/[pcre]/code/trunk/doc/pcreapi.3
ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 654 - (show annotations)
Tue Aug 2 11:00:40 2011 UTC (3 years, 9 months ago) by ph10
File size: 102880 byte(s)
Error occurred while calculating annotation data.
Documentation and general text tidies in preparation for test release.
1 .TH PCREAPI 3
2 .SH NAME
3 PCRE - Perl-compatible regular expressions
4 .SH "PCRE NATIVE API"
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 .
112 .SH "PCRE API OVERVIEW"
113 .rs
114 .sp
115 PCRE has its own native API, which is described in this document. There are
116 also some wrapper functions that correspond to the POSIX regular expression
117 API. These are described in the
118 .\" HREF
119 \fBpcreposix\fP
120 .\"
121 documentation. Both of these APIs define a set of C function calls. A C++
122 wrapper is distributed with PCRE. It is documented in the
123 .\" HREF
124 \fBpcrecpp\fP
125 .\"
126 page.
127 .P
128 The native API C function prototypes are defined in the header file
129 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132 and PCRE_MINOR to contain the major and minor release numbers for the library.
133 Applications can use these to include support for different releases of PCRE.
134 .P
135 In a Windows environment, if you want to statically link an application program
136 against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
137 including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
138 \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
139 \fB__declspec(dllimport)\fP, with unwanted results.
140 .P
141 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
142 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
143 in a Perl-compatible manner. A sample program that demonstrates the simplest
144 way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
145 source distribution. A listing of this program is given in the
146 .\" HREF
147 \fBpcredemo\fP
148 .\"
149 documentation, and the
150 .\" HREF
151 \fBpcresample\fP
152 .\"
153 documentation describes how to compile and run it.
154 .P
155 A second matching function, \fBpcre_dfa_exec()\fP, which is not
156 Perl-compatible, is also provided. This uses a different algorithm for the
157 matching. The alternative algorithm finds all possible matches (at a given
158 point in the subject), and scans the subject just once (unless there are
159 lookbehind assertions). However, this algorithm does not return captured
160 substrings. A description of the two matching algorithms and their advantages
161 and disadvantages is given in the
162 .\" HREF
163 \fBpcrematching\fP
164 .\"
165 documentation.
166 .P
167 In addition to the main compiling and matching functions, there are convenience
168 functions for extracting captured substrings from a subject string that is
169 matched by \fBpcre_exec()\fP. They are:
170 .sp
171 \fBpcre_copy_substring()\fP
172 \fBpcre_copy_named_substring()\fP
173 \fBpcre_get_substring()\fP
174 \fBpcre_get_named_substring()\fP
175 \fBpcre_get_substring_list()\fP
176 \fBpcre_get_stringnumber()\fP
177 \fBpcre_get_stringtable_entries()\fP
178 .sp
179 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
180 provided, to free the memory used for extracted strings.
181 .P
182 The function \fBpcre_maketables()\fP is used to build a set of character tables
183 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
184 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
185 specialist use. Most commonly, no special tables are passed, in which case
186 internal tables that are generated when PCRE is built are used.
187 .P
188 The function \fBpcre_fullinfo()\fP is used to find out information about a
189 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
190 some of the available information, but is retained for backwards compatibility.
191 The function \fBpcre_version()\fP returns a pointer to a string containing the
192 version of PCRE and its date of release.
193 .P
194 The function \fBpcre_refcount()\fP maintains a reference count in a data block
195 containing a compiled pattern. This is provided for the benefit of
196 object-oriented applications.
197 .P
198 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
199 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
200 respectively. PCRE calls the memory management functions via these variables,
201 so a calling program can replace them if it wishes to intercept the calls. This
202 should be done before calling any PCRE functions.
203 .P
204 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
205 indirections to memory management functions. These special functions are used
206 only when PCRE is compiled to use the heap for remembering data, instead of
207 recursive function calls, when running the \fBpcre_exec()\fP function. See the
208 .\" HREF
209 \fBpcrebuild\fP
210 .\"
211 documentation for details of how to do this. It is a non-standard way of
212 building PCRE, for use in environments that have limited stacks. Because of the
213 greater use of memory management, it runs more slowly. Separate functions are
214 provided so that special-purpose external code can be used for this case. When
215 used, these functions are always called in a stack-like manner (last obtained,
216 first freed), and always for memory blocks of the same size. There is a
217 discussion about PCRE's stack usage in the
218 .\" HREF
219 \fBpcrestack\fP
220 .\"
221 documentation.
222 .P
223 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
224 by the caller to a "callout" function, which PCRE will then call at specified
225 points during a matching operation. Details are given in the
226 .\" HREF
227 \fBpcrecallout\fP
228 .\"
229 documentation.
230 .
231 .
232 .\" HTML <a name="newlines"></a>
233 .SH NEWLINES
234 .rs
235 .sp
236 PCRE supports five different conventions for indicating line breaks in
237 strings: a single CR (carriage return) character, a single LF (linefeed)
238 character, the two-character sequence CRLF, any of the three preceding, or any
239 Unicode newline sequence. The Unicode newline sequences are the three just
240 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
241 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
242 (paragraph separator, U+2029).
243 .P
244 Each of the first three conventions is used by at least one operating system as
245 its standard newline sequence. When PCRE is built, a default can be specified.
246 The default default is LF, which is the Unix standard. When PCRE is run, the
247 default can be overridden, either when a pattern is compiled, or when it is
248 matched.
249 .P
250 At compile time, the newline convention can be specified by the \fIoptions\fP
251 argument of \fBpcre_compile()\fP, or it can be specified by special text at the
252 start of the pattern itself; this overrides any other settings. See the
253 .\" HREF
254 \fBpcrepattern\fP
255 .\"
256 page for details of the special character sequences.
257 .P
258 In the PCRE documentation the word "newline" is used to mean "the character or
259 pair of characters that indicate a line break". The choice of newline
260 convention affects the handling of the dot, circumflex, and dollar
261 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
262 recognized line ending sequence, the match position advancement for a
263 non-anchored pattern. There is more detail about this in the
264 .\" HTML <a href="#execoptions">
265 .\" </a>
266 section on \fBpcre_exec()\fP options
267 .\"
268 below.
269 .P
270 The choice of newline convention does not affect the interpretation of
271 the \en or \er escape sequences, nor does it affect what \eR matches, which is
272 controlled in a similar way, but by separate options.
273 .
274 .
275 .SH MULTITHREADING
276 .rs
277 .sp
278 The PCRE functions can be used in multi-threading applications, with the
279 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
280 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
281 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
282 .P
283 The compiled form of a regular expression is not altered during matching, so
284 the same compiled pattern can safely be used by several threads at once.
285 .
286 .
287 .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
288 .rs
289 .sp
290 The compiled form of a regular expression can be saved and re-used at a later
291 time, possibly by a different program, and even on a host other than the one on
292 which it was compiled. Details are given in the
293 .\" HREF
294 \fBpcreprecompile\fP
295 .\"
296 documentation. However, compiling a regular expression with one version of PCRE
297 for use with a different version is not guaranteed to work and may cause
298 crashes.
299 .
300 .
301 .SH "CHECKING BUILD-TIME OPTIONS"
302 .rs
303 .sp
304 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
305 .PP
306 The function \fBpcre_config()\fP makes it possible for a PCRE client to
307 discover which optional features have been compiled into the PCRE library. The
308 .\" HREF
309 \fBpcrebuild\fP
310 .\"
311 documentation has more details about these optional features.
312 .P
313 The first argument for \fBpcre_config()\fP is an integer, specifying which
314 information is required; the second argument is a pointer to a variable into
315 which the information is placed. The following information is available:
316 .sp
317 PCRE_CONFIG_UTF8
318 .sp
319 The output is an integer that is set to one if UTF-8 support is available;
320 otherwise it is set to zero.
321 .sp
322 PCRE_CONFIG_UNICODE_PROPERTIES
323 .sp
324 The output is an integer that is set to one if support for Unicode character
325 properties is available; otherwise it is set to zero.
326 .sp
327 PCRE_CONFIG_NEWLINE
328 .sp
329 The output is an integer whose value specifies the default character sequence
330 that is recognized as meaning "newline". The four values that are supported
331 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
332 Though they are derived from ASCII, the same values are returned in EBCDIC
333 environments. The default should normally correspond to the standard sequence
334 for your operating system.
335 .sp
336 PCRE_CONFIG_BSR
337 .sp
338 The output is an integer whose value indicates what character sequences the \eR
339 escape sequence matches by default. A value of 0 means that \eR matches any
340 Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
341 or CRLF. The default can be overridden when a pattern is compiled or matched.
342 .sp
343 PCRE_CONFIG_LINK_SIZE
344 .sp
345 The output is an integer that contains the number of bytes used for internal
346 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
347 allow larger regular expressions to be compiled, at the expense of slower
348 matching. The default value of 2 is sufficient for all but the most massive
349 patterns, since it allows the compiled pattern to be up to 64K in size.
350 .sp
351 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
352 .sp
353 The output is an integer that contains the threshold above which the POSIX
354 interface uses \fBmalloc()\fP for output vectors. Further details are given in
355 the
356 .\" HREF
357 \fBpcreposix\fP
358 .\"
359 documentation.
360 .sp
361 PCRE_CONFIG_MATCH_LIMIT
362 .sp
363 The output is a long integer that gives the default limit for the number of
364 internal matching function calls in a \fBpcre_exec()\fP execution. Further
365 details are given with \fBpcre_exec()\fP below.
366 .sp
367 PCRE_CONFIG_MATCH_LIMIT_RECURSION
368 .sp
369 The output is a long integer that gives the default limit for the depth of
370 recursion when calling the internal matching function in a \fBpcre_exec()\fP
371 execution. Further details are given with \fBpcre_exec()\fP below.
372 .sp
373 PCRE_CONFIG_STACKRECURSE
374 .sp
375 The output is an integer that is set to one if internal recursion when running
376 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
377 to remember their state. This is the usual way that PCRE is compiled. The
378 output is zero if PCRE was compiled to use blocks of data on the heap instead
379 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
380 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
381 avoiding the use of the stack.
382 .
383 .
