicode newline sequence. The Unicode newline sequences are the three just
234 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
235 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
236 (paragraph separator, U+2029).
237 .P
238 Each of the first three conventions is used by at least one operating system as
239 its standard newline sequence. When PCRE is built, a default can be specified.
240 The default default is LF, which is the Unix standard. When PCRE is run, the
241 default can be overridden, either when a pattern is compiled, or when it is
242 matched.
243 .P
244 At compile time, the newline convention can be specified by the \fIoptions\fP
245 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
246 start of the pattern itself; this overrides any other settings. See the
247 .\" HREF
248 \fBpcrepattern\fP
249 .\"
250 page for details of the special character sequences.
251 .P
252 In the PCRE documentation the word "newline" is used to mean "the character or
253 pair of characters that indicate a line break". The choice of newline
254 convention affects the handling of the dot, circumflex, and dollar
255 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
256 recognized line ending sequence, the match position advancement for a
257 non-anchored pattern. There is more detail about this in the
258 .\" HTML <a href="#execoptions">
259 .\" </a>
260 section on \fBpcre_exec()\fP options
261 .\"
262 below.
263 .P
264 The choice of newline convention does not affect the interpretation of
265 the \en or \er escape sequences, nor does it affect what \eR matches, which is
266 controlled in a similar way, but by separate options.
267 .
268 .
270 .rs
271 .sp
272 The PCRE functions can be used in multi-threading applications, with the
273 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
274 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
275 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
276 .P
277 The compiled form of a regular expression is not altered during matching, so
278 the same compiled pattern can safely be used by several threads at once.
279 .
280 .
282 .rs
283 .sp
284 The compiled form of a regular expression can be saved and re-used at a later
285 time, possibly by a different program, and even on a host other than the one on
286 which it was compiled. Details are given in the
287 .\" HREF
288 \fBpcreprecompile\fP
289 .\"
290 documentation. However, compiling a regular expression with one version of PCRE
291 for use with a different version is not guaranteed to work and may cause
292 crashes.
293 .
294 .
296 .rs
297 .sp
298 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
299 .PP
300 The function \fBpcre_config()\fP makes it possible for a PCRE client to
301 discover which optional features have been compiled into the PCRE library. The
302 .\" HREF
303 \fBpcrebuild\fP
304 .\"
305 documentation has more details about these optional features.
306 .P
307 The first argument for \fBpcre_config()\fP is an integer, specifying which
308 information is required; the second argument is a pointer to a variable into
309 which the information is placed. The following information is available:
310 .sp
312 .sp
313 The output is an integer that is set to one if UTF-8 support is available;
314 otherwise it is set to zero.
315 .sp
317 .sp
318 The output is an integer that is set to one if support for Unicode character
319 properties is available; otherwise it is set to zero.
320 .sp
322 .sp
323 The output is an integer whose value specifies the default character sequence
324 that is recognized as meaning "newline". The four values that are supported
325 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
326 Though they are derived from ASCII, the same values are returned in EBCDIC
327 environments. The default should normally correspond to the standard sequence
328 for your operating system.
329 .sp
331 .sp
332 The output is an integer whose value indicates what character sequences the \eR
333 escape sequence matches by default. A value of 0 means that \eR matches any
334 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
335 or CRLF. The default can be overridden when a pattern is compiled or matched.
336 .sp
338 .sp
339 The output is an integer that contains the number of bytes used for internal
340 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
341 allow larger regular expressions to be compiled, at the expense of slower
342 matching. The default value of 2 is sufficient for all but the most massive
343 patterns, since it allows the compiled pattern to be up to 64K in size.
344 .sp
346 .sp
347 The output is an integer that contains the threshold above which the POSIX
348 interface uses \fBmalloc()\fP for output vectors. Further details are given in
349 the
350 .\" HREF
351 \fBpcreposix\fP
352 .\"
353 documentation.
354 .sp
356 .sp
357 The output is a long integer that gives the default limit for the number of
358 internal matching function calls in a \fBpcre_exec()\fP execution. Further
359 details are given with \fBpcre_exec()\fP below.
360 .sp
362 .sp
363 The output is a long integer that gives the default limit for the depth of
364 recursion when calling the internal matching function in a \fBpcre_exec()\fP
365 execution. Further details are given with \fBpcre_exec()\fP below.
366 .sp
368 .sp
369 The output is an integer that is set to one if internal recursion when running
370 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
371 to remember their state. This is the usual way that PCRE is compiled. The
372 output is zero if PCRE was compiled to use blocks of data on the heap instead
373 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
374 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
375 avoiding the use of the stack.
376 .
377 .
379 .rs
380 .sp
381 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
382 .ti +5n
383 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
384 .ti +5n
385 .B const unsigned char *\fItableptr\fP);
386 .sp
387 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
388 .ti +5n
389 .B int *\fIerrorcodeptr\fP,
390 .ti +5n
391 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
392 .ti +5n
393 .B const unsigned char *\fItableptr\fP);
394 .P
395 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
396 called to compile a pattern into an internal form. The only difference between
397 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
398 \fIerrorcodeptr\fP, via which a numerical error code can be returned.
399 .P
400 The pattern is a C string terminated by a binary zero, and is passed in the
401 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
402 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
403 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
404 for a structure whose contents are not externally defined. It is up to the
405 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
406 .P
407 Although the compiled code of a PCRE regex is relocatable, that is, it does not
408 depend on memory location, the complete \fBpcre\fP data block is not
409 fully relocatable, because it may contain a copy of the \fItableptr\fP
410 argument, which is an address (see below).
411 .P
412 The \fIoptions\fP argument contains various bit settings that affect the
413 compilation. It should be zero if no options are required. The available
414 options are described below. Some of them (in particular, those that are
415 compatible with Perl, but also some others) can also be set and unset from
416 within the pattern (see the detailed description in the
417 .\" HREF
418 \fBpcrepattern\fP
419 .\"
420 documentation). For those options that can be different in different parts of
421 the pattern, the contents of the \fIoptions\fP argument specifies their initial
422 settings at the start of compilation and execution. The PCRE_ANCHORED and
423 PCRE_NEWLINE_\fIxxx\fP options can be set at the time of matching as well as at
424 compile time.
425 .P
426 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
427 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
428 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
429 error message. This is a static string that is part of the library. You must
430 not try to free it. The offset from the start of the pattern to the character
431 where the error was discovered is placed in the variable pointed to by
432 \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
433 .P
434 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
435 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
436 returned via this argument in the event of an error. This is in addition to the
437 textual error message. Error codes and messages are listed below.
438 .P
439 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
440 character tables that are built when PCRE is compiled, using the default C
441 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
442 call to \fBpcre_maketables()\fP. This value is stored with the compiled
443 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
444 passed to it. For more discussion, see the section on locale support below.
445 .P
446 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
447 .sp
448 pcre *re;
449 const char *error;
450 int erroffset;
451 re = pcre_compile(
452 "^A.*Z", /* the pattern */
453 0, /* default options */
454 &error, /* for error message */
455 &erroffset, /* for error offset */
456 NULL); /* use default character tables */
457 .sp
458 The following names for option bits are defined in the \fBpcre.h\fP header
459 file:
460 .sp
462 .sp
463 If this bit is set, the pattern is forced to be "anchored", that is, it is
464 constrained to match only at the first matching point in the string that is
465 being searched (the "subject string"). This effect can also be achieved by
466 appropriate constructs in the pattern itself, which is the only way to do it in
467 Perl.
468 .sp
470 .sp
471 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
472 all with number 255, before each pattern item. For discussion of the callout
473 facility, see the
474 .\" HREF
475 \fBpcrecallout\fP
476 .\"
477 documentation.
