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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 579 - (show annotations)
Wed Nov 24 17:39:25 2010 UTC (8 years, 8 months ago) by ph10
File size: 98796 byte(s)
Tidies of documenation and code while preparing for 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. The offset from the start of the pattern to the byte that
440 was being processed when the error was discovered is placed in the variable
441 pointed to by \fIerroffset\fP, which must not be NULL. If it is, an immediate
442 error is given. Some errors are not detected until checks are carried out when
443 the whole pattern has been scanned; in this case the offset is set to the end
444 of the pattern.
445 .P
446 Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
447 point into the middle of a UTF-8 character (for example, when
448 PCRE_ERROR_BADUTF8 is returned for an invalid UTF-8 string).
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
556 is controlled by the options passed to \fBpcre_compile()\fP or by a special
557 sequence at the start of the 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 .SH "INFORMATION ABOUT A PATTERN"
950 .rs
951 .sp
952 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
953 .ti +5n
954 .B int \fIwhat\fP, void *\fIwhere\fP);
955 .PP
956 The \fBpcre_fullinfo()\fP function returns information about a compiled
957 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
958 nevertheless retained for backwards compability (and is documented below).
959 .P
960 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
961 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
962 the pattern was not studied. The third argument specifies which piece of
963 information is required, and the fourth argument is a pointer to a variable
964 to receive the data. The yield of the function is zero for success, or one of
965 the following negative numbers:
966 .sp
967 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
968 the argument \fIwhere\fP was NULL
969 PCRE_ERROR_BADMAGIC the "magic number" was not found
970 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
971 .sp
972 The "magic number" is placed at the start of each compiled pattern as an simple
973 check against passing an arbitrary memory pointer. Here is a typical call of
974 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
975 .sp
976 int rc;
977 size_t length;
978 rc = pcre_fullinfo(
979 re, /* result of pcre_compile() */
980 pe, /* result of pcre_study(), or NULL */
981 PCRE_INFO_SIZE, /* what is required */
982 &length); /* where to put the data */
983 .sp
984 The possible values for the third argument are defined in \fBpcre.h\fP, and are
985 as follows:
986 .sp
987 PCRE_INFO_BACKREFMAX
988 .sp
989 Return the number of the highest back reference in the pattern. The fourth
990 argument should point to an \fBint\fP variable. Zero is returned if there are
991 no back references.
992 .sp
993 PCRE_INFO_CAPTURECOUNT
994 .sp
995 Return the number of capturing subpatterns in the pattern. The fourth argument
996 should point to an \fBint\fP variable.
997 .sp
998 PCRE_INFO_DEFAULT_TABLES
999 .sp
1000 Return a pointer to the internal default character tables within PCRE. The
1001 fourth argument should point to an \fBunsigned char *\fP variable. This
1002 information call is provided for internal use by the \fBpcre_study()\fP
1003 function. External callers can cause PCRE to use its internal tables by passing
1004 a NULL table pointer.
1005 .sp
1006 PCRE_INFO_FIRSTBYTE
1007 .sp
1008 Return information about the first byte of any matched string, for a
1009 non-anchored pattern. The fourth argument should point to an \fBint\fP
1010 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1011 still recognized for backwards compatibility.)
1012 .P
1013 If there is a fixed first byte, for example, from a pattern such as
1014 (cat|cow|coyote), its value is returned. Otherwise, if either
1015 .sp
1016 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1017 starts with "^", or
1018 .sp
1019 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1020 (if it were set, the pattern would be anchored),
1021 .sp
1022 -1 is returned, indicating that the pattern matches only at the start of a
1023 subject string or after any newline within the string. Otherwise -2 is
1024 returned. For anchored patterns, -2 is returned.
1025 .sp
1026 PCRE_INFO_FIRSTTABLE
1027 .sp
1028 If the pattern was studied, and this resulted in the construction of a 256-bit
1029 table indicating a fixed set of bytes for the first byte in any matching
1030 string, a pointer to the table is returned. Otherwise NULL is returned. The
1031 fourth argument should point to an \fBunsigned char *\fP variable.