384 .SH "COMPILING A PATTERN"
385 .rs
386 .sp
387 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
388 .ti +5n
389 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
390 .ti +5n
391 .B const unsigned char *\fItableptr\fP);
392 .sp
393 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
394 .ti +5n
395 .B int *\fIerrorcodeptr\fP,
396 .ti +5n
397 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
398 .ti +5n
399 .B const unsigned char *\fItableptr\fP);
400 .P
401 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
402 called to compile a pattern into an internal form. The only difference between
403 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
404 \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
405 too much repetition, we refer just to \fBpcre_compile()\fP below, but the
406 information applies equally to \fBpcre_compile2()\fP.
407 .P
408 The pattern is a C string terminated by a binary zero, and is passed in the
409 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
410 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
411 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
412 for a structure whose contents are not externally defined. It is up to the
413 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
414 .P
415 Although the compiled code of a PCRE regex is relocatable, that is, it does not
416 depend on memory location, the complete \fBpcre\fP data block is not
417 fully relocatable, because it may contain a copy of the \fItableptr\fP
418 argument, which is an address (see below).
419 .P
420 The \fIoptions\fP argument contains various bit settings that affect the
421 compilation. It should be zero if no options are required. The available
422 options are described below. Some of them (in particular, those that are
423 compatible with Perl, but some others as well) can also be set and unset from
424 within the pattern (see the detailed description in the
425 .\" HREF
426 \fBpcrepattern\fP
427 .\"
428 documentation). For those options that can be different in different parts of
429 the pattern, the contents of the \fIoptions\fP argument specifies their
430 settings at the start of compilation and execution. The PCRE_ANCHORED,
431 PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
432 PCRE_NO_START_OPT options can be set at the time of matching as well as at
433 compile time.
434 .P
435 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
436 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
437 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
438 error message. This is a static string that is part of the library. You must
439 not try to free it. Normally, the offset from the start of the pattern to the
440 byte that was being processed when the error was discovered is placed in the
441 variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
442 immediate error is given). However, for an invalid UTF-8 string, the offset is
443 that of the first byte of the failing character. Also, some errors are not
444 detected until checks are carried out when the whole pattern has been scanned;
445 in these cases the offset passed back is the length of the pattern.
446 .P
447 Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
448 sometimes point into the middle of a UTF-8 character.
449 .P
450 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
451 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
452 returned via this argument in the event of an error. This is in addition to the
453 textual error message. Error codes and messages are listed below.
454 .P
455 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
456 character tables that are built when PCRE is compiled, using the default C
457 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
458 call to \fBpcre_maketables()\fP. This value is stored with the compiled
459 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
460 passed to it. For more discussion, see the section on locale support below.
461 .P
462 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
463 .sp
464 pcre *re;
465 const char *error;
466 int erroffset;
467 re = pcre_compile(
468 "^A.*Z", /* the pattern */
469 0, /* default options */
470 &error, /* for error message */
471 &erroffset, /* for error offset */
472 NULL); /* use default character tables */
473 .sp
474 The following names for option bits are defined in the \fBpcre.h\fP header
475 file:
476 .sp
477 PCRE_ANCHORED
478 .sp
479 If this bit is set, the pattern is forced to be "anchored", that is, it is
480 constrained to match only at the first matching point in the string that is
481 being searched (the "subject string"). This effect can also be achieved by
482 appropriate constructs in the pattern itself, which is the only way to do it in
483 Perl.
484 .sp
485 PCRE_AUTO_CALLOUT
486 .sp
487 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
488 all with number 255, before each pattern item. For discussion of the callout
489 facility, see the
490 .\" HREF
491 \fBpcrecallout\fP
492 .\"
493 documentation.
494 .sp
495 PCRE_BSR_ANYCRLF
496 PCRE_BSR_UNICODE
497 .sp
498 These options (which are mutually exclusive) control what the \eR escape
499 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
500 match any Unicode newline sequence. The default is specified when PCRE is
501 built. It can be overridden from within the pattern, or by setting an option
502 when a compiled pattern is matched.
503 .sp
504 PCRE_CASELESS
505 .sp
506 If this bit is set, letters in the pattern match both upper and lower case
507 letters. It is equivalent to Perl's /i option, and it can be changed within a
508 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
509 concept of case for characters whose values are less than 128, so caseless
510 matching is always possible. For characters with higher values, the concept of
511 case is supported if PCRE is compiled with Unicode property support, but not
512 otherwise. If you want to use caseless matching for characters 128 and above,
513 you must ensure that PCRE is compiled with Unicode property support as well as
514 with UTF-8 support.
515 .sp
516 PCRE_DOLLAR_ENDONLY
517 .sp
518 If this bit is set, a dollar metacharacter in the pattern matches only at the
519 end of the subject string. Without this option, a dollar also matches
520 immediately before a newline at the end of the string (but not before any other
521 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
522 There is no equivalent to this option in Perl, and no way to set it within a
523 pattern.
524 .sp
525 PCRE_DOTALL
526 .sp
527 If this bit is set, a dot metacharacter in the pattern matches a character of
528 any value, including one that indicates a newline. However, it only ever
529 matches one character, even if newlines are coded as CRLF. Without this option,
530 a dot does not match when the current position is at a newline. This option is
531 equivalent to Perl's /s option, and it can be changed within a pattern by a
532 (?s) option setting. A negative class such as [^a] always matches newline
533 characters, independent of the setting of this option.
534 .sp
535 PCRE_DUPNAMES
536 .sp
537 If this bit is set, names used to identify capturing subpatterns need not be
538 unique. This can be helpful for certain types of pattern when it is known that
539 only one instance of the named subpattern can ever be matched. There are more
540 details of named subpatterns below; see also the
541 .\" HREF
542 \fBpcrepattern\fP
543 .\"
544 documentation.
545 .sp
546 PCRE_EXTENDED
547 .sp
548 If this bit is set, whitespace data characters in the pattern are totally
549 ignored except when escaped or inside a character class. Whitespace does not
550 include the VT character (code 11). In addition, characters between an
551 unescaped # outside a character class and the next newline, inclusive, are also
552 ignored. This is equivalent to Perl's /x option, and it can be changed within a
553 pattern by a (?x) option setting.
554 .P
555 Which characters are interpreted as newlines is controlled by the options
556 passed to \fBpcre_compile()\fP or by a special sequence at the start of the
557 pattern, as described in the section entitled
558 .\" HTML <a href="pcrepattern.html#newlines">
559 .\" </a>
560 "Newline conventions"
561 .\"
562 in the \fBpcrepattern\fP documentation. Note that the end of this type of
563 comment is a literal newline sequence in the pattern; escape sequences that
564 happen to represent a newline do not count.
565 .P
566 This option makes it possible to include comments inside complicated patterns.
567 Note, however, that this applies only to data characters. Whitespace characters
568 may never appear within special character sequences in a pattern, for example
569 within the sequence (?( that introduces a conditional subpattern.
570 .sp
571 PCRE_EXTRA
572 .sp
573 This option was invented in order to turn on additional functionality of PCRE
574 that is incompatible with Perl, but it is currently of very little use. When
575 set, any backslash in a pattern that is followed by a letter that has no
576 special meaning causes an error, thus reserving these combinations for future
577 expansion. By default, as in Perl, a backslash followed by a letter with no
578 special meaning is treated as a literal. (Perl can, however, be persuaded to
579 give an error for this, by running it with the -w option.) There are at present
580 no other features controlled by this option. It can also be set by a (?X)
581 option setting within a pattern.
582 .sp
583 PCRE_FIRSTLINE
584 .sp
585 If this option is set, an unanchored pattern is required to match before or at
586 the first newline in the subject string, though the matched text may continue
587 over the newline.
588 .sp
589 PCRE_JAVASCRIPT_COMPAT
590 .sp
591 If this option is set, PCRE's behaviour is changed in some ways so that it is
592 compatible with JavaScript rather than Perl. The changes are as follows:
593 .P
594 (1) A lone closing square bracket in a pattern causes a compile-time error,
595 because this is illegal in JavaScript (by default it is treated as a data
596 character). Thus, the pattern AB]CD becomes illegal when this option is set.
597 .P
598 (2) At run time, a back reference to an unset subpattern group matches an empty
599 string (by default this causes the current matching alternative to fail). A
600 pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
601 an "a" in the subject), whereas it fails by default, for Perl compatibility.
602 .sp
603 PCRE_MULTILINE
604 .sp
605 By default, PCRE treats the subject string as consisting of a single line of
606 characters (even if it actually contains newlines). The "start of line"
607 metacharacter (^) matches only at the start of the string, while the "end of
608 line" metacharacter ($) matches only at the end of the string, or before a
609 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
610 Perl.
611 .P
612 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
613 match immediately following or immediately before internal newlines in the
614 subject string, respectively, as well as at the very start and end. This is
615 equivalent to Perl's /m option, and it can be changed within a pattern by a
616 (?m) option setting. If there are no newlines in a subject string, or no
617 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
618 .sp
619 PCRE_NEWLINE_CR
620 PCRE_NEWLINE_LF
621 PCRE_NEWLINE_CRLF
622 PCRE_NEWLINE_ANYCRLF
623 PCRE_NEWLINE_ANY
624 .sp
625 These options override the default newline definition that was chosen when PCRE
626 was built. Setting the first or the second specifies that a newline is
627 indicated by a single character (CR or LF, respectively). Setting
628 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
629 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
630 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
631 that any Unicode newline sequence should be recognized. The Unicode newline
632 sequences are the three just mentioned, plus the single characters VT (vertical
633 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
634 separator, U+2028), and PS (paragraph separator, U+2029). The last two are
635 recognized only in UTF-8 mode.
636 .P
637 The newline setting in the options word uses three bits that are treated
638 as a number, giving eight possibilities. Currently only six are used (default
639 plus the five values above). This means that if you set more than one newline
640 option, the combination may or may not be sensible. For example,
641 PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
642 other combinations may yield unused numbers and cause an error.
643 .P
644 The only time that a line break in a pattern is specially recognized when
645 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
646 and so are ignored in this mode. Also, an unescaped # outside a character class
647 indicates a comment that lasts until after the next line break sequence. In
648 other circumstances, line break sequences in patterns are treated as literal
649 data.