478 .sp
481 .sp
482 These options (which are mutually exclusive) control what the \eR escape
483 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
484 match any Unicode newline sequence. The default is specified when PCRE is
485 built. It can be overridden from within the pattern, or by setting an option
486 when a compiled pattern is matched.
487 .sp
489 .sp
490 If this bit is set, letters in the pattern match both upper and lower case
491 letters. It is equivalent to Perl's /i option, and it can be changed within a
492 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
493 concept of case for characters whose values are less than 128, so caseless
494 matching is always possible. For characters with higher values, the concept of
495 case is supported if PCRE is compiled with Unicode property support, but not
496 otherwise. If you want to use caseless matching for characters 128 and above,
497 you must ensure that PCRE is compiled with Unicode property support as well as
498 with UTF-8 support.
499 .sp
501 .sp
502 If this bit is set, a dollar metacharacter in the pattern matches only at the
503 end of the subject string. Without this option, a dollar also matches
504 immediately before a newline at the end of the string (but not before any other
505 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
506 There is no equivalent to this option in Perl, and no way to set it within a
507 pattern.
508 .sp
510 .sp
511 If this bit is set, a dot metacharater in the pattern matches all characters,
512 including those that indicate newline. Without it, a dot does not match when
513 the current position is at a newline. This option is equivalent to Perl's /s
514 option, and it can be changed within a pattern by a (?s) option setting. A
515 negative class such as [^a] always matches newline characters, independent of
516 the setting of this option.
517 .sp
519 .sp
520 If this bit is set, names used to identify capturing subpatterns need not be
521 unique. This can be helpful for certain types of pattern when it is known that
522 only one instance of the named subpattern can ever be matched. There are more
523 details of named subpatterns below; see also the
524 .\" HREF
525 \fBpcrepattern\fP
526 .\"
527 documentation.
528 .sp
530 .sp
531 If this bit is set, whitespace data characters in the pattern are totally
532 ignored except when escaped or inside a character class. Whitespace does not
533 include the VT character (code 11). In addition, characters between an
534 unescaped # outside a character class and the next newline, inclusive, are also
535 ignored. This is equivalent to Perl's /x option, and it can be changed within a
536 pattern by a (?x) option setting.
537 .P
538 This option makes it possible to include comments inside complicated patterns.
539 Note, however, that this applies only to data characters. Whitespace characters
540 may never appear within special character sequences in a pattern, for example
541 within the sequence (?( which introduces a conditional subpattern.
542 .sp
544 .sp
545 This option was invented in order to turn on additional functionality of PCRE
546 that is incompatible with Perl, but it is currently of very little use. When
547 set, any backslash in a pattern that is followed by a letter that has no
548 special meaning causes an error, thus reserving these combinations for future
549 expansion. By default, as in Perl, a backslash followed by a letter with no
550 special meaning is treated as a literal. (Perl can, however, be persuaded to
551 give a warning for this.) There are at present no other features controlled by
552 this option. It can also be set by a (?X) option setting within a pattern.
553 .sp
555 .sp
556 If this option is set, an unanchored pattern is required to match before or at
557 the first newline in the subject string, though the matched text may continue
558 over the newline.
559 .sp
561 .sp
562 If this option is set, PCRE's behaviour is changed in some ways so that it is
563 compatible with JavaScript rather than Perl. The changes are as follows:
564 .P
565 (1) A lone closing square bracket in a pattern causes a compile-time error,
566 because this is illegal in JavaScript (by default it is treated as a data
567 character). Thus, the pattern AB]CD becomes illegal when this option is set.
568 .P
569 (2) At run time, a back reference to an unset subpattern group matches an empty
570 string (by default this causes the current matching alternative to fail). A
571 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
572 an "a" in the subject), whereas it fails by default, for Perl compatibility.
573 .sp
575 .sp
576 By default, PCRE treats the subject string as consisting of a single line of
577 characters (even if it actually contains newlines). The "start of line"
578 metacharacter (^) matches only at the start of the string, while the "end of
579 line" metacharacter ($) matches only at the end of the string, or before a
580 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
581 Perl.
582 .P
583 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
584 match immediately following or immediately before internal newlines in the
585 subject string, respectively, as well as at the very start and end. This is
586 equivalent to Perl's /m option, and it can be changed within a pattern by a
587 (?m) option setting. If there are no newlines in a subject string, or no
588 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
589 .sp
595 .sp
596 These options override the default newline definition that was chosen when PCRE
597 was built. Setting the first or the second specifies that a newline is
598 indicated by a single character (CR or LF, respectively). Setting
599 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
600 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
601 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
602 that any Unicode newline sequence should be recognized. The Unicode newline
603 sequences are the three just mentioned, plus the single characters VT (vertical
604 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
605 separator, U+2028), and PS (paragraph separator, U+2029). The last two are
606 recognized only in UTF-8 mode.
607 .P
608 The newline setting in the options word uses three bits that are treated
609 as a number, giving eight possibilities. Currently only six are used (default
610 plus the five values above). This means that if you set more than one newline
611 option, the combination may or may not be sensible. For example,
613 other combinations may yield unused numbers and cause an error.
614 .P
615 The only time that a line break is specially recognized when compiling a
616 pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
617 class is encountered. This indicates a comment that lasts until after the next
618 line break sequence. In other circumstances, line break sequences are treated
619 as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
620 as whitespace characters and are therefore ignored.
621 .P
622 The newline option that is set at compile time becomes the default that is used
623 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
624 .sp
626 .sp
627 If this option is set, it disables the use of numbered capturing parentheses in
628 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
629 were followed by ?: but named parentheses can still be used for capturing (and
630 they acquire numbers in the usual way). There is no equivalent of this option
631 in Perl.
632 .sp
634 .sp
635 This option inverts the "greediness" of the quantifiers so that they are not
636 greedy by default, but become greedy if followed by "?". It is not compatible
637 with Perl. It can also be set by a (?U) option setting within the pattern.
638 .sp
640 .sp
641 This option causes PCRE to regard both the pattern and the subject as strings
642 of UTF-8 characters instead of single-byte character strings. However, it is
643 available only when PCRE is built to include UTF-8 support. If not, the use
644 of this option provokes an error. Details of how this option changes the
645 behaviour of PCRE are given in the
646 .\" HTML <a href="pcre.html#utf8support">
647 .\" </a>
648 section on UTF-8 support
649 .\"
650 in the main
651 .\" HREF
652 \fBpcre\fP
653 .\"
654 page.
655 .sp
657 .sp
658 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
659 automatically checked. There is a discussion about the
660 .\" HTML <a href="pcre.html#utf8strings">
661 .\" </a>
662 validity of UTF-8 strings
663 .\"
664 in the main
665 .\" HREF
666 \fBpcre\fP
667 .\"
668 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
669 returns an error. If you already know that your pattern is valid, and you want
670 to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
671 option. When it is set, the effect of passing an invalid UTF-8 string as a
672 pattern is undefined. It may cause your program to crash. Note that this option
673 can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
674 the UTF-8 validity checking of subject strings.