1032 .sp
1033 PCRE_INFO_HASCRORLF
1034 .sp
1035 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1036 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1037 explicit match is either a literal CR or LF character, or \er or \en.
1038 .sp
1039 PCRE_INFO_JCHANGED
1040 .sp
1041 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1042 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1043 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1044 .sp
1045 PCRE_INFO_LASTLITERAL
1046 .sp
1047 Return the value of the rightmost literal byte that must exist in any matched
1048 string, other than at its start, if such a byte has been recorded. The fourth
1049 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1050 returned. For anchored patterns, a last literal byte is recorded only if it
1051 follows something of variable length. For example, for the pattern
1052 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1053 is -1.
1054 .sp
1055 PCRE_INFO_MINLENGTH
1056 .sp
1057 If the pattern was studied and a minimum length for matching subject strings
1058 was computed, its value is returned. Otherwise the returned value is -1. The
1059 value is a number of characters, not bytes (this may be relevant in UTF-8
1060 mode). The fourth argument should point to an \fBint\fP variable. A
1061 non-negative value is a lower bound to the length of any matching string. There
1062 may not be any strings of that length that do actually match, but every string
1063 that does match is at least that long.
1064 .sp
1065 PCRE_INFO_NAMECOUNT
1066 PCRE_INFO_NAMEENTRYSIZE
1067 PCRE_INFO_NAMETABLE
1068 .sp
1069 PCRE supports the use of named as well as numbered capturing parentheses. The
1070 names are just an additional way of identifying the parentheses, which still
1071 acquire numbers. Several convenience functions such as
1072 \fBpcre_get_named_substring()\fP are provided for extracting captured
1073 substrings by name. It is also possible to extract the data directly, by first
1074 converting the name to a number in order to access the correct pointers in the
1075 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1076 you need to use the name-to-number map, which is described by these three
1077 values.
1078 .P
1079 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1080 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1081 entry; both of these return an \fBint\fP value. The entry size depends on the
1082 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1083 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1084 are the number of the capturing parenthesis, most significant byte first. The
1085 rest of the entry is the corresponding name, zero terminated.
1086 .P
1087 The names are in alphabetical order. Duplicate names may appear if (?| is used
1088 to create multiple groups with the same number, as described in the
1089 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1090 .\" </a>
1091 section on duplicate subpattern numbers
1092 .\"
1093 in the
1094 .\" HREF
1095 \fBpcrepattern\fP
1096 .\"
1097 page. Duplicate names for subpatterns with different numbers are permitted only
1098 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1099 table in the order in which they were found in the pattern. In the absence of
1100 (?| this is the order of increasing number; when (?| is used this is not
1101 necessarily the case because later subpatterns may have lower numbers.
1102 .P
1103 As a simple example of the name/number table, consider the following pattern
1104 (assume PCRE_EXTENDED is set, so white space - including newlines - is
1105 ignored):
1106 .sp
1107 .\" JOIN
1108 (?<date> (?<year>(\ed\ed)?\ed\ed) -
1109 (?<month>\ed\ed) - (?<day>\ed\ed) )
1110 .sp
1111 There are four named subpatterns, so the table has four entries, and each entry
1112 in the table is eight bytes long. The table is as follows, with non-printing
1113 bytes shows in hexadecimal, and undefined bytes shown as ??:
1114 .sp
1115 00 01 d a t e 00 ??
1116 00 05 d a y 00 ?? ??
1117 00 04 m o n t h 00
1118 00 02 y e a r 00 ??
1119 .sp
1120 When writing code to extract data from named subpatterns using the
1121 name-to-number map, remember that the length of the entries is likely to be
1122 different for each compiled pattern.