650 .P
651 The newline option that is set at compile time becomes the default that is used
652 for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
653 .sp
654 PCRE_NO_AUTO_CAPTURE
655 .sp
656 If this option is set, it disables the use of numbered capturing parentheses in
657 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
658 were followed by ?: but named parentheses can still be used for capturing (and
659 they acquire numbers in the usual way). There is no equivalent of this option
660 in Perl.
661 .sp
662 NO_START_OPTIMIZE
663 .sp
664 This is an option that acts at matching time; that is, it is really an option
665 for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
666 it is remembered with the compiled pattern and assumed at matching time. For
667 details see the discussion of PCRE_NO_START_OPTIMIZE
668 .\" HTML <a href="#execoptions">
669 .\" </a>
670 below.
671 .\"
672 .sp
673 PCRE_UCP
674 .sp
675 This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
676 \ew, and some of the POSIX character classes. By default, only ASCII characters
677 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
678 classify characters. More details are given in the section on
679 .\" HTML <a href="pcre.html#genericchartypes">
680 .\" </a>
681 generic character types
682 .\"
683 in the
684 .\" HREF
685 \fBpcrepattern\fP
686 .\"
687 page. If you set PCRE_UCP, matching one of the items it affects takes much
688 longer. The option is available only if PCRE has been compiled with Unicode
689 property support.
690 .sp
691 PCRE_UNGREEDY
692 .sp
693 This option inverts the "greediness" of the quantifiers so that they are not
694 greedy by default, but become greedy if followed by "?". It is not compatible
695 with Perl. It can also be set by a (?U) option setting within the pattern.
696 .sp
697 PCRE_UTF8
698 .sp
699 This option causes PCRE to regard both the pattern and the subject as strings
700 of UTF-8 characters instead of single-byte character strings. However, it is
701 available only when PCRE is built to include UTF-8 support. If not, the use
702 of this option provokes an error. Details of how this option changes the
703 behaviour of PCRE are given in the
704 .\" HTML <a href="pcre.html#utf8support">
705 .\" </a>
706 section on UTF-8 support
707 .\"
708 in the main
709 .\" HREF
710 \fBpcre\fP
711 .\"
712 page.
713 .sp
714 PCRE_NO_UTF8_CHECK
715 .sp
716 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
717 automatically checked. There is a discussion about the
718 .\" HTML <a href="pcre.html#utf8strings">
719 .\" </a>
720 validity of UTF-8 strings
721 .\"
722 in the main
723 .\" HREF
724 \fBpcre\fP
725 .\"
726 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
727 returns an error. If you already know that your pattern is valid, and you want
728 to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
729 option. When it is set, the effect of passing an invalid UTF-8 string as a
730 pattern is undefined. It may cause your program to crash. Note that this option
731 can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
732 the UTF-8 validity checking of subject strings.
733 .
734 .
735 .SH "COMPILATION ERROR CODES"
736 .rs
737 .sp
738 The following table lists the error codes than may be returned by
739 \fBpcre_compile2()\fP, along with the error messages that may be returned by
740 both compiling functions. As PCRE has developed, some error codes have fallen
741 out of use. To avoid confusion, they have not been re-used.
742 .sp
743 0 no error
744 1 \e at end of pattern
745 2 \ec at end of pattern
746 3 unrecognized character follows \e
747 4 numbers out of order in {} quantifier
748 5 number too big in {} quantifier
749 6 missing terminating ] for character class
750 7 invalid escape sequence in character class
751 8 range out of order in character class
752 9 nothing to repeat
753 10 [this code is not in use]
754 11 internal error: unexpected repeat
755 12 unrecognized character after (? or (?-
756 13 POSIX named classes are supported only within a class
757 14 missing )
758 15 reference to non-existent subpattern
759 16 erroffset passed as NULL
760 17 unknown option bit(s) set
761 18 missing ) after comment
762 19 [this code is not in use]
763 20 regular expression is too large
764 21 failed to get memory
765 22 unmatched parentheses
766 23 internal error: code overflow
767 24 unrecognized character after (?<
768 25 lookbehind assertion is not fixed length
769 26 malformed number or name after (?(
770 27 conditional group contains more than two branches
771 28 assertion expected after (?(
772 29 (?R or (?[+-]digits must be followed by )
773 30 unknown POSIX class name
774 31 POSIX collating elements are not supported
775 32 this version of PCRE is not compiled with PCRE_UTF8 support
776 33 [this code is not in use]
777 34 character value in \ex{...} sequence is too large
778 35 invalid condition (?(0)
779 36 \eC not allowed in lookbehind assertion
780 37 PCRE does not support \eL, \el, \eN, \eU, or \eu
781 38 number after (?C is > 255
782 39 closing ) for (?C expected
783 40 recursive call could loop indefinitely
784 41 unrecognized character after (?P
785 42 syntax error in subpattern name (missing terminator)
786 43 two named subpatterns have the same name
787 44 invalid UTF-8 string
788 45 support for \eP, \ep, and \eX has not been compiled
789 46 malformed \eP or \ep sequence
790 47 unknown property name after \eP or \ep
791 48 subpattern name is too long (maximum 32 characters)
792 49 too many named subpatterns (maximum 10000)
793 50 [this code is not in use]
794 51 octal value is greater than \e377 (not in UTF-8 mode)
795 52 internal error: overran compiling workspace
796 53 internal error: previously-checked referenced subpattern
797 not found
798 54 DEFINE group contains more than one branch
799 55 repeating a DEFINE group is not allowed
800 56 inconsistent NEWLINE options
801 57 \eg is not followed by a braced, angle-bracketed, or quoted
802 name/number or by a plain number
803 58 a numbered reference must not be zero
804 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
805 60 (*VERB) not recognized
806 61 number is too big
807 62 subpattern name expected
808 63 digit expected after (?+
809 64 ] is an invalid data character in JavaScript compatibility mode
810 65 different names for subpatterns of the same number are
811 not allowed
812 66 (*MARK) must have an argument
813 67 this version of PCRE is not compiled with PCRE_UCP support
814 .sp
815 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
816 be used if the limits were changed when PCRE was built.
817 .
818 .
819 .SH "STUDYING A PATTERN"
820 .rs
821 .sp
822 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
823 .ti +5n
824 .B const char **\fIerrptr\fP);
825 .PP
826 If a compiled pattern is going to be used several times, it is worth spending
827 more time analyzing it in order to speed up the time taken for matching. The
828 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
829 argument. If studying the pattern produces additional information that will
830 help speed up matching, \fBpcre_study()\fP returns a pointer to a
831 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
832 results of the study.
833 .P
834 The returned value from \fBpcre_study()\fP can be passed directly to
835 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
836 also contains other fields that can be set by the caller before the block is
837 passed; these are described
838 .\" HTML <a href="#extradata">
839 .\" </a>
840 below
841 .\"
842 in the section on matching a pattern.
843 .P
844 If studying the pattern does not produce any useful information,
845 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
846 wants to pass any of the other fields to \fBpcre_exec()\fP or
847 \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
848 .P
849 The second argument of \fBpcre_study()\fP contains option bits. At present, no
850 options are defined, and this argument should always be zero.
851 .P
852 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
853 studying succeeds (even if no data is returned), the variable it points to is
854 set to NULL. Otherwise it is set to point to a textual error message. This is a
855 static string that is part of the library. You must not try to free it. You
856 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
857 sure that it has run successfully.
858 .P
859 This is a typical call to \fBpcre_study\fP():
860 .sp
861 pcre_extra *pe;
862 pe = pcre_study(
863 re, /* result of pcre_compile() */
864 0, /* no options exist */
865 &error); /* set to NULL or points to a message */
866 .sp
867 Studying a pattern does two things: first, a lower bound for the length of
868 subject string that is needed to match the pattern is computed. This does not
869 mean that there are any strings of that length that match, but it does
870 guarantee that no shorter strings match. The value is used by
871 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
872 match strings that are shorter than the lower bound. You can find out the value
873 in a calling program via the \fBpcre_fullinfo()\fP function.
874 .P
875 Studying a pattern is also useful for non-anchored patterns that do not have a
876 single fixed starting character. A bitmap of possible starting bytes is
877 created. This speeds up finding a position in the subject at which to start
878 matching.
879 .P
880 The two optimizations just described can be disabled by setting the
881 PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
882 \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
883 callouts or (*MARK), and you want to make use of these facilities in cases
884 where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
885 .\" HTML <a href="#execoptions">
886 .\" </a>
887 below.
888 .\"
889 .
890 .
891 .\" HTML <a name="localesupport"></a>
892 .SH "LOCALE SUPPORT"
893 .rs
894 .sp
895 PCRE handles caseless matching, and determines whether characters are letters,
896 digits, or whatever, by reference to a set of tables, indexed by character
897 value. When running in UTF-8 mode, this applies only to characters with codes
898 less than 128. By default, higher-valued codes never match escapes such as \ew
899 or \ed, but they can be tested with \ep if PCRE is built with Unicode character
900 property support. Alternatively, the PCRE_UCP option can be set at compile
901 time; this causes \ew and friends to use Unicode property support instead of
902 built-in tables. The use of locales with Unicode is discouraged. If you are
903 handling characters with codes greater than 128, you should either use UTF-8
904 and Unicode, or use locales, but not try to mix the two.
905 .P
906 PCRE contains an internal set of tables that are used when the final argument
907 of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
908 Normally, the internal tables recognize only ASCII characters. However, when
909 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
910 default "C" locale of the local system, which may cause them to be different.
911 .P
912 The internal tables can always be overridden by tables supplied by the
913 application that calls PCRE. These may be created in a different locale from
914 the default. As more and more applications change to using Unicode, the need
915 for this locale support is expected to die away.
916 .P
917 External tables are built by calling the \fBpcre_maketables()\fP function,
918 which has no arguments, in the relevant locale. The result can then be passed
919 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
920 example, to build and use tables that are appropriate for the French locale
921 (where accented characters with values greater than 128 are treated as letters),
922 the following code could be used:
923 .sp
924 setlocale(LC_CTYPE, "fr_FR");
925 tables = pcre_maketables();
926 re = pcre_compile(..., tables);
927 .sp
928 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
929 are using Windows, the name for the French locale is "french".