675 .
676 .
678 .rs
679 .sp
680 The following table lists the error codes than may be returned by
681 \fBpcre_compile2()\fP, along with the error messages that may be returned by
682 both compiling functions. As PCRE has developed, some error codes have fallen
683 out of use. To avoid confusion, they have not been re-used.
684 .sp
685 0 no error
686 1 \e at end of pattern
687 2 \ec at end of pattern
688 3 unrecognized character follows \e
689 4 numbers out of order in {} quantifier
690 5 number too big in {} quantifier
691 6 missing terminating ] for character class
692 7 invalid escape sequence in character class
693 8 range out of order in character class
694 9 nothing to repeat
695 10 [this code is not in use]
696 11 internal error: unexpected repeat
697 12 unrecognized character after (? or (?-
698 13 POSIX named classes are supported only within a class
699 14 missing )
700 15 reference to non-existent subpattern
701 16 erroffset passed as NULL
702 17 unknown option bit(s) set
703 18 missing ) after comment
704 19 [this code is not in use]
705 20 regular expression is too large
706 21 failed to get memory
707 22 unmatched parentheses
708 23 internal error: code overflow
709 24 unrecognized character after (?<
710 25 lookbehind assertion is not fixed length
711 26 malformed number or name after (?(
712 27 conditional group contains more than two branches
713 28 assertion expected after (?(
714 29 (?R or (?[+-]digits must be followed by )
715 30 unknown POSIX class name
716 31 POSIX collating elements are not supported
717 32 this version of PCRE is not compiled with PCRE_UTF8 support
718 33 [this code is not in use]
719 34 character value in \ex{...} sequence is too large
720 35 invalid condition (?(0)
721 36 \eC not allowed in lookbehind assertion
722 37 PCRE does not support \eL, \el, \eN, \eU, or \eu
723 38 number after (?C is > 255
724 39 closing ) for (?C expected
725 40 recursive call could loop indefinitely
726 41 unrecognized character after (?P
727 42 syntax error in subpattern name (missing terminator)
728 43 two named subpatterns have the same name
729 44 invalid UTF-8 string
730 45 support for \eP, \ep, and \eX has not been compiled
731 46 malformed \eP or \ep sequence
732 47 unknown property name after \eP or \ep
733 48 subpattern name is too long (maximum 32 characters)
734 49 too many named subpatterns (maximum 10000)
735 50 [this code is not in use]
736 51 octal value is greater than \e377 (not in UTF-8 mode)
737 52 internal error: overran compiling workspace
738 53 internal error: previously-checked referenced subpattern not found
739 54 DEFINE group contains more than one branch
740 55 repeating a DEFINE group is not allowed
741 56 inconsistent NEWLINE options
742 57 \eg is not followed by a braced, angle-bracketed, or quoted
743 name/number or by a plain number
744 58 a numbered reference must not be zero
745 59 (*VERB) with an argument is not supported
746 60 (*VERB) not recognized
747 61 number is too big
748 62 subpattern name expected
749 63 digit expected after (?+
750 64 ] is an invalid data character in JavaScript compatibility mode
751 .sp
752 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
753 be used if the limits were changed when PCRE was built.
754 .
755 .
757 .rs
758 .sp
759 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
760 .ti +5n
761 .B const char **\fIerrptr\fP);
762 .PP
763 If a compiled pattern is going to be used several times, it is worth spending
764 more time analyzing it in order to speed up the time taken for matching. The
765 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
766 argument. If studying the pattern produces additional information that will
767 help speed up matching, \fBpcre_study()\fP returns a pointer to a
768 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
769 results of the study.
770 .P
771 The returned value from \fBpcre_study()\fP can be passed directly to
772 \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
773 fields that can be set by the caller before the block is passed; these are
774 described
775 .\" HTML <a href="#extradata">
776 .\" </a>
777 below
778 .\"
779 in the section on matching a pattern.
780 .P
781 If studying the pattern does not produce any additional information
782 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
783 wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
784 own \fBpcre_extra\fP block.
785 .P
786 The second argument of \fBpcre_study()\fP contains option bits. At present, no
787 options are defined, and this argument should always be zero.
788 .P
789 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
790 studying succeeds (even if no data is returned), the variable it points to is
791 set to NULL. Otherwise it is set to point to a textual error message. This is a
792 static string that is part of the library. You must not try to free it. You
793 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
794 sure that it has run successfully.
795 .P
796 This is a typical call to \fBpcre_study\fP():
797 .sp
798 pcre_extra *pe;
799 pe = pcre_study(
800 re, /* result of pcre_compile() */
801 0, /* no options exist */
802 &error); /* set to NULL or points to a message */
803 .sp
804 At present, studying a pattern is useful only for non-anchored patterns that do
805 not have a single fixed starting character. A bitmap of possible starting
806 bytes is created.
807 .
808 .
809 .\" HTML <a name="localesupport"></a>
811 .rs
812 .sp
813 PCRE handles caseless matching, and determines whether characters are letters,
814 digits, or whatever, by reference to a set of tables, indexed by character
815 value. When running in UTF-8 mode, this applies only to characters with codes
816 less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
817 can be tested with \ep if PCRE is built with Unicode character property
818 support. The use of locales with Unicode is discouraged. If you are handling
819 characters with codes greater than 128, you should either use UTF-8 and
820 Unicode, or use locales, but not try to mix the two.
821 .P
822 PCRE contains an internal set of tables that are used when the final argument
823 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
824 Normally, the internal tables recognize only ASCII characters. However, when
825 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
826 default "C" locale of the local system, which may cause them to be different.
827 .P
828 The internal tables can always be overridden by tables supplied by the
829 application that calls PCRE. These may be created in a different locale from
830 the default. As more and more applications change to using Unicode, the need
831 for this locale support is expected to die away.
832 .P
833 External tables are built by calling the \fBpcre_maketables()\fP function,
834 which has no arguments, in the relevant locale. The result can then be passed
835 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
836 example, to build and use tables that are appropriate for the French locale
837 (where accented characters with values greater than 128 are treated as letters),
838 the following code could be used:
839 .sp
840 setlocale(LC_CTYPE, "fr_FR");
841 tables = pcre_maketables();
842 re = pcre_compile(..., tables);
843 .sp
844 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
845 are using Windows, the name for the French locale is "french".
846 .P
847 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
848 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
849 that the memory containing the tables remains available for as long as it is
850 needed.
851 .P
852 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
853 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
854 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
855 pattern, compilation, studying and matching all happen in the same locale, but
856 different patterns can be compiled in different locales.
857 .P
858 It is possible to pass a table pointer or NULL (indicating the use of the
859 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
860 this facility could be used to match a pattern in a different locale from the
861 one in which it was compiled. Passing table pointers at run time is discussed
862 below in the section on matching a pattern.
863 .
864 .
866 .rs
867 .sp
868 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
869 .ti +5n
870 .B int \fIwhat\fP, void *\fIwhere\fP);
871 .PP
872 The \fBpcre_fullinfo()\fP function returns information about a compiled
873 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
874 nevertheless retained for backwards compability (and is documented below).