1123 .sp
1124 PCRE_INFO_OKPARTIAL
1125 .sp
1126 Return 1 if the pattern can be used for partial matching with
1127 \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1128 \fBint\fP variable. From release 8.00, this always returns 1, because the
1129 restrictions that previously applied to partial matching have been lifted. The
1130 .\" HREF
1131 \fBpcrepartial\fP
1132 .\"
1133 documentation gives details of partial matching.
1134 .sp
1135 PCRE_INFO_OPTIONS
1136 .sp
1137 Return a copy of the options with which the pattern was compiled. The fourth
1138 argument should point to an \fBunsigned long int\fP variable. These option bits
1139 are those specified in the call to \fBpcre_compile()\fP, modified by any
1140 top-level option settings at the start of the pattern itself. In other words,
1141 they are the options that will be in force when matching starts. For example,
1142 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1143 result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1144 .P
1145 A pattern is automatically anchored by PCRE if all of its top-level
1146 alternatives begin with one of the following:
1147 .sp
1148 ^ unless PCRE_MULTILINE is set
1149 \eA always
1150 \eG always
1151 .\" JOIN
1152 .* if PCRE_DOTALL is set and there are no back
1153 references to the subpattern in which .* appears
1154 .sp
1155 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1156 \fBpcre_fullinfo()\fP.
1157 .sp
1158 PCRE_INFO_SIZE
1159 .sp
1160 Return the size of the compiled pattern, that is, the value that was passed as
1161 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1162 place the compiled data. The fourth argument should point to a \fBsize_t\fP
1163 variable.
1164 .sp
1165 PCRE_INFO_STUDYSIZE
1166 .sp
1167 Return the size of the data block pointed to by the \fIstudy_data\fP field in
1168 a \fBpcre_extra\fP block. That is, it is the value that was passed to
1169 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1170 created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no
1171 study data, zero is returned. The fourth argument should point to a
1172 \fBsize_t\fP variable.
1173 .
1174 .
1175 .SH "OBSOLETE INFO FUNCTION"
1176 .rs
1177 .sp
1178 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1179 .B *\fIfirstcharptr\fP);
1180 .PP
1181 The \fBpcre_info()\fP function is now obsolete because its interface is too
1182 restrictive to return all the available data about a compiled pattern. New
1183 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1184 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1185 following negative numbers:
1186 .sp
1187 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1188 PCRE_ERROR_BADMAGIC the "magic number" was not found
1189 .sp
1190 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1191 pattern was compiled is placed in the integer it points to (see
1192 PCRE_INFO_OPTIONS above).
1193 .P
1194 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1195 it is used to pass back information about the first character of any matched
1196 string (see PCRE_INFO_FIRSTBYTE above).
1197 .
1198 .
1199 .SH "REFERENCE COUNTS"
1200 .rs
1201 .sp
1202 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1203 .PP
1204 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1205 data block that contains a compiled pattern. It is provided for the benefit of
1206 applications that operate in an object-oriented manner, where different parts
1207 of the application may be using the same compiled pattern, but you want to free
1208 the block when they are all done.
1209 .P
1210 When a pattern is compiled, the reference count field is initialized to zero.
1211 It is changed only by calling this function, whose action is to add the
1212 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1213 function is the new value. However, the value of the count is constrained to
1214 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1215 it is forced to the appropriate limit value.
1216 .P
1217 Except when it is zero, the reference count is not correctly preserved if a
1218 pattern is compiled on one host and then transferred to a host whose byte-order
1219 is different. (This seems a highly unlikely scenario.)
1220 .
1221 .
1222 .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1223 .rs
1224 .sp
1225 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1226 .ti +5n
1227 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1228 .ti +5n
1229 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1230 .P
1231 The function \fBpcre_exec()\fP is called to match a subject string against a
1232 compiled pattern, which is passed in the \fIcode\fP argument. If the
1233 pattern was studied, the result of the study should be passed in the
1234 \fIextra\fP argument. This function is the main matching facility of the
1235 library, and it operates in a Perl-like manner. For specialist use there is
1236 also an alternative matching function, which is described
1237 .\" HTML <a href="#dfamatch">
1238 .\" </a>
1239 below
1240 .\"
1241 in the section about the \fBpcre_dfa_exec()\fP function.