930 .P
931 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
932 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
933 that the memory containing the tables remains available for as long as it is
934 needed.
935 .P
936 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
937 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
938 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
939 pattern, compilation, studying and matching all happen in the same locale, but
940 different patterns can be compiled in different locales.
941 .P
942 It is possible to pass a table pointer or NULL (indicating the use of the
943 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
944 this facility could be used to match a pattern in a different locale from the
945 one in which it was compiled. Passing table pointers at run time is discussed
946 below in the section on matching a pattern.
947 .
948 .
949 .\" HTML <a name="infoaboutpattern"></a>
950 .SH "INFORMATION ABOUT A PATTERN"
951 .rs
952 .sp
953 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
954 .ti +5n
955 .B int \fIwhat\fP, void *\fIwhere\fP);
956 .PP
957 The \fBpcre_fullinfo()\fP function returns information about a compiled
958 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
959 nevertheless retained for backwards compability (and is documented below).
960 .P
961 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
962 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
963 the pattern was not studied. The third argument specifies which piece of
964 information is required, and the fourth argument is a pointer to a variable
965 to receive the data. The yield of the function is zero for success, or one of
966 the following negative numbers:
967 .sp
968 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
969 the argument \fIwhere\fP was NULL
970 PCRE_ERROR_BADMAGIC the "magic number" was not found
971 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
972 .sp
973 The "magic number" is placed at the start of each compiled pattern as an simple
974 check against passing an arbitrary memory pointer. Here is a typical call of
975 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
976 .sp
977 int rc;
978 size_t length;
979 rc = pcre_fullinfo(
980 re, /* result of pcre_compile() */
981 pe, /* result of pcre_study(), or NULL */
982 PCRE_INFO_SIZE, /* what is required */
983 &length); /* where to put the data */
984 .sp
985 The possible values for the third argument are defined in \fBpcre.h\fP, and are
986 as follows:
987 .sp
988 PCRE_INFO_BACKREFMAX
989 .sp
990 Return the number of the highest back reference in the pattern. The fourth
991 argument should point to an \fBint\fP variable. Zero is returned if there are
992 no back references.
993 .sp
994 PCRE_INFO_CAPTURECOUNT
995 .sp
996 Return the number of capturing subpatterns in the pattern. The fourth argument
997 should point to an \fBint\fP variable.
998 .sp
999 PCRE_INFO_DEFAULT_TABLES
1000 .sp
1001 Return a pointer to the internal default character tables within PCRE. The
1002 fourth argument should point to an \fBunsigned char *\fP variable. This
1003 information call is provided for internal use by the \fBpcre_study()\fP
1004 function. External callers can cause PCRE to use its internal tables by passing
1005 a NULL table pointer.
1006 .sp
1007 PCRE_INFO_FIRSTBYTE
1008 .sp
1009 Return information about the first byte of any matched string, for a
1010 non-anchored pattern. The fourth argument should point to an \fBint\fP
1011 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1012 still recognized for backwards compatibility.)
1013 .P
1014 If there is a fixed first byte, for example, from a pattern such as
1015 (cat|cow|coyote), its value is returned. Otherwise, if either
1016 .sp
1017 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1018 starts with "^", or
1019 .sp
1020 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1021 (if it were set, the pattern would be anchored),
1022 .sp
1023 -1 is returned, indicating that the pattern matches only at the start of a
1024 subject string or after any newline within the string. Otherwise -2 is
1025 returned. For anchored patterns, -2 is returned.
1026 .sp
1027 PCRE_INFO_FIRSTTABLE
1028 .sp
1029 If the pattern was studied, and this resulted in the construction of a 256-bit
1030 table indicating a fixed set of bytes for the first byte in any matching
1031 string, a pointer to the table is returned. Otherwise NULL is returned. The
1032 fourth argument should point to an \fBunsigned char *\fP variable.
1033 .sp
1034 PCRE_INFO_HASCRORLF
1035 .sp
1036 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1037 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1038 explicit match is either a literal CR or LF character, or \er or \en.
1039 .sp
1040 PCRE_INFO_JCHANGED
1041 .sp
1042 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1043 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1044 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1045 .sp
1046 PCRE_INFO_LASTLITERAL
1047 .sp
1048 Return the value of the rightmost literal byte that must exist in any matched
1049 string, other than at its start, if such a byte has been recorded. The fourth
1050 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1051 returned. For anchored patterns, a last literal byte is recorded only if it
1052 follows something of variable length. For example, for the pattern
1053 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1054 is -1.
1055 .sp
1056 PCRE_INFO_MINLENGTH
1057 .sp
1058 If the pattern was studied and a minimum length for matching subject strings
1059 was computed, its value is returned. Otherwise the returned value is -1. The
1060 value is a number of characters, not bytes (this may be relevant in UTF-8
1061 mode). The fourth argument should point to an \fBint\fP variable. A
1062 non-negative value is a lower bound to the length of any matching string. There
1063 may not be any strings of that length that do actually match, but every string
1064 that does match is at least that long.
1065 .sp
1066 PCRE_INFO_NAMECOUNT
1067 PCRE_INFO_NAMEENTRYSIZE
1068 PCRE_INFO_NAMETABLE
1069 .sp
1070 PCRE supports the use of named as well as numbered capturing parentheses. The
1071 names are just an additional way of identifying the parentheses, which still
1072 acquire numbers. Several convenience functions such as
1073 \fBpcre_get_named_substring()\fP are provided for extracting captured
1074 substrings by name. It is also possible to extract the data directly, by first
1075 converting the name to a number in order to access the correct pointers in the
1076 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1077 you need to use the name-to-number map, which is described by these three
1078 values.
1079 .P
1080 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1081 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1082 entry; both of these return an \fBint\fP value. The entry size depends on the
1083 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1084 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1085 are the number of the capturing parenthesis, most significant byte first. The
1086 rest of the entry is the corresponding name, zero terminated.
1087 .P
1088 The names are in alphabetical order. Duplicate names may appear if (?| is used
1089 to create multiple groups with the same number, as described in the
1090 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1091 .\" </a>
1092 section on duplicate subpattern numbers
1093 .\"
1094 in the
1095 .\" HREF
1096 \fBpcrepattern\fP
1097 .\"
1098 page. Duplicate names for subpatterns with different numbers are permitted only
1099 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1100 table in the order in which they were found in the pattern. In the absence of
1101 (?| this is the order of increasing number; when (?| is used this is not
1102 necessarily the case because later subpatterns may have lower numbers.
1103 .P
1104 As a simple example of the name/number table, consider the following pattern
1105 (assume PCRE_EXTENDED is set, so white space - including newlines - is
1106 ignored):
1107 .sp
1108 .\" JOIN
1109 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1110 (?<month>\ed\ed) - (?<day>\ed\ed) )
1111 .sp
1112 There are four named subpatterns, so the table has four entries, and each entry
1113 in the table is eight bytes long. The table is as follows, with non-printing
1114 bytes shows in hexadecimal, and undefined bytes shown as ??:
1115 .sp
1116 00 01 d a t e 00 ??
1117 00 05 d a y 00 ?? ??
1118 00 04 m o n t h 00
1119 00 02 y e a r 00 ??
1120 .sp
1121 When writing code to extract data from named subpatterns using the
1122 name-to-number map, remember that the length of the entries is likely to be
1123 different for each compiled pattern.
1124 .sp
1125 PCRE_INFO_OKPARTIAL
1126 .sp
1127 Return 1 if the pattern can be used for partial matching with
1128 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1129 \fBint\fP variable. From release 8.00, this always returns 1, because the
1130 restrictions that previously applied to partial matching have been lifted. The
1131 .\" HREF
1132 \fBpcrepartial\fP
1133 .\"
1134 documentation gives details of partial matching.
1135 .sp
1136 PCRE_INFO_OPTIONS
1137 .sp
1138 Return a copy of the options with which the pattern was compiled. The fourth
1139 argument should point to an \fBunsigned long int\fP variable. These option bits
1140 are those specified in the call to \fBpcre_compile()\fP, modified by any
1141 top-level option settings at the start of the pattern itself. In other words,
1142 they are the options that will be in force when matching starts. For example,
1143 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1144 result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1145 .P
1146 A pattern is automatically anchored by PCRE if all of its top-level
1147 alternatives begin with one of the following:
1148 .sp
1149 ^ unless PCRE_MULTILINE is set
1150 \eA always
1151 \eG always
1152 .\" JOIN
1153 .* if PCRE_DOTALL is set and there are no back
1154 references to the subpattern in which .* appears
1155 .sp
1156 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1157 \fBpcre_fullinfo()\fP.
1158 .sp
1159 PCRE_INFO_SIZE
1160 .sp
1161 Return the size of the compiled pattern, that is, the value that was passed as
1162 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1163 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1164 variable.
1165 .sp
1166 PCRE_INFO_STUDYSIZE
1167 .sp
1168 Return the size of the data block pointed to by the \fIstudy_data\fP field in
1169 a \fBpcre_extra\fP block. That is, it is the value that was passed to
1170 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1171 created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no
1172 study data, zero is returned. The fourth argument should point to a
1173 \fBsize_t\fP variable.
1174 .
1175 .
1176 .SH "OBSOLETE INFO FUNCTION"
1177 .rs
1178 .sp
1179 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1180 .B *\fIfirstcharptr\fP);
1181 .PP
1182 The \fBpcre_info()\fP function is now obsolete because its interface is too
1183 restrictive to return all the available data about a compiled pattern. New
1184 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1185 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1186 following negative numbers:
1187 .sp
1188 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1189 PCRE_ERROR_BADMAGIC the "magic number" was not found
1190 .sp
1191 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1192 pattern was compiled is placed in the integer it points to (see
1193 PCRE_INFO_OPTIONS above).
1194 .P
1195 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1196 it is used to pass back information about the first character of any matched
1197 string (see PCRE_INFO_FIRSTBYTE above).
1198 .
1199 .
1200 .SH "REFERENCE COUNTS"
1201 .rs
1202 .sp
1203 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1204 .PP
1205 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1206 data block that contains a compiled pattern. It is provided for the benefit of
1207 applications that operate in an object-oriented manner, where different parts
1208 of the application may be using the same compiled pattern, but you want to free
1209 the block when they are all done.