875 .P
876 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
877 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
878 the pattern was not studied. The third argument specifies which piece of
879 information is required, and the fourth argument is a pointer to a variable
880 to receive the data. The yield of the function is zero for success, or one of
881 the following negative numbers:
882 .sp
883 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
884 the argument \fIwhere\fP was NULL
885 PCRE_ERROR_BADMAGIC the "magic number" was not found
886 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
887 .sp
888 The "magic number" is placed at the start of each compiled pattern as an simple
889 check against passing an arbitrary memory pointer. Here is a typical call of
890 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
891 .sp
892 int rc;
893 size_t length;
894 rc = pcre_fullinfo(
895 re, /* result of pcre_compile() */
896 pe, /* result of pcre_study(), or NULL */
897 PCRE_INFO_SIZE, /* what is required */
898 &length); /* where to put the data */
899 .sp
900 The possible values for the third argument are defined in \fBpcre.h\fP, and are
901 as follows:
902 .sp
904 .sp
905 Return the number of the highest back reference in the pattern. The fourth
906 argument should point to an \fBint\fP variable. Zero is returned if there are
907 no back references.
908 .sp
910 .sp
911 Return the number of capturing subpatterns in the pattern. The fourth argument
912 should point to an \fBint\fP variable.
913 .sp
915 .sp
916 Return a pointer to the internal default character tables within PCRE. The
917 fourth argument should point to an \fBunsigned char *\fP variable. This
918 information call is provided for internal use by the \fBpcre_study()\fP
919 function. External callers can cause PCRE to use its internal tables by passing
920 a NULL table pointer.
921 .sp
923 .sp
924 Return information about the first byte of any matched string, for a
925 non-anchored pattern. The fourth argument should point to an \fBint\fP
926 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
927 still recognized for backwards compatibility.)
928 .P
929 If there is a fixed first byte, for example, from a pattern such as
930 (cat|cow|coyote), its value is returned. Otherwise, if either
931 .sp
932 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
933 starts with "^", or
934 .sp
935 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
936 (if it were set, the pattern would be anchored),
937 .sp
938 -1 is returned, indicating that the pattern matches only at the start of a
939 subject string or after any newline within the string. Otherwise -2 is
940 returned. For anchored patterns, -2 is returned.
941 .sp
943 .sp
944 If the pattern was studied, and this resulted in the construction of a 256-bit
945 table indicating a fixed set of bytes for the first byte in any matching
946 string, a pointer to the table is returned. Otherwise NULL is returned. The
947 fourth argument should point to an \fBunsigned char *\fP variable.
948 .sp
950 .sp
951 Return 1 if the pattern contains any explicit matches for CR or LF characters,
952 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
953 explicit match is either a literal CR or LF character, or \er or \en.
954 .sp
956 .sp
957 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
958 0. The fourth argument should point to an \fBint\fP variable. (?J) and
959 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
960 .sp
962 .sp
963 Return the value of the rightmost literal byte that must exist in any matched
964 string, other than at its start, if such a byte has been recorded. The fourth
965 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
966 returned. For anchored patterns, a last literal byte is recorded only if it
967 follows something of variable length. For example, for the pattern
968 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
969 is -1.
970 .sp
974 .sp
975 PCRE supports the use of named as well as numbered capturing parentheses. The
976 names are just an additional way of identifying the parentheses, which still
977 acquire numbers. Several convenience functions such as
978 \fBpcre_get_named_substring()\fP are provided for extracting captured
979 substrings by name. It is also possible to extract the data directly, by first
980 converting the name to a number in order to access the correct pointers in the
981 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
982 you need to use the name-to-number map, which is described by these three
983 values.
984 .P
985 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
986 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
987 entry; both of these return an \fBint\fP value. The entry size depends on the
988 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
989 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
990 are the number of the capturing parenthesis, most significant byte first. The
991 rest of the entry is the corresponding name, zero terminated. The names are in
992 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
993 their parentheses numbers. For example, consider the following pattern (assume
994 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
995 .sp
996 .\" JOIN
997 (?<date> (?<year>(\ed\ed)?\ed\ed) -
998 (?<month>\ed\ed) - (?<day>\ed\ed) )
999 .sp
1000 There are four named subpatterns, so the table has four entries, and each entry
1001 in the table is eight bytes long. The table is as follows, with non-printing
1002 bytes shows in hexadecimal, and undefined bytes shown as ??:
1003 .sp
1004 00 01 d a t e 00 ??
1005 00 05 d a y 00 ?? ??
1006 00 04 m o n t h 00
1007 00 02 y e a r 00 ??
1008 .sp
1009 When writing code to extract data from named subpatterns using the
1010 name-to-number map, remember that the length of the entries is likely to be
1011 different for each compiled pattern.
1012 .sp
1014 .sp
1015 Return 1 if the pattern can be used for partial matching with
1016 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1017 \fBint\fP variable. From release 8.00, this always returns 1, because the
1018 restrictions that previously applied to partial matching have been lifted. The
1019 .\" HREF
1020 \fBpcrepartial\fP
1021 .\"
1022 documentation gives details of partial matching.
1023 .sp
1025 .sp
1026 Return a copy of the options with which the pattern was compiled. The fourth
1027 argument should point to an \fBunsigned long int\fP variable. These option bits
1028 are those specified in the call to \fBpcre_compile()\fP, modified by any
1029 top-level option settings at the start of the pattern itself. In other words,
1030 they are the options that will be in force when matching starts. For example,
1031 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1033 .P
1034 A pattern is automatically anchored by PCRE if all of its top-level
1035 alternatives begin with one of the following:
1036 .sp
1037 ^ unless PCRE_MULTILINE is set
1038 \eA always
1039 \eG always
1040 .\" JOIN
1041 .* if PCRE_DOTALL is set and there are no back
1042 references to the subpattern in which .* appears
1043 .sp
1044 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1045 \fBpcre_fullinfo()\fP.
1046 .sp
1048 .sp
1049 Return the size of the compiled pattern, that is, the value that was passed as
1050 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1051 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1052 variable.
1053 .sp
1055 .sp
1056 Return the size of the data block pointed to by the \fIstudy_data\fP field in
1057 a \fBpcre_extra\fP block. That is, it is the value that was passed to
1058 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1059 created by \fBpcre_study()\fP. The fourth argument should point to a
1060 \fBsize_t\fP variable.
1061 .
1062 .
1064 .rs
1065 .sp
1066 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1067 .B *\fIfirstcharptr\fP);
1068 .PP
1069 The \fBpcre_info()\fP function is now obsolete because its interface is too
1070 restrictive to return all the available data about a compiled pattern. New
1071 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1072 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1073 following negative numbers:
1074 .sp
1075 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1076 PCRE_ERROR_BADMAGIC the "magic number" was not found
1077 .sp
1078 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1079 pattern was compiled is placed in the integer it points to (see
1080 PCRE_INFO_OPTIONS above).
1081 .P
1082 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1083 it is used to pass back information about the first character of any matched
1084 string (see PCRE_INFO_FIRSTBYTE above).
1085 .
1086 .