1242 .P
1243 In most applications, the pattern will have been compiled (and optionally
1244 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1245 possible to save compiled patterns and study data, and then use them later
1246 in different processes, possibly even on different hosts. For a discussion
1247 about this, see the
1248 .\" HREF
1249 \fBpcreprecompile\fP
1250 .\"
1251 documentation.
1252 .P
1253 Here is an example of a simple call to \fBpcre_exec()\fP:
1254 .sp
1255 int rc;
1256 int ovector[30];
1257 rc = pcre_exec(
1258 re, /* result of pcre_compile() */
1259 NULL, /* we didn't study the pattern */
1260 "some string", /* the subject string */
1261 11, /* the length of the subject string */
1262 0, /* start at offset 0 in the subject */
1263 0, /* default options */
1264 ovector, /* vector of integers for substring information */
1265 30); /* number of elements (NOT size in bytes) */
1266 .
1267 .\" HTML <a name="extradata"></a>
1268 .SS "Extra data for \fBpcre_exec()\fR"
1269 .rs
1270 .sp
1271 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1272 data block. The \fBpcre_study()\fP function returns such a block (when it
1273 doesn't return NULL), but you can also create one for yourself, and pass
1274 additional information in it. The \fBpcre_extra\fP block contains the following
1275 fields (not necessarily in this order):
1276 .sp
1277 unsigned long int \fIflags\fP;
1278 void *\fIstudy_data\fP;
1279 unsigned long int \fImatch_limit\fP;
1280 unsigned long int \fImatch_limit_recursion\fP;
1281 void *\fIcallout_data\fP;
1282 const unsigned char *\fItables\fP;
1283 unsigned char **\fImark\fP;
1284 .sp
1285 The \fIflags\fP field is a bitmap that specifies which of the other fields
1286 are set. The flag bits are:
1287 .sp
1288 PCRE_EXTRA_STUDY_DATA
1289 PCRE_EXTRA_MATCH_LIMIT
1290 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1291 PCRE_EXTRA_CALLOUT_DATA
1292 PCRE_EXTRA_TABLES
1293 PCRE_EXTRA_MARK
1294 .sp
1295 Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1296 \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1297 the appropriate flag bit. You should not set this yourself, but you may add to
1298 the block by setting the other fields and their corresponding flag bits.
1299 .P
1300 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1301 vast amount of resources when running patterns that are not going to match,
1302 but which have a very large number of possibilities in their search trees. The
1303 classic example is a pattern that uses nested unlimited repeats.
1304 .P
1305 Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1306 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1307 number of times this function is called during a match, which has the effect of
1308 limiting the amount of backtracking that can take place. For patterns that are
1309 not anchored, the count restarts from zero for each position in the subject
1310 string.
1311 .P
1312 The default value for the limit can be set when PCRE is built; the default
1313 default is 10 million, which handles all but the most extreme cases. You can
1314 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1315 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1316 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1317 PCRE_ERROR_MATCHLIMIT.
1318 .P
1319 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1320 instead of limiting the total number of times that \fBmatch()\fP is called, it
1321 limits the depth of recursion. The recursion depth is a smaller number than the
1322 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1323 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1324 .P
1325 Limiting the recursion depth limits the amount of stack that can be used, or,
1326 when PCRE has been compiled to use memory on the heap instead of the stack, the
1327 amount of heap memory that can be used.
1328 .P
1329 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1330 built; the default default is the same value as the default for
1331 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1332 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1333 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1334 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1335 .P
1336 The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1337 and is described in the
1338 .\" HREF
1339 \fBpcrecallout\fP
1340 .\"
1341 documentation.