1210 .P
1211 When a pattern is compiled, the reference count field is initialized to zero.
1212 It is changed only by calling this function, whose action is to add the
1213 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1214 function is the new value. However, the value of the count is constrained to
1215 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1216 it is forced to the appropriate limit value.
1217 .P
1218 Except when it is zero, the reference count is not correctly preserved if a
1219 pattern is compiled on one host and then transferred to a host whose byte-order
1220 is different. (This seems a highly unlikely scenario.)
1221 .
1222 .
1223 .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1224 .rs
1225 .sp
1226 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1227 .ti +5n
1228 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1229 .ti +5n
1230 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1231 .P
1232 The function \fBpcre_exec()\fP is called to match a subject string against a
1233 compiled pattern, which is passed in the \fIcode\fP argument. If the
1234 pattern was studied, the result of the study should be passed in the
1235 \fIextra\fP argument. This function is the main matching facility of the
1236 library, and it operates in a Perl-like manner. For specialist use there is
1237 also an alternative matching function, which is described
1238 .\" HTML <a href="#dfamatch">
1239 .\" </a>
1240 below
1241 .\"
1242 in the section about the \fBpcre_dfa_exec()\fP function.
1243 .P
1244 In most applications, the pattern will have been compiled (and optionally
1245 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1246 possible to save compiled patterns and study data, and then use them later
1247 in different processes, possibly even on different hosts. For a discussion
1248 about this, see the
1249 .\" HREF
1250 \fBpcreprecompile\fP
1251 .\"
1252 documentation.
1253 .P
1254 Here is an example of a simple call to \fBpcre_exec()\fP:
1255 .sp
1256 int rc;
1257 int ovector[30];
1258 rc = pcre_exec(
1259 re, /* result of pcre_compile() */
1260 NULL, /* we didn't study the pattern */
1261 "some string", /* the subject string */
1262 11, /* the length of the subject string */
1263 0, /* start at offset 0 in the subject */
1264 0, /* default options */
1265 ovector, /* vector of integers for substring information */
1266 30); /* number of elements (NOT size in bytes) */
1267 .
1268 .\" HTML <a name="extradata"></a>
1269 .SS "Extra data for \fBpcre_exec()\fR"
1270 .rs
1271 .sp
1272 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1273 data block. The \fBpcre_study()\fP function returns such a block (when it
1274 doesn't return NULL), but you can also create one for yourself, and pass
1275 additional information in it. The \fBpcre_extra\fP block contains the following
1276 fields (not necessarily in this order):
1277 .sp
1278 unsigned long int \fIflags\fP;
1279 void *\fIstudy_data\fP;
1280 unsigned long int \fImatch_limit\fP;
1281 unsigned long int \fImatch_limit_recursion\fP;
1282 void *\fIcallout_data\fP;
1283 const unsigned char *\fItables\fP;
1284 unsigned char **\fImark\fP;
1285 .sp
1286 The \fIflags\fP field is a bitmap that specifies which of the other fields
1287 are set. The flag bits are:
1288 .sp
1289 PCRE_EXTRA_STUDY_DATA
1290 PCRE_EXTRA_MATCH_LIMIT
1291 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1292 PCRE_EXTRA_CALLOUT_DATA
1293 PCRE_EXTRA_TABLES
1294 PCRE_EXTRA_MARK
1295 .sp
1296 Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1297 \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1298 the appropriate flag bit. You should not set this yourself, but you may add to
1299 the block by setting the other fields and their corresponding flag bits.
1300 .P
1301 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1302 vast amount of resources when running patterns that are not going to match,
1303 but which have a very large number of possibilities in their search trees. The
1304 classic example is a pattern that uses nested unlimited repeats.
1305 .P
1306 Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1307 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1308 number of times this function is called during a match, which has the effect of
1309 limiting the amount of backtracking that can take place. For patterns that are
1310 not anchored, the count restarts from zero for each position in the subject
1311 string.
1312 .P
1313 The default value for the limit can be set when PCRE is built; the default
1314 default is 10 million, which handles all but the most extreme cases. You can
1315 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1316 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1317 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1318 PCRE_ERROR_MATCHLIMIT.
1319 .P
1320 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1321 instead of limiting the total number of times that \fBmatch()\fP is called, it
1322 limits the depth of recursion. The recursion depth is a smaller number than the
1323 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1324 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1325 .P
1326 Limiting the recursion depth limits the amount of stack that can be used, or,
1327 when PCRE has been compiled to use memory on the heap instead of the stack, the
1328 amount of heap memory that can be used.
1329 .P
1330 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1331 built; the default default is the same value as the default for
1332 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1333 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1334 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1335 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1336 .P
1337 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1338 and is described in the
1339 .\" HREF
1340 \fBpcrecallout\fP
1341 .\"
1342 documentation.
1343 .P
1344 The \fItables\fP field is used to pass a character tables pointer to
1345 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1346 pattern. A non-NULL value is stored with the compiled pattern only if custom
1347 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1348 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1349 internal tables to be used. This facility is helpful when re-using patterns
1350 that have been saved after compiling with an external set of tables, because
1351 the external tables might be at a different address when \fBpcre_exec()\fP is
1352 called. See the
1353 .\" HREF
1354 \fBpcreprecompile\fP
1355 .\"
1356 documentation for a discussion of saving compiled patterns for later use.
1357 .P
1358 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1359 be set to point to a \fBchar *\fP variable. If the pattern contains any
1360 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1361 a name to pass back, a pointer to the name string (zero terminated) is placed
1362 in the variable pointed to by the \fImark\fP field. The names are within the
1363 compiled pattern; if you wish to retain such a name you must copy it before
1364 freeing the memory of a compiled pattern. If there is no name to pass back, the
1365 variable pointed to by the \fImark\fP field set to NULL. For details of the
1366 backtracking control verbs, see the section entitled
1367 .\" HTML <a href="pcrepattern#backtrackcontrol">
1368 .\" </a>
1369 "Backtracking control"
1370 .\"
1371 in the
1372 .\" HREF
1373 \fBpcrepattern\fP
1374 .\"
1375 documentation.
1376 .
1377 .
1378 .\" HTML <a name="execoptions"></a>
1379 .SS "Option bits for \fBpcre_exec()\fP"
1380 .rs
1381 .sp
1382 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1383 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1384 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1385 PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1386 PCRE_PARTIAL_HARD.
1387 .sp
1388 PCRE_ANCHORED
1389 .sp
1390 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1391 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1392 to be anchored by virtue of its contents, it cannot be made unachored at
1393 matching time.
1394 .sp
1395 PCRE_BSR_ANYCRLF
1396 PCRE_BSR_UNICODE
1397 .sp
1398 These options (which are mutually exclusive) control what the \eR escape
1399 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1400 match any Unicode newline sequence. These options override the choice that was
1401 made or defaulted when the pattern was compiled.
1402 .sp
1403 PCRE_NEWLINE_CR
1404 PCRE_NEWLINE_LF
1405 PCRE_NEWLINE_CRLF
1406 PCRE_NEWLINE_ANYCRLF
1407 PCRE_NEWLINE_ANY
1408 .sp
1409 These options override the newline definition that was chosen or defaulted when
1410 the pattern was compiled. For details, see the description of
1411 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1412 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1413 the way the match position is advanced after a match failure for an unanchored
1414 pattern.
1415 .P
1416 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1417 match attempt for an unanchored pattern fails when the current position is at a
1418 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1419 characters, the match position is advanced by two characters instead of one, in
1420 other words, to after the CRLF.
1421 .P
1422 The above rule is a compromise that makes the most common cases work as
1423 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1424 set), it does not match the string "\er\enA" because, after failing at the
1425 start, it skips both the CR and the LF before retrying. However, the pattern
1426 [\er\en]A does match that string, because it contains an explicit CR or LF
1427 reference, and so advances only by one character after the first failure.
1428 .P
1429 An explicit match for CR of LF is either a literal appearance of one of those
1430 characters, or one of the \er or \en escape sequences. Implicit matches such as
1431 [^X] do not count, nor does \es (which includes CR and LF in the characters
1432 that it matches).
1433 .P
1434 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1435 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1436 .sp
1437 PCRE_NOTBOL
1438 .sp
1439 This option specifies that first character of the subject string is not the
1440 beginning of a line, so the circumflex metacharacter should not match before
1441 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1442 never to match. This option affects only the behaviour of the circumflex
1443 metacharacter. It does not affect \eA.
1444 .sp
1445 PCRE_NOTEOL
1446 .sp
1447 This option specifies that the end of the subject string is not the end of a
1448 line, so the dollar metacharacter should not match it nor (except in multiline
1449 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1450 compile time) causes dollar never to match. This option affects only the
1451 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1452 .sp
1453 PCRE_NOTEMPTY
1454 .sp
1455 An empty string is not considered to be a valid match if this option is set. If
1456 there are alternatives in the pattern, they are tried. If all the alternatives
1457 match the empty string, the entire match fails. For example, if the pattern
1458 .sp
1459 a?b?
1460 .sp
1461 is applied to a string not beginning with "a" or "b", it matches an empty
1462 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1463 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1464 .sp
1465 PCRE_NOTEMPTY_ATSTART
1466 .sp
1467 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1468 the start of the subject is permitted. If the pattern is anchored, such a match
1469 can occur only if the pattern contains \eK.
1470 .P
1471 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1472 does make a special case of a pattern match of the empty string within its
1473 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1474 emulate Perl's behaviour after matching a null string by first trying the match
1475 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1476 if that fails, by advancing the starting offset (see below) and trying an
1477 ordinary match again. There is some code that demonstrates how to do this in
1478 the
1479 .\" HREF
1480 \fBpcredemo\fP
1481 .\"
1482 sample program. In the most general case, you have to check to see if the
1483 newline convention recognizes CRLF as a newline, and if so, and the current
1484 character is CR followed by LF, advance the starting offset by two characters
1485 instead of one.