1088 .rs
1089 .sp
1090 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1091 .PP
1092 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1093 data block that contains a compiled pattern. It is provided for the benefit of
1094 applications that operate in an object-oriented manner, where different parts
1095 of the application may be using the same compiled pattern, but you want to free
1096 the block when they are all done.
1097 .P
1098 When a pattern is compiled, the reference count field is initialized to zero.
1099 It is changed only by calling this function, whose action is to add the
1100 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1101 function is the new value. However, the value of the count is constrained to
1102 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1103 it is forced to the appropriate limit value.
1104 .P
1105 Except when it is zero, the reference count is not correctly preserved if a
1106 pattern is compiled on one host and then transferred to a host whose byte-order
1107 is different. (This seems a highly unlikely scenario.)
1108 .
1109 .
1111 .rs
1112 .sp
1113 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1114 .ti +5n
1115 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1116 .ti +5n
1117 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1118 .P
1119 The function \fBpcre_exec()\fP is called to match a subject string against a
1120 compiled pattern, which is passed in the \fIcode\fP argument. If the
1121 pattern has been studied, the result of the study should be passed in the
1122 \fIextra\fP argument. This function is the main matching facility of the
1123 library, and it operates in a Perl-like manner. For specialist use there is
1124 also an alternative matching function, which is described
1125 .\" HTML <a href="#dfamatch">
1126 .\" </a>
1127 below
1128 .\"
1129 in the section about the \fBpcre_dfa_exec()\fP function.
1130 .P
1131 In most applications, the pattern will have been compiled (and optionally
1132 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1133 possible to save compiled patterns and study data, and then use them later
1134 in different processes, possibly even on different hosts. For a discussion
1135 about this, see the
1136 .\" HREF
1137 \fBpcreprecompile\fP
1138 .\"
1139 documentation.
1140 .P
1141 Here is an example of a simple call to \fBpcre_exec()\fP:
1142 .sp
1143 int rc;
1144 int ovector[30];
1145 rc = pcre_exec(
1146 re, /* result of pcre_compile() */
1147 NULL, /* we didn't study the pattern */
1148 "some string", /* the subject string */
1149 11, /* the length of the subject string */
1150 0, /* start at offset 0 in the subject */
1151 0, /* default options */
1152 ovector, /* vector of integers for substring information */
1153 30); /* number of elements (NOT size in bytes) */
1154 .
1155 .\" HTML <a name="extradata"></a>
1156 .SS "Extra data for \fBpcre_exec()\fR"
1157 .rs
1158 .sp
1159 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1160 data block. The \fBpcre_study()\fP function returns such a block (when it
1161 doesn't return NULL), but you can also create one for yourself, and pass
1162 additional information in it. The \fBpcre_extra\fP block contains the following
1163 fields (not necessarily in this order):
1164 .sp
1165 unsigned long int \fIflags\fP;
1166 void *\fIstudy_data\fP;
1167 unsigned long int \fImatch_limit\fP;
1168 unsigned long int \fImatch_limit_recursion\fP;
1169 void *\fIcallout_data\fP;
1170 const unsigned char *\fItables\fP;
1171 .sp
1172 The \fIflags\fP field is a bitmap that specifies which of the other fields
1173 are set. The flag bits are:
1174 .sp
1180 .sp
1181 Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1182 \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1183 the appropriate flag bit. You should not set this yourself, but you may add to
1184 the block by setting the other fields and their corresponding flag bits.
1185 .P
1186 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1187 vast amount of resources when running patterns that are not going to match,
1188 but which have a very large number of possibilities in their search trees. The
1189 classic example is the use of nested unlimited repeats.
1190 .P
1191 Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1192 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1193 number of times this function is called during a match, which has the effect of
1194 limiting the amount of backtracking that can take place. For patterns that are
1195 not anchored, the count restarts from zero for each position in the subject
1196 string.
1197 .P
1198 The default value for the limit can be set when PCRE is built; the default
1199 default is 10 million, which handles all but the most extreme cases. You can
1200 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1201 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1202 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1204 .P
1205 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1206 instead of limiting the total number of times that \fBmatch()\fP is called, it
1207 limits the depth of recursion. The recursion depth is a smaller number than the
1208 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1209 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1210 .P
1211 Limiting the recursion depth limits the amount of stack that can be used, or,
1212 when PCRE has been compiled to use memory on the heap instead of the stack, the
1213 amount of heap memory that can be used.
1214 .P
1215 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1216 built; the default default is the same value as the default for
1217 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1218 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1219 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1220 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1221 .P
1222 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1223 and is described in the
1224 .\" HREF
1225 \fBpcrecallout\fP
1226 .\"
1227 documentation.
1228 .P
1229 The \fItables\fP field is used to pass a character tables pointer to
1230 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1231 pattern. A non-NULL value is stored with the compiled pattern only if custom
1232 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1233 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1234 internal tables to be used. This facility is helpful when re-using patterns
1235 that have been saved after compiling with an external set of tables, because
1236 the external tables might be at a different address when \fBpcre_exec()\fP is
1237 called. See the
1238 .\" HREF
1239 \fBpcreprecompile\fP
1240 .\"
1241 documentation for a discussion of saving compiled patterns for later use.
1242 .
1243 .\" HTML <a name="execoptions"></a>
1244 .SS "Option bits for \fBpcre_exec()\fP"
1245 .rs
1246 .sp
1247 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1248 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1251 .sp
1253 .sp
1254 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1255 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1256 to be anchored by virtue of its contents, it cannot be made unachored at
1257 matching time.
1258 .sp
1261 .sp
1262 These options (which are mutually exclusive) control what the \eR escape
1263 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1264 match any Unicode newline sequence. These options override the choice that was
1265 made or defaulted when the pattern was compiled.
1266 .sp
1272 .sp
1273 These options override the newline definition that was chosen or defaulted when
1274 the pattern was compiled. For details, see the description of
1275 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1276 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1277 the way the match position is advanced after a match failure for an unanchored
1278 pattern.
1279 .P
1281 match attempt for an unanchored pattern fails when the current position is at a
1282 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1283 characters, the match position is advanced by two characters instead of one, in
1284 other words, to after the CRLF.
1285 .P
1286 The above rule is a compromise that makes the most common cases work as
1287 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1288 set), it does not match the string "\er\enA" because, after failing at the
1289 start, it skips both the CR and the LF before retrying. However, the pattern
1290 [\er\en]A does match that string, because it contains an explicit CR or LF
1291 reference, and so advances only by one character after the first failure.
1292 .P
1293 An explicit match for CR of LF is either a literal appearance of one of those
1294 characters, or one of the \er or \en escape sequences. Implicit matches such as
1295 [^X] do not count, nor does \es (which includes CR and LF in the characters
1296 that it matches).
1297 .P
1298 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1299 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1300 .sp
1302 .sp
1303 This option specifies that first character of the subject string is not the
1304 beginning of a line, so the circumflex metacharacter should not match before
1305 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1306 never to match. This option affects only the behaviour of the circumflex
1307 metacharacter. It does not affect \eA.
1308 .sp
1310 .sp
1311 This option specifies that the end of the subject string is not the end of a
1312 line, so the dollar metacharacter should not match it nor (except in multiline
1313 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1314 compile time) causes dollar never to match. This option affects only the
1315 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1316 .sp
1318 .sp
1319 An empty string is not considered to be a valid match if this option is set. If
1320 there are alternatives in the pattern, they are tried. If all the alternatives
1321 match the empty string, the entire match fails. For example, if the pattern
1322 .sp
1323 a?b?