1342 .P
1343 The \fItables\fP field is used to pass a character tables pointer to
1344 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1345 pattern. A non-NULL value is stored with the compiled pattern only if custom
1346 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1347 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1348 internal tables to be used. This facility is helpful when re-using patterns
1349 that have been saved after compiling with an external set of tables, because
1350 the external tables might be at a different address when \fBpcre_exec()\fP is
1351 called. See the
1352 .\" HREF
1353 \fBpcreprecompile\fP
1354 .\"
1355 documentation for a discussion of saving compiled patterns for later use.
1356 .P
1357 If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1358 be set to point to a \fBchar *\fP variable. If the pattern contains any
1359 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1360 a name to pass back, a pointer to the name string (zero terminated) is placed
1361 in the variable pointed to by the \fImark\fP field. The names are within the
1362 compiled pattern; if you wish to retain such a name you must copy it before
1363 freeing the memory of a compiled pattern. If there is no name to pass back, the
1364 variable pointed to by the \fImark\fP field set to NULL. For details of the
1365 backtracking control verbs, see the section entitled
1366 .\" HTML <a href="pcrepattern#backtrackcontrol">
1367 .\" </a>
1368 "Backtracking control"
1369 .\"
1370 in the
1371 .\" HREF
1372 \fBpcrepattern\fP
1373 .\"
1374 documentation.
1375 .
1376 .
1377 .\" HTML <a name="execoptions"></a>
1378 .SS "Option bits for \fBpcre_exec()\fP"
1379 .rs
1380 .sp
1381 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1382 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1383 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1384 PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1385 PCRE_PARTIAL_HARD.
1386 .sp
1387 PCRE_ANCHORED
1388 .sp
1389 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1390 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1391 to be anchored by virtue of its contents, it cannot be made unachored at
1392 matching time.
1393 .sp
1394 PCRE_BSR_ANYCRLF
1395 PCRE_BSR_UNICODE
1396 .sp
1397 These options (which are mutually exclusive) control what the \eR escape
1398 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1399 match any Unicode newline sequence. These options override the choice that was
1400 made or defaulted when the pattern was compiled.
1401 .sp
1402 PCRE_NEWLINE_CR
1403 PCRE_NEWLINE_LF
1404 PCRE_NEWLINE_CRLF
1405 PCRE_NEWLINE_ANYCRLF
1406 PCRE_NEWLINE_ANY
1407 .sp
1408 These options override the newline definition that was chosen or defaulted when
1409 the pattern was compiled. For details, see the description of
1410 \fBpcre_compile()\fP above. During matching, the newline choice affects the
1411 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1412 the way the match position is advanced after a match failure for an unanchored
1413 pattern.
1414 .P
1415 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1416 match attempt for an unanchored pattern fails when the current position is at a
1417 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1418 characters, the match position is advanced by two characters instead of one, in
1419 other words, to after the CRLF.
1420 .P
1421 The above rule is a compromise that makes the most common cases work as
1422 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1423 set), it does not match the string "\er\enA" because, after failing at the
1424 start, it skips both the CR and the LF before retrying. However, the pattern
1425 [\er\en]A does match that string, because it contains an explicit CR or LF
1426 reference, and so advances only by one character after the first failure.
1427 .P
1428 An explicit match for CR of LF is either a literal appearance of one of those
1429 characters, or one of the \er or \en escape sequences. Implicit matches such as
1430 [^X] do not count, nor does \es (which includes CR and LF in the characters
1431 that it matches).
1432 .P
1433 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1434 valid newline sequence and explicit \er or \en escapes appear in the pattern.
1435 .sp
1436 PCRE_NOTBOL
1437 .sp
1438 This option specifies that first character of the subject string is not the
1439 beginning of a line, so the circumflex metacharacter should not match before
1440 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1441 never to match. This option affects only the behaviour of the circumflex
1442 metacharacter. It does not affect \eA.