1486 .sp
1487 PCRE_NO_START_OPTIMIZE
1488 .sp
1489 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1490 a match, in order to speed up the process. For example, if it is known that an
1491 unanchored match must start with a specific character, it searches the subject
1492 for that character, and fails immediately if it cannot find it, without
1493 actually running the main matching function. This means that a special item
1494 such as (*COMMIT) at the start of a pattern is not considered until after a
1495 suitable starting point for the match has been found. When callouts or (*MARK)
1496 items are in use, these "start-up" optimizations can cause them to be skipped
1497 if the pattern is never actually used. The start-up optimizations are in effect
1498 a pre-scan of the subject that takes place before the pattern is run.
1499 .P
1500 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1501 causing performance to suffer, but ensuring that in cases where the result is
1502 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1503 are considered at every possible starting position in the subject string. If
1504 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1505 time.
1506 .P
1507 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1508 Consider the pattern
1509 .sp
1510 (*COMMIT)ABC
1511 .sp
1512 When this is compiled, PCRE records the fact that a match must start with the
1513 character "A". Suppose the subject string is "DEFABC". The start-up
1514 optimization scans along the subject, finds "A" and runs the first match
1515 attempt from there. The (*COMMIT) item means that the pattern must match the
1516 current starting position, which in this case, it does. However, if the same
1517 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1518 subject string does not happen. The first match attempt is run starting from
1519 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1520 the overall result is "no match". If the pattern is studied, more start-up
1521 optimizations may be used. For example, a minimum length for the subject may be
1522 recorded. Consider the pattern
1523 .sp
1524 (*MARK:A)(X|Y)
1525 .sp
1526 The minimum length for a match is one character. If the subject is "ABC", there
1527 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1528 If the pattern is studied, the final attempt does not take place, because PCRE
1529 knows that the subject is too short, and so the (*MARK) is never encountered.
1530 In this case, studying the pattern does not affect the overall match result,
1531 which is still "no match", but it does affect the auxiliary information that is
1532 returned.
1533 .sp
1534 PCRE_NO_UTF8_CHECK
1535 .sp
1536 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1537 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1538 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1539 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1540 strings in the
1541 .\" HTML <a href="pcre.html#utf8strings">
1542 .\" </a>
1543 section on UTF-8 support
1544 .\"
1545 in the main
1546 .\" HREF
1547 \fBpcre\fP
1548 .\"
1549 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1550 the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1551 a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1552 both cases, information about the precise nature of the error may also be
1553 returned (see the descriptions of these errors in the section entitled \fIError
1554 return values from\fP \fBpcre_exec()\fP
1555 .\" HTML <a href="#errorlist">
1556 .\" </a>
1557 below).
1558 .\"
1559 If \fIstartoffset\fP contains a value that does not point to the start of a
1560 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1561 returned.
1562 .P
1563 If you already know that your subject is valid, and you want to skip these
1564 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1565 calling \fBpcre_exec()\fP. You might want to do this for the second and
1566 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1567 all the matches in a single subject string. However, you should be sure that
1568 the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1569 end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1570 invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1571 undefined. Your program may crash.
1572 .sp
1573 PCRE_PARTIAL_HARD
1574 PCRE_PARTIAL_SOFT
1575 .sp
1576 These options turn on the partial matching feature. For backwards
1577 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1578 occurs if the end of the subject string is reached successfully, but there are
1579 not enough subject characters to complete the match. If this happens when
1580 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1581 testing any remaining alternatives. Only if no complete match can be found is
1582 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1583 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1584 but only if no complete match can be found.
1585 .P
1586 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1587 partial match is found, \fBpcre_exec()\fP immediately returns
1588 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1589 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1590 important that an alternative complete match.
1591 .P
1592 In both cases, the portion of the string that was inspected when the partial
1593 match was found is set as the first matching string. There is a more detailed
1594 discussion of partial and multi-segment matching, with examples, in the
1595 .\" HREF
1596 \fBpcrepartial\fP
1597 .\"
1598 documentation.
1599 .
1600 .
1601 .SS "The string to be matched by \fBpcre_exec()\fP"
1602 .rs
1603 .sp
1604 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1605 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1606 in \fIstartoffset\fP. If this is negative or greater than the length of the
1607 subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1608 offset is zero, the search for a match starts at the beginning of the subject,
1609 and this is by far the most common case. In UTF-8 mode, the byte offset must
1610 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1611 pattern string, the subject may contain binary zero bytes.
1612 .P
1613 A non-zero starting offset is useful when searching for another match in the
1614 same subject by calling \fBpcre_exec()\fP again after a previous success.
1615 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1616 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1617 lookbehind. For example, consider the pattern
1618 .sp
1619 \eBiss\eB
1620 .sp
1621 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1622 the current position in the subject is not a word boundary.) When applied to
1623 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1624 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1625 subject, namely "issipi", it does not match, because \eB is always false at the
1626 start of the subject, which is deemed to be a word boundary. However, if
1627 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1628 set to 4, it finds the second occurrence of "iss" because it is able to look
1629 behind the starting point to discover that it is preceded by a letter.
1630 .P
1631 Finding all the matches in a subject is tricky when the pattern can match an
1632 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1633 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1634 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1635 and trying an ordinary match again. There is some code that demonstrates how to
1636 do this in the
1637 .\" HREF
1638 \fBpcredemo\fP
1639 .\"
1640 sample program. In the most general case, you have to check to see if the
1641 newline convention recognizes CRLF as a newline, and if so, and the current
1642 character is CR followed by LF, advance the starting offset by two characters
1643 instead of one.
1644 .P
1645 If a non-zero starting offset is passed when the pattern is anchored, one
1646 attempt to match at the given offset is made. This can only succeed if the
1647 pattern does not require the match to be at the start of the subject.
1648 .
1649 .
1650 .SS "How \fBpcre_exec()\fP returns captured substrings"
1651 .rs
1652 .sp
1653 In general, a pattern matches a certain portion of the subject, and in
1654 addition, further substrings from the subject may be picked out by parts of the
1655 pattern. Following the usage in Jeffrey Friedl's book, this is called
1656 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1657 a fragment of a pattern that picks out a substring. PCRE supports several other
1658 kinds of parenthesized subpattern that do not cause substrings to be captured.
1659 .P
1660 Captured substrings are returned to the caller via a vector of integers whose
1661 address is passed in \fIovector\fP. The number of elements in the vector is
1662 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1663 argument is NOT the size of \fIovector\fP in bytes.
1664 .P
1665 The first two-thirds of the vector is used to pass back captured substrings,
1666 each substring using a pair of integers. The remaining third of the vector is
1667 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1668 and is not available for passing back information. The number passed in
1669 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1670 rounded down.
1671 .P
1672 When a match is successful, information about captured substrings is returned
1673 in pairs of integers, starting at the beginning of \fIovector\fP, and
1674 continuing up to two-thirds of its length at the most. The first element of
1675 each pair is set to the byte offset of the first character in a substring, and
1676 the second is set to the byte offset of the first character after the end of a
1677 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1678 mode. They are not character counts.
1679 .P
1680 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1681 portion of the subject string matched by the entire pattern. The next pair is
1682 used for the first capturing subpattern, and so on. The value returned by
1683 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1684 For example, if two substrings have been captured, the returned value is 3. If
1685 there are no capturing subpatterns, the return value from a successful match is
1686 1, indicating that just the first pair of offsets has been set.
1687 .P
1688 If a capturing subpattern is matched repeatedly, it is the last portion of the
1689 string that it matched that is returned.
1690 .P
1691 If the vector is too small to hold all the captured substring offsets, it is
1692 used as far as possible (up to two-thirds of its length), and the function
1693 returns a value of zero. If the substring offsets are not of interest,
1694 \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1695 \fIovecsize\fP as zero. However, if the pattern contains back references and
1696 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1697 has to get additional memory for use during matching. Thus it is usually
1698 advisable to supply an \fIovector\fP.
1699 .P
1700 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1701 subpatterns there are in a compiled pattern. The smallest size for
1702 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1703 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1704 .P
1705 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1706 the subject when subpattern \fIn\fP has not been used at all. For example, if
1707 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1708 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1709 happens, both values in the offset pairs corresponding to unused subpatterns
1710 are set to -1.
1711 .P
1712 Offset values that correspond to unused subpatterns at the end of the
1713 expression are also set to -1. For example, if the string "abc" is matched
1714 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1715 return from the function is 2, because the highest used capturing subpattern
1716 number is 1, and the offsets for for the second and third capturing subpatterns
1717 (assuming the vector is large enough, of course) are set to -1.
1718 .P
1719 \fBNote\fP: Elements of \fIovector\fP that do not correspond to capturing
1720 parentheses in the pattern are never changed. That is, if a pattern contains
1721 \fIn\fP capturing parentheses, no more than \fIovector[0]\fP to
1722 \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other elements retain
1723 whatever values they previously had.
1724 .P
1725 Some convenience functions are provided for extracting the captured substrings
1726 as separate strings. These are described below.
1727 .
1728 .
1729 .\" HTML <a name="errorlist"></a>
1730 .SS "Error return values from \fBpcre_exec()\fP"
1731 .rs
1732 .sp
1733 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1734 defined in the header file:
1735 .sp
1736 PCRE_ERROR_NOMATCH (-1)
1737 .sp
1738 The subject string did not match the pattern.
1739 .sp
1740 PCRE_ERROR_NULL (-2)
1741 .sp
1742 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1743 NULL and \fIovecsize\fP was not zero.
1744 .sp
1745 PCRE_ERROR_BADOPTION (-3)
1746 .sp
1747 An unrecognized bit was set in the \fIoptions\fP argument.
1748 .sp
1749 PCRE_ERROR_BADMAGIC (-4)
1750 .sp
1751 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1752 the case when it is passed a junk pointer and to detect when a pattern that was
1753 compiled in an environment of one endianness is run in an environment with the
1754 other endianness. This is the error that PCRE gives when the magic number is
1755 not present.
1756 .sp
1757 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1758 .sp
1759 While running the pattern match, an unknown item was encountered in the
1760 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1761 of the compiled pattern.
1762 .sp
1763 PCRE_ERROR_NOMEMORY (-6)
1764 .sp
1765 If a pattern contains back references, but the \fIovector\fP that is passed to
1766 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1767 gets a block of memory at the start of matching to use for this purpose. If the
1768 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1769 automatically freed at the end of matching.