1324 .sp
1325 is applied to a string not beginning with "a" or "b", it matches the empty
1326 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1327 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1328 .P
1329 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1330 of a pattern match of the empty string within its \fBsplit()\fP function, and
1331 when using the /g modifier. It is possible to emulate Perl's behaviour after
1332 matching a null string by first trying the match again at the same offset with
1333 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1334 starting offset (see below) and trying an ordinary match again. There is some
1335 code that demonstrates how to do this in the
1336 .\" HREF
1337 \fBpcredemo\fP
1338 .\"
1339 sample program.
1340 .sp
1342 .sp
1343 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1344 a match, in order to speed up the process. For example, if it is known that a
1345 match must start with a specific character, it searches the subject for that
1346 character, and fails immediately if it cannot find it, without actually running
1347 the main matching function. When callouts are in use, these optimizations can
1348 cause them to be skipped. This option disables the "start-up" optimizations,
1349 causing performance to suffer, but ensuring that the callouts do occur.
1350 .sp
1352 .sp
1353 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1354 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1355 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1356 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1357 strings in the
1358 .\" HTML <a href="pcre.html#utf8strings">
1359 .\" </a>
1360 section on UTF-8 support
1361 .\"
1362 in the main
1363 .\" HREF
1364 \fBpcre\fP
1365 .\"
1366 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1367 the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1368 PCRE_ERROR_BADUTF8_OFFSET is returned.
1369 .P
1370 If you already know that your subject is valid, and you want to skip these
1371 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1372 calling \fBpcre_exec()\fP. You might want to do this for the second and
1373 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1374 all the matches in a single subject string. However, you should be sure that
1375 the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1376 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1377 subject, or a value of \fIstartoffset\fP that does not point to the start of a
1378 UTF-8 character, is undefined. Your program may crash.
1379 .sp
1382 .sp
1383 These options turn on the partial matching feature. For backwards
1384 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1385 occurs if the end of the subject string is reached successfully, but there are
1386 not enough subject characters to complete the match. If this happens when
1387 PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns
1388 PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues
1389 by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL
1390 returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that
1391 was inspected when the partial match was found is set as the first matching
1392 string. There is a more detailed discussion in the
1393 .\" HREF
1394 \fBpcrepartial\fP
1395 .\"
1396 documentation.
1397 .
1398 .SS "The string to be matched by \fBpcre_exec()\fP"
1399 .rs
1400 .sp
1401 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1402 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1403 in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1404 a UTF-8 character. Unlike the pattern string, the subject may contain binary
1405 zero bytes. When the starting offset is zero, the search for a match starts at
1406 the beginning of the subject, and this is by far the most common case.
1407 .P
1408 A non-zero starting offset is useful when searching for another match in the
1409 same subject by calling \fBpcre_exec()\fP again after a previous success.
1410 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1411 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1412 lookbehind. For example, consider the pattern
1413 .sp
1414 \eBiss\eB
1415 .sp
1416 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1417 the current position in the subject is not a word boundary.) When applied to
1418 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1419 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1420 subject, namely "issipi", it does not match, because \eB is always false at the
1421 start of the subject, which is deemed to be a word boundary. However, if
1422 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1423 set to 4, it finds the second occurrence of "iss" because it is able to look
1424 behind the starting point to discover that it is preceded by a letter.
1425 .P
1426 If a non-zero starting offset is passed when the pattern is anchored, one
1427 attempt to match at the given offset is made. This can only succeed if the
1428 pattern does not require the match to be at the start of the subject.
1429 .
1430 .SS "How \fBpcre_exec()\fP returns captured substrings"
1431 .rs
1432 .sp
1433 In general, a pattern matches a certain portion of the subject, and in
1434 addition, further substrings from the subject may be picked out by parts of the
1435 pattern. Following the usage in Jeffrey Friedl's book, this is called
1436 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1437 a fragment of a pattern that picks out a substring. PCRE supports several other
1438 kinds of parenthesized subpattern that do not cause substrings to be captured.
1439 .P
1440 Captured substrings are returned to the caller via a vector of integers whose
1441 address is passed in \fIovector\fP. The number of elements in the vector is
1442 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1443 argument is NOT the size of \fIovector\fP in bytes.
1444 .P
1445 The first two-thirds of the vector is used to pass back captured substrings,
1446 each substring using a pair of integers. The remaining third of the vector is
1447 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1448 and is not available for passing back information. The number passed in
1449 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1450 rounded down.
1451 .P
1452 When a match is successful, information about captured substrings is returned
1453 in pairs of integers, starting at the beginning of \fIovector\fP, and
1454 continuing up to two-thirds of its length at the most. The first element of
1455 each pair is set to the byte offset of the first character in a substring, and
1456 the second is set to the byte offset of the first character after the end of a
1457 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1458 mode. They are not character counts.
1459 .P
1460 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1461 portion of the subject string matched by the entire pattern. The next pair is
1462 used for the first capturing subpattern, and so on. The value returned by
1463 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1464 For example, if two substrings have been captured, the returned value is 3. If
1465 there are no capturing subpatterns, the return value from a successful match is
1466 1, indicating that just the first pair of offsets has been set.
1467 .P
1468 If a capturing subpattern is matched repeatedly, it is the last portion of the
1469 string that it matched that is returned.
1470 .P
1471 If the vector is too small to hold all the captured substring offsets, it is
1472 used as far as possible (up to two-thirds of its length), and the function
1473 returns a value of zero. If the substring offsets are not of interest,
1474 \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1475 \fIovecsize\fP as zero. However, if the pattern contains back references and
1476 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1477 has to get additional memory for use during matching. Thus it is usually
1478 advisable to supply an \fIovector\fP.
1479 .P
1480 The \fBpcre_info()\fP function can be used to find out how many capturing
1481 subpatterns there are in a compiled pattern. The smallest size for
1482 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1483 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1484 .P
1485 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1486 the subject when subpattern \fIn\fP has not been used at all. For example, if
1487 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1488 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1489 happens, both values in the offset pairs corresponding to unused subpatterns
1490 are set to -1.
1491 .P
1492 Offset values that correspond to unused subpatterns at the end of the
1493 expression are also set to -1. For example, if the string "abc" is matched
1494 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1495 return from the function is 2, because the highest used capturing subpattern
1496 number is 1. However, you can refer to the offsets for the second and third
1497 capturing subpatterns if you wish (assuming the vector is large enough, of
1498 course).
1499 .P
1500 Some convenience functions are provided for extracting the captured substrings
1501 as separate strings. These are described below.
1502 .
1503 .\" HTML <a name="errorlist"></a>
1504 .SS "Error return values from \fBpcre_exec()\fP"
1505 .rs
1506 .sp
1507 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1508 defined in the header file:
1509 .sp
1511 .sp
1512 The subject string did not match the pattern.
1513 .sp
1515 .sp
1516 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1517 NULL and \fIovecsize\fP was not zero.