1770 .P
1771 This error is also given if \fBpcre_stack_malloc()\fP fails in
1772 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1773 \fB--disable-stack-for-recursion\fP.
1774 .sp
1775 PCRE_ERROR_NOSUBSTRING (-7)
1776 .sp
1777 This error is used by the \fBpcre_copy_substring()\fP,
1778 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1779 below). It is never returned by \fBpcre_exec()\fP.
1780 .sp
1781 PCRE_ERROR_MATCHLIMIT (-8)
1782 .sp
1783 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1784 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1785 above.
1786 .sp
1787 PCRE_ERROR_CALLOUT (-9)
1788 .sp
1789 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1790 use by callout functions that want to yield a distinctive error code. See the
1791 .\" HREF
1792 \fBpcrecallout\fP
1793 .\"
1794 documentation for details.
1795 .sp
1796 PCRE_ERROR_BADUTF8 (-10)
1797 .sp
1798 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1799 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1800 (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1801 UTF-8 character is placed in the first element, and a reason code is placed in
1802 the second element. The reason codes are listed in the
1803 .\" HTML <a href="#badutf8reasons">
1804 .\" </a>
1805 following section.
1806 .\"
1807 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1808 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1809 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1810 .sp
1811 PCRE_ERROR_BADUTF8_OFFSET (-11)
1812 .sp
1813 The UTF-8 byte sequence that was passed as a subject was checked and found to
1814 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
1815 \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1816 end of the subject.
1817 .sp
1818 PCRE_ERROR_PARTIAL (-12)
1819 .sp
1820 The subject string did not match, but it did match partially. See the
1821 .\" HREF
1822 \fBpcrepartial\fP
1823 .\"
1824 documentation for details of partial matching.
1825 .sp
1826 PCRE_ERROR_BADPARTIAL (-13)
1827 .sp
1828 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1829 option was used with a compiled pattern containing items that were not
1830 supported for partial matching. From release 8.00 onwards, there are no
1831 restrictions on partial matching.
1832 .sp
1833 PCRE_ERROR_INTERNAL (-14)
1834 .sp
1835 An unexpected internal error has occurred. This error could be caused by a bug
1836 in PCRE or by overwriting of the compiled pattern.
1837 .sp
1838 PCRE_ERROR_BADCOUNT (-15)
1839 .sp
1840 This error is given if the value of the \fIovecsize\fP argument is negative.
1841 .sp
1842 PCRE_ERROR_RECURSIONLIMIT (-21)
1843 .sp
1844 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1845 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1846 description above.
1847 .sp
1848 PCRE_ERROR_BADNEWLINE (-23)
1849 .sp
1850 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1851 .sp
1852 PCRE_ERROR_BADOFFSET (-24)
1853 .sp
1854 The value of \fIstartoffset\fP was negative or greater than the length of the
1855 subject, that is, the value in \fIlength\fP.
1856 .sp
1857 PCRE_ERROR_SHORTUTF8 (-25)
1858 .sp
1859 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
1860 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
1861 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
1862 fact sufficient to detect this case, but this special error code for
1863 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
1864 retained for backwards compatibility.
1865 .sp
1866 PCRE_ERROR_RECURSELOOP (-26)
1867 .sp
1868 This error is returned when \fBpcre_exec()\fP detects a recursion loop within
1869 the pattern. Specifically, it means that either the whole pattern or a
1870 subpattern has been called recursively for the second time at the same position
1871 in the subject string. Some simple patterns that might do this are detected and
1872 faulted at compile time, but more complicated cases, in particular mutual
1873 recursions between two different subpatterns, cannot be detected until run
1874 time.
1875 .P
1876 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1877 .
1878 .
1879 .\" HTML <a name="badutf8reasons"></a>
1880 .SS "Reason codes for invalid UTF-8 strings"
1881 .rs
1882 .sp
1883 When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
1884 PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
1885 least 2, the offset of the start of the invalid UTF-8 character is placed in
1886 the first output vector element (\fIovector[0]\fP) and a reason code is placed
1887 in the second element (\fIovector[1]\fP). The reason codes are given names in
1888 the \fBpcre.h\fP header file:
1889 .sp
1890 PCRE_UTF8_ERR1
1891 PCRE_UTF8_ERR2
1892 PCRE_UTF8_ERR3
1893 PCRE_UTF8_ERR4
1894 PCRE_UTF8_ERR5
1895 .sp
1896 The string ends with a truncated UTF-8 character; the code specifies how many
1897 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
1898 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
1899 allows for up to 6 bytes, and this is checked first; hence the possibility of
1900 4 or 5 missing bytes.
1901 .sp
1902 PCRE_UTF8_ERR6
1903 PCRE_UTF8_ERR7
1904 PCRE_UTF8_ERR8
1905 PCRE_UTF8_ERR9
1906 PCRE_UTF8_ERR10
1907 .sp
1908 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
1909 character do not have the binary value 0b10 (that is, either the most
1910 significant bit is 0, or the next bit is 1).
1911 .sp
1912 PCRE_UTF8_ERR11
1913 PCRE_UTF8_ERR12
1914 .sp
1915 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
1916 these code points are excluded by RFC 3629.
1917 .sp
1918 PCRE_UTF8_ERR13
1919 .sp
1920 A 4-byte character has a value greater than 0x10fff; these code points are
1921 excluded by RFC 3629.
1922 .sp
1923 PCRE_UTF8_ERR14
1924 .sp
1925 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
1926 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
1927 from UTF-8.
1928 .sp
1929 PCRE_UTF8_ERR15
1930 PCRE_UTF8_ERR16
1931 PCRE_UTF8_ERR17
1932 PCRE_UTF8_ERR18
1933 PCRE_UTF8_ERR19
1934 .sp
1935 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
1936 value that can be represented by fewer bytes, which is invalid. For example,
1937 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
1938 one byte.
1939 .sp
1940 PCRE_UTF8_ERR20
1941 .sp
1942 The two most significant bits of the first byte of a character have the binary
1943 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
1944 byte can only validly occur as the second or subsequent byte of a multi-byte
1945 character.
1946 .sp
1947 PCRE_UTF8_ERR21
1948 .sp
1949 The first byte of a character has the value 0xfe or 0xff. These values can
1950 never occur in a valid UTF-8 string.
1951 .
1952 .
1953 .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1954 .rs
1955 .sp
1956 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1957 .ti +5n
1958 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1959 .ti +5n
1960 .B int \fIbuffersize\fP);
1961 .PP
1962 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1963 .ti +5n
1964 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1965 .ti +5n
1966 .B const char **\fIstringptr\fP);
1967 .PP
1968 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1969 .ti +5n
1970 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1971 .PP
1972 Captured substrings can be accessed directly by using the offsets returned by
1973 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1974 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1975 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1976 as new, separate, zero-terminated strings. These functions identify substrings
1977 by number. The next section describes functions for extracting named
1978 substrings.
1979 .P
1980 A substring that contains a binary zero is correctly extracted and has a
1981 further zero added on the end, but the result is not, of course, a C string.
1982 However, you can process such a string by referring to the length that is
1983 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1984 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1985 for handling strings containing binary zeros, because the end of the final
1986 string is not independently indicated.
1987 .P
1988 The first three arguments are the same for all three of these functions:
1989 \fIsubject\fP is the subject string that has just been successfully matched,
1990 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1991 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1992 captured by the match, including the substring that matched the entire regular
1993 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1994 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1995 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1996 number of elements in the vector divided by three.
1997 .P
1998 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1999 extract a single substring, whose number is given as \fIstringnumber\fP. A
2000 value of zero extracts the substring that matched the entire pattern, whereas
2001 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2002 the string is placed in \fIbuffer\fP, whose length is given by
2003 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2004 obtained via \fBpcre_malloc\fP, and its address is returned via
2005 \fIstringptr\fP. The yield of the function is the length of the string, not
2006 including the terminating zero, or one of these error codes:
2007 .sp
2008 PCRE_ERROR_NOMEMORY (-6)
2009 .sp
2010 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2011 memory failed for \fBpcre_get_substring()\fP.
2012 .sp
2013 PCRE_ERROR_NOSUBSTRING (-7)
2014 .sp
2015 There is no substring whose number is \fIstringnumber\fP.
2016 .P
2017 The \fBpcre_get_substring_list()\fP function extracts all available substrings
2018 and builds a list of pointers to them. All this is done in a single block of
2019 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2020 is returned via \fIlistptr\fP, which is also the start of the list of string
2021 pointers. The end of the list is marked by a NULL pointer. The yield of the
2022 function is zero if all went well, or the error code
2023 .sp
2024 PCRE_ERROR_NOMEMORY (-6)
2025 .sp
2026 if the attempt to get the memory block failed.
2027 .P
2028 When any of these functions encounter a substring that is unset, which can
2029 happen when capturing subpattern number \fIn+1\fP matches some part of the
2030 subject, but subpattern \fIn\fP has not been used at all, they return an empty
2031 string. This can be distinguished from a genuine zero-length substring by
2032 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2033 substrings.
2034 .P
2035 The two convenience functions \fBpcre_free_substring()\fP and
2036 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2037 a previous call of \fBpcre_get_substring()\fP or
2038 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2039 the function pointed to by \fBpcre_free\fP, which of course could be called
2040 directly from a C program. However, PCRE is used in some situations where it is
2041 linked via a special interface to another programming language that cannot use
2042 \fBpcre_free\fP directly; it is for these cases that the functions are
2043 provided.
2044 .
2045 .
2046 .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
2047 .rs
2048 .sp
2049 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2050 .ti +5n
2051 .B const char *\fIname\fP);
2052 .PP
2053 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2054 .ti +5n
2055 .B const char *\fIsubject\fP, int *\fIovector\fP,
2056 .ti +5n
2057 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2058 .ti +5n
2059 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2060 .PP
2061 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2062 .ti +5n
2063 .B const char *\fIsubject\fP, int *\fIovector\fP,
2064 .ti +5n
2065 .B int \fIstringcount\fP, const char *\fIstringname\fP,
2066 .ti +5n
2067 .B const char **\fIstringptr\fP);
2068 .PP
2069 To extract a substring by name, you first have to find associated number.
2070 For example, for this pattern
2071 .sp
2072 (a+)b(?<xxx>\ed+)...