1518 .sp
1520 .sp
1521 An unrecognized bit was set in the \fIoptions\fP argument.
1522 .sp
1524 .sp
1525 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1526 the case when it is passed a junk pointer and to detect when a pattern that was
1527 compiled in an environment of one endianness is run in an environment with the
1528 other endianness. This is the error that PCRE gives when the magic number is
1529 not present.
1530 .sp
1532 .sp
1533 While running the pattern match, an unknown item was encountered in the
1534 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1535 of the compiled pattern.
1536 .sp
1538 .sp
1539 If a pattern contains back references, but the \fIovector\fP that is passed to
1540 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1541 gets a block of memory at the start of matching to use for this purpose. If the
1542 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1543 automatically freed at the end of matching.
1544 .sp
1546 .sp
1547 This error is used by the \fBpcre_copy_substring()\fP,
1548 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1549 below). It is never returned by \fBpcre_exec()\fP.
1550 .sp
1552 .sp
1553 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1554 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1555 above.
1556 .sp
1558 .sp
1559 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1560 use by callout functions that want to yield a distinctive error code. See the
1561 .\" HREF
1562 \fBpcrecallout\fP
1563 .\"
1564 documentation for details.
1565 .sp
1567 .sp
1568 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1569 .sp
1571 .sp
1572 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1573 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1574 .sp
1576 .sp
1577 The subject string did not match, but it did match partially. See the
1578 .\" HREF
1579 \fBpcrepartial\fP
1580 .\"
1581 documentation for details of partial matching.
1582 .sp
1584 .sp
1585 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1586 option was used with a compiled pattern containing items that were not
1587 supported for partial matching. From release 8.00 onwards, there are no
1588 restrictions on partial matching.
1589 .sp
1591 .sp
1592 An unexpected internal error has occurred. This error could be caused by a bug
1593 in PCRE or by overwriting of the compiled pattern.
1594 .sp
1596 .sp
1597 This error is given if the value of the \fIovecsize\fP argument is negative.
1598 .sp
1600 .sp
1601 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1602 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1603 description above.
1604 .sp
1606 .sp
1607 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1608 .P
1609 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1610 .
1611 .
1613 .rs
1614 .sp
1615 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1616 .ti +5n
1617 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1618 .ti +5n
1619 .B int \fIbuffersize\fP);
1620 .PP
1621 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1622 .ti +5n
1623 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1624 .ti +5n
1625 .B const char **\fIstringptr\fP);
1626 .PP
1627 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1628 .ti +5n
1629 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1630 .PP
1631 Captured substrings can be accessed directly by using the offsets returned by
1632 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1633 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1634 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1635 as new, separate, zero-terminated strings. These functions identify substrings
1636 by number. The next section describes functions for extracting named
1637 substrings.
1638 .P
1639 A substring that contains a binary zero is correctly extracted and has a
1640 further zero added on the end, but the result is not, of course, a C string.
1641 However, you can process such a string by referring to the length that is
1642 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1643 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1644 for handling strings containing binary zeros, because the end of the final
1645 string is not independently indicated.
1646 .P
1647 The first three arguments are the same for all three of these functions:
1648 \fIsubject\fP is the subject string that has just been successfully matched,
1649 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1650 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1651 captured by the match, including the substring that matched the entire regular
1652 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1653 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1654 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1655 number of elements in the vector divided by three.
1656 .P
1657 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1658 extract a single substring, whose number is given as \fIstringnumber\fP. A
1659 value of zero extracts the substring that matched the entire pattern, whereas
1660 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1661 the string is placed in \fIbuffer\fP, whose length is given by
1662 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1663 obtained via \fBpcre_malloc\fP, and its address is returned via
1664 \fIstringptr\fP. The yield of the function is the length of the string, not
1665 including the terminating zero, or one of these error codes:
1666 .sp
1668 .sp
1669 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1670 memory failed for \fBpcre_get_substring()\fP.
1671 .sp
1673 .sp
1674 There is no substring whose number is \fIstringnumber\fP.
1675 .P
1676 The \fBpcre_get_substring_list()\fP function extracts all available substrings
1677 and builds a list of pointers to them. All this is done in a single block of
1678 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1679 is returned via \fIlistptr\fP, which is also the start of the list of string
1680 pointers. The end of the list is marked by a NULL pointer. The yield of the
1681 function is zero if all went well, or the error code
1682 .sp
1684 .sp
1685 if the attempt to get the memory block failed.
1686 .P
1687 When any of these functions encounter a substring that is unset, which can
1688 happen when capturing subpattern number \fIn+1\fP matches some part of the
1689 subject, but subpattern \fIn\fP has not been used at all, they return an empty
1690 string. This can be distinguished from a genuine zero-length substring by
1691 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1692 substrings.
1693 .P
1694 The two convenience functions \fBpcre_free_substring()\fP and
1695 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1696 a previous call of \fBpcre_get_substring()\fP or
1697 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1698 the function pointed to by \fBpcre_free\fP, which of course could be called
1699 directly from a C program. However, PCRE is used in some situations where it is
1700 linked via a special interface to another programming language that cannot use
1701 \fBpcre_free\fP directly; it is for these cases that the functions are
1702 provided.
1703 .
1704 .
1706 .rs
1707 .sp
1708 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1709 .ti +5n
1710 .B const char *\fIname\fP);
1711 .PP
1712 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1713 .ti +5n
1714 .B const char *\fIsubject\fP, int *\fIovector\fP,
1715 .ti +5n
1716 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1717 .ti +5n
1718 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1719 .PP
1720 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1721 .ti +5n
1722 .B const char *\fIsubject\fP, int *\fIovector\fP,
1723 .ti +5n
1724 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1725 .ti +5n
1726 .B const char **\fIstringptr\fP);
1727 .PP
1728 To extract a substring by name, you first have to find associated number.
1729 For example, for this pattern
1730 .sp
1731 (a+)b(?<xxx>\ed+)...
1732 .sp
1733 the number of the subpattern called "xxx" is 2. If the name is known to be
1734 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1735 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1736 pattern, and the second is the name. The yield of the function is the
1737 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1738 that name.
1739 .P
1740 Given the number, you can extract the substring directly, or use one of the
1741 functions described in the previous section. For convenience, there are also
1742 two functions that do the whole job.
1743 .P
1744 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1745 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1746 functions that extract by number. As these are described in the previous
1747 section, they are not re-described here. There are just two differences:
1748 .P
1749 First, instead of a substring number, a substring name is given. Second, there
1750 is an extra argument, given at the start, which is a pointer to the compiled
1751 pattern. This is needed in order to gain access to the name-to-number
1752 translation table.
1753 .P
1754 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1755 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1756 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1757 the behaviour may not be what you want (see the next section).
1758 .P
1759 \fBWarning:\fP If the pattern uses the "(?|" feature to set up multiple
1760 subpatterns with the same number, you cannot use names to distinguish them,
1761 because names are not included in the compiled code. The matching process uses
1762 only numbers.
1763 .
1765 .rs
1766 .sp
1767 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1768 .ti +5n
1769 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1770 .PP
1771 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1772 are not required to be unique. Normally, patterns with duplicate names are such
1773 that in any one match, only one of the named subpatterns participates. An
1774 example is shown in the
1775 .\" HREF
1776 \fBpcrepattern\fP
1777 .\"
1778 documentation.
1779 .P
1780 When duplicates are present, \fBpcre_copy_named_substring()\fP and
1781 \fBpcre_get_named_substring()\fP return the first substring corresponding to
1782 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1783 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1784 returns one of the numbers that are associated with the name, but it is not
1785 defined which it is.
1786 .P
1787 If you want to get full details of all captured substrings for a given name,
1788 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1789 argument is the compiled pattern, and the second is the name. The third and
1790 fourth are pointers to variables which are updated by the function. After it
1791 has run, they point to the first and last entries in the name-to-number table
1792 for the given name. The function itself returns the length of each entry, or
1793 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1794 described above in the section entitled \fIInformation about a pattern\fP.
1795 Given all the relevant entries for the name, you can extract each of their
1796 numbers, and hence the captured data, if any.
1797 .
1798 .
1800 .rs
1801 .sp
1802 The traditional matching function uses a similar algorithm to Perl, which stops
1803 when it finds the first match, starting at a given point in the subject. If you
1804 want to find all possible matches, or the longest possible match, consider
1805 using the alternative matching function (see below) instead. If you cannot use
1806 the alternative function, but still need to find all possible matches, you
1807 can kludge it up by making use of the callout facility, which is described in
1808 the
1809 .\" HREF
1810 \fBpcrecallout\fP
1811 .\"
1812 documentation.
1813 .P
1814 What you have to do is to insert a callout right at the end of the pattern.
1815 When your callout function is called, extract and save the current matched
1816 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1817 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1818 will yield PCRE_ERROR_NOMATCH.
1819 .
1820 .
1821 .\" HTML <a name="dfamatch"></a>
1823 .rs
1824 .sp
1825 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1826 .ti +5n
1827 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1828 .ti +5n
1829 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1830 .ti +5n
1831 .B int *\fIworkspace\fP, int \fIwscount\fP);
1832 .P
1833 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1834 a compiled pattern, using a matching algorithm that scans the subject string
1835 just once, and does not backtrack. This has different characteristics to the
1836 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1837 patterns are not supported. Nevertheless, there are times when this kind of
1838 matching can be useful. For a discussion of the two matching algorithms, and a
1839 list of features that \fBpcre_dfa_exec()\fP does not support, see the
1840 .\" HREF
1841 \fBpcrematching\fP
1842 .\"
1843 documentation.
1844 .P
1845 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1846 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1847 different way, and this is described below. The other common arguments are used
1848 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1849 here.
1850 .P
1851 The two additional arguments provide workspace for the function. The workspace
1852 vector should contain at least 20 elements. It is used for keeping track of
1853 multiple paths through the pattern tree. More workspace will be needed for
1854 patterns and subjects where there are a lot of potential matches.
1855 .P
1856 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1857 .sp
1858 int rc;
1859 int ovector[10];
1860 int wspace[20];
1861 rc = pcre_dfa_exec(
1862 re, /* result of pcre_compile() */
1863 NULL, /* we didn't study the pattern */
1864 "some string", /* the subject string */
1865 11, /* the length of the subject string */
1866 0, /* start at offset 0 in the subject */
1867 0, /* default options */
1868 ovector, /* vector of integers for substring information */
1869 10, /* number of elements (NOT size in bytes) */
1870 wspace, /* working space vector */
1871 20); /* number of elements (NOT size in bytes) */
1872 .
1873 .SS "Option bits for \fBpcre_dfa_exec()\fP"
1874 .rs
1875 .sp
1876 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1877 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1880 four of these are exactly the same as for \fBpcre_exec()\fP, so their
1881 description is not repeated here.
1882 .sp
1885 .sp
1886 These have the same general effect as they do for \fBpcre_exec()\fP, but the
1887 details are slightly different. When PCRE_PARTIAL_HARD is set for
1888 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
1889 is reached and there is still at least one matching possibility that requires
1890 additional characters. This happens even if some complete matches have also
1891 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
1892 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
1893 there have been no complete matches, but there is still at least one matching
1894 possibility. The portion of the string that was inspected when the longest
1895 partial match was found is set as the first matching string in both cases.
1896 .sp
1898 .sp
1899 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1900 soon as it has found one match. Because of the way the alternative algorithm
1901 works, this is necessarily the shortest possible match at the first possible
1902 matching point in the subject string.
1903 .sp
1905 .sp
1906 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
1907 again, with additional subject characters, and have it continue with the same
1908 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
1909 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
1910 before because data about the match so far is left in them after a partial
1911 match. There is more discussion of this facility in the
1912 .\" HREF
1913 \fBpcrepartial\fP
1914 .\"
1915 documentation.
1916 .
1917 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1918 .rs
1919 .sp
1920 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1921 substring in the subject. Note, however, that all the matches from one run of
1922 the function start at the same point in the subject. The shorter matches are
1923 all initial substrings of the longer matches. For example, if the pattern
1924 .sp
1925 <.*>
1926 .sp
1927 is matched against the string
1928 .sp
1929 This is <something> <something else> <something further> no more
1930 .sp
1931 the three matched strings are
1932 .sp
1933 <something>
1934 <something> <something else>
1935 <something> <something else> <something further>
1936 .sp
1937 On success, the yield of the function is a number greater than zero, which is
1938 the number of matched substrings. The substrings themselves are returned in
1939 \fIovector\fP. Each string uses two elements; the first is the offset to the
1940 start, and the second is the offset to the end. In fact, all the strings have
1941 the same start offset. (Space could have been saved by giving this only once,
1942 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1943 returns data, even though the meaning of the strings is different.)
1944 .P
1945 The strings are returned in reverse order of length; that is, the longest
1946 matching string is given first. If there were too many matches to fit into
1947 \fIovector\fP, the yield of the function is zero, and the vector is filled with
1948 the longest matches.
1949 .
1950 .SS "Error returns from \fBpcre_dfa_exec()\fP"
1951 .rs
1952 .sp
1953 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1954 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1955 described
1956 .\" HTML <a href="#errorlist">
1957 .\" </a>
1958 above.
1959 .\"
1960 There are in addition the following errors that are specific to
1961 \fBpcre_dfa_exec()\fP:
1962 .sp
1964 .sp
1965 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1966 that it does not support, for instance, the use of \eC or a back reference.
1967 .sp
1969 .sp
1970 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1971 uses a back reference for the condition, or a test for recursion in a specific
1972 group. These are not supported.
1973 .sp
1975 .sp
1976 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1977 block that contains a setting of the \fImatch_limit\fP field. This is not
1978 supported (it is meaningless).
1979 .sp
1981 .sp
1982 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1983 \fIworkspace\fP vector.
1984 .sp
1986 .sp
1987 When a recursive subpattern is processed, the matching function calls itself
1988 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1989 error is given if the output vector is not large enough. This should be
1990 extremely rare, as a vector of size 1000 is used.
1991 .
1992 .
1993 .SH "SEE ALSO"
1994 .rs
1995 .sp
1996 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1997 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1998 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1999 .
2000 .
2002 .rs
2003 .sp
2004 .nf
2005 Philip Hazel
2006 University Computing Service
2007 Cambridge CB2 3QH, England.
2008 .fi
2009 .
2010 .
2012 .rs
2013 .sp
2014 .nf
2015 Last updated: 09 September 2009
2016 Copyright (c) 1997-2009 University of Cambridge.
2017 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5