2073 .sp
2074 the number of the subpattern called "xxx" is 2. If the name is known to be
2075 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2076 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2077 pattern, and the second is the name. The yield of the function is the
2078 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2079 that name.
2080 .P
2081 Given the number, you can extract the substring directly, or use one of the
2082 functions described in the previous section. For convenience, there are also
2083 two functions that do the whole job.
2084 .P
2085 Most of the arguments of \fBpcre_copy_named_substring()\fP and
2086 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2087 functions that extract by number. As these are described in the previous
2088 section, they are not re-described here. There are just two differences:
2089 .P
2090 First, instead of a substring number, a substring name is given. Second, there
2091 is an extra argument, given at the start, which is a pointer to the compiled
2092 pattern. This is needed in order to gain access to the name-to-number
2093 translation table.
2094 .P
2095 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2096 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2097 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2098 the behaviour may not be what you want (see the next section).
2099 .P
2100 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2101 subpatterns with the same number, as described in the
2102 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2103 .\" </a>
2104 section on duplicate subpattern numbers
2105 .\"
2106 in the
2107 .\" HREF
2108 \fBpcrepattern\fP
2109 .\"
2110 page, you cannot use names to distinguish the different subpatterns, because
2111 names are not included in the compiled code. The matching process uses only
2112 numbers. For this reason, the use of different names for subpatterns of the
2113 same number causes an error at compile time.
2114 .
2115 .SH "DUPLICATE SUBPATTERN NAMES"
2116 .rs
2117 .sp
2118 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2119 .ti +5n
2120 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2121 .PP
2122 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2123 are not required to be unique. (Duplicate names are always allowed for
2124 subpatterns with the same number, created by using the (?| feature. Indeed, if
2125 such subpatterns are named, they are required to use the same names.)
2126 .P
2127 Normally, patterns with duplicate names are such that in any one match, only
2128 one of the named subpatterns participates. An example is shown in the
2129 .\" HREF
2130 \fBpcrepattern\fP
2131 .\"
2132 documentation.
2133 .P
2134 When duplicates are present, \fBpcre_copy_named_substring()\fP and
2135 \fBpcre_get_named_substring()\fP return the first substring corresponding to
2136 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2137 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2138 returns one of the numbers that are associated with the name, but it is not
2139 defined which it is.
2140 .P
2141 If you want to get full details of all captured substrings for a given name,
2142 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2143 argument is the compiled pattern, and the second is the name. The third and
2144 fourth are pointers to variables which are updated by the function. After it
2145 has run, they point to the first and last entries in the name-to-number table
2146 for the given name. The function itself returns the length of each entry, or
2147 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2148 described above in the section entitled \fIInformation about a pattern\fP
2149 .\" HTML <a href="#infoaboutpattern">
2150 .\" </a>
2151 above.
2152 .\"
2153 Given all the relevant entries for the name, you can extract each of their
2154 numbers, and hence the captured data, if any.
2155 .
2156 .
2157 .SH "FINDING ALL POSSIBLE MATCHES"
2158 .rs
2159 .sp
2160 The traditional matching function uses a similar algorithm to Perl, which stops
2161 when it finds the first match, starting at a given point in the subject. If you
2162 want to find all possible matches, or the longest possible match, consider
2163 using the alternative matching function (see below) instead. If you cannot use
2164 the alternative function, but still need to find all possible matches, you
2165 can kludge it up by making use of the callout facility, which is described in
2166 the
2167 .\" HREF
2168 \fBpcrecallout\fP
2169 .\"
2170 documentation.
2171 .P
2172 What you have to do is to insert a callout right at the end of the pattern.
2173 When your callout function is called, extract and save the current matched
2174 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2175 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2176 will yield PCRE_ERROR_NOMATCH.
2177 .
2178 .
2179 .\" HTML <a name="dfamatch"></a>
2180 .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2181 .rs
2182 .sp
2183 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2184 .ti +5n
2185 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2186 .ti +5n
2187 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2188 .ti +5n
2189 .B int *\fIworkspace\fP, int \fIwscount\fP);
2190 .P
2191 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2192 a compiled pattern, using a matching algorithm that scans the subject string
2193 just once, and does not backtrack. This has different characteristics to the
2194 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2195 patterns are not supported. Nevertheless, there are times when this kind of
2196 matching can be useful. For a discussion of the two matching algorithms, and a
2197 list of features that \fBpcre_dfa_exec()\fP does not support, see the
2198 .\" HREF
2199 \fBpcrematching\fP
2200 .\"
2201 documentation.
2202 .P
2203 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2204 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2205 different way, and this is described below. The other common arguments are used
2206 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2207 here.
2208 .P
2209 The two additional arguments provide workspace for the function. The workspace
2210 vector should contain at least 20 elements. It is used for keeping track of
2211 multiple paths through the pattern tree. More workspace will be needed for
2212 patterns and subjects where there are a lot of potential matches.
2213 .P
2214 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2215 .sp
2216 int rc;
2217 int ovector[10];
2218 int wspace[20];
2219 rc = pcre_dfa_exec(
2220 re, /* result of pcre_compile() */
2221 NULL, /* we didn't study the pattern */
2222 "some string", /* the subject string */
2223 11, /* the length of the subject string */
2224 0, /* start at offset 0 in the subject */
2225 0, /* default options */
2226 ovector, /* vector of integers for substring information */
2227 10, /* number of elements (NOT size in bytes) */
2228 wspace, /* working space vector */
2229 20); /* number of elements (NOT size in bytes) */
2230 .
2231 .SS "Option bits for \fBpcre_dfa_exec()\fP"
2232 .rs
2233 .sp
2234 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2235 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2236 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2237 PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2238 PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2239 All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2240 so their description is not repeated here.
2241 .sp
2242 PCRE_PARTIAL_HARD
2243 PCRE_PARTIAL_SOFT
2244 .sp
2245 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2246 details are slightly different. When PCRE_PARTIAL_HARD is set for
2247 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2248 is reached and there is still at least one matching possibility that requires
2249 additional characters. This happens even if some complete matches have also
2250 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2251 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2252 there have been no complete matches, but there is still at least one matching
2253 possibility. The portion of the string that was inspected when the longest
2254 partial match was found is set as the first matching string in both cases.
2255 There is a more detailed discussion of partial and multi-segment matching, with
2256 examples, in the
2257 .\" HREF
2258 \fBpcrepartial\fP
2259 .\"
2260 documentation.
2261 .sp
2262 PCRE_DFA_SHORTEST
2263 .sp
2264 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2265 soon as it has found one match. Because of the way the alternative algorithm
2266 works, this is necessarily the shortest possible match at the first possible
2267 matching point in the subject string.
2268 .sp
2269 PCRE_DFA_RESTART
2270 .sp
2271 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2272 again, with additional subject characters, and have it continue with the same
2273 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2274 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2275 before because data about the match so far is left in them after a partial
2276 match. There is more discussion of this facility in the
2277 .\" HREF
2278 \fBpcrepartial\fP
2279 .\"
2280 documentation.
2281 .
2282 .
2283 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2284 .rs
2285 .sp
2286 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2287 substring in the subject. Note, however, that all the matches from one run of
2288 the function start at the same point in the subject. The shorter matches are
2289 all initial substrings of the longer matches. For example, if the pattern
2290 .sp
2291 <.*>
2292 .sp
2293 is matched against the string
2294 .sp
2295 This is <something> <something else> <something further> no more
2296 .sp
2297 the three matched strings are
2298 .sp
2299 <something>
2300 <something> <something else>
2301 <something> <something else> <something further>
2302 .sp
2303 On success, the yield of the function is a number greater than zero, which is
2304 the number of matched substrings. The substrings themselves are returned in
2305 \fIovector\fP. Each string uses two elements; the first is the offset to the
2306 start, and the second is the offset to the end. In fact, all the strings have
2307 the same start offset. (Space could have been saved by giving this only once,
2308 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2309 returns data, even though the meaning of the strings is different.)
2310 .P
2311 The strings are returned in reverse order of length; that is, the longest
2312 matching string is given first. If there were too many matches to fit into
2313 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2314 the longest matches.
2315 .
2316 .
2317 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2318 .rs
2319 .sp
2320 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2321 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2322 described
2323 .\" HTML <a href="#errorlist">
2324 .\" </a>
2325 above.
2326 .\"
2327 There are in addition the following errors that are specific to
2328 \fBpcre_dfa_exec()\fP:
2329 .sp
2330 PCRE_ERROR_DFA_UITEM (-16)
2331 .sp
2332 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2333 that it does not support, for instance, the use of \eC or a back reference.
2334 .sp
2335 PCRE_ERROR_DFA_UCOND (-17)
2336 .sp
2337 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2338 uses a back reference for the condition, or a test for recursion in a specific
2339 group. These are not supported.
2340 .sp
2341 PCRE_ERROR_DFA_UMLIMIT (-18)
2342 .sp
2343 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2344 block that contains a setting of the \fImatch_limit\fP field. This is not
2345 supported (it is meaningless).
2346 .sp
2347 PCRE_ERROR_DFA_WSSIZE (-19)
2348 .sp
2349 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2350 \fIworkspace\fP vector.
2351 .sp
2352 PCRE_ERROR_DFA_RECURSE (-20)
2353 .sp
2354 When a recursive subpattern is processed, the matching function calls itself
2355 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2356 error is given if the output vector is not large enough. This should be
2357 extremely rare, as a vector of size 1000 is used.
2358 .
2359 .
2360 .SH "SEE ALSO"
2361 .rs
2362 .sp
2363 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2364 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2365 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2366 .
2367 .
2368 .SH AUTHOR
2369 .rs
2370 .sp
2371 .nf
2372 Philip Hazel
2373 University Computing Service
2374 Cambridge CB2 3QH, England.
2375 .fi
2376 .
2377 .
2378 .SH REVISION
2379 .rs
2380 .sp
2381 .nf
2382 Last updated: 28 July 2011
2383 Copyright (c) 1997-2011 University of Cambridge.
2384 .fi

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5