1443 .sp
1444 PCRE_NOTEOL
1445 .sp
1446 This option specifies that the end of the subject string is not the end of a
1447 line, so the dollar metacharacter should not match it nor (except in multiline
1448 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1449 compile time) causes dollar never to match. This option affects only the
1450 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1451 .sp
1452 PCRE_NOTEMPTY
1453 .sp
1454 An empty string is not considered to be a valid match if this option is set. If
1455 there are alternatives in the pattern, they are tried. If all the alternatives
1456 match the empty string, the entire match fails. For example, if the pattern
1457 .sp
1458 a?b?
1459 .sp
1460 is applied to a string not beginning with "a" or "b", it matches an empty
1461 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1462 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1463 .sp
1464 PCRE_NOTEMPTY_ATSTART
1465 .sp
1466 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1467 the start of the subject is permitted. If the pattern is anchored, such a match
1468 can occur only if the pattern contains \eK.
1469 .P
1470 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1471 does make a special case of a pattern match of the empty string within its
1472 \fBsplit()\fP function, and when using the /g modifier. It is possible to
1473 emulate Perl's behaviour after matching a null string by first trying the match
1474 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1475 if that fails, by advancing the starting offset (see below) and trying an
1476 ordinary match again. There is some code that demonstrates how to do this in
1477 the
1478 .\" HREF
1479 \fBpcredemo\fP
1480 .\"
1481 sample program. In the most general case, you have to check to see if the
1482 newline convention recognizes CRLF as a newline, and if so, and the current
1483 character is CR followed by LF, advance the starting offset by two characters
1484 instead of one.
1485 .sp
1486 PCRE_NO_START_OPTIMIZE
1487 .sp
1488 There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1489 a match, in order to speed up the process. For example, if it is known that an
1490 unanchored match must start with a specific character, it searches the subject
1491 for that character, and fails immediately if it cannot find it, without
1492 actually running the main matching function. This means that a special item
1493 such as (*COMMIT) at the start of a pattern is not considered until after a
1494 suitable starting point for the match has been found. When callouts or (*MARK)
1495 items are in use, these "start-up" optimizations can cause them to be skipped
1496 if the pattern is never actually used. The start-up optimizations are in effect
1497 a pre-scan of the subject that takes place before the pattern is run.
1498 .P
1499 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1500 causing performance to suffer, but ensuring that in cases where the result is
1501 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1502 are considered at every possible starting position in the subject string. If
1503 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1504 time.
1505 .P
1506 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1507 Consider the pattern
1508 .sp
1509 (*COMMIT)ABC
1510 .sp
1511 When this is compiled, PCRE records the fact that a match must start with the
1512 character "A". Suppose the subject string is "DEFABC". The start-up
1513 optimization scans along the subject, finds "A" and runs the first match
1514 attempt from there. The (*COMMIT) item means that the pattern must match the
1515 current starting position, which in this case, it does. However, if the same
1516 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1517 subject string does not happen. The first match attempt is run starting from
1518 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1519 the overall result is "no match". If the pattern is studied, more start-up
1520 optimizations may be used. For example, a minimum length for the subject may be
1521 recorded. Consider the pattern
1522 .sp
1523 (*MARK:A)(X|Y)
1524 .sp
1525 The minimum length for a match is one character. If the subject is "ABC", there
1526 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1527 If the pattern is studied, the final attempt does not take place, because PCRE
1528 knows that the subject is too short, and so the (*MARK) is never encountered.
1529 In this case, studying the pattern does not affect the overall match result,
1530 which is still "no match", but it does affect the auxiliary information that is
1531 returned.
1532 .sp
1533 PCRE_NO_UTF8_CHECK
1534 .sp
1535 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1536 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1537 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1538 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1539 strings in the
1540 .\" HTML <a href="pcre.html#utf8strings">
1541 .\" </a>
1542 section on UTF-8 support
1543 .\"
1544 in the main
1545 .\" HREF
1546 \fBpcre\fP
1547 .\"
1548 page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1549 the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1550 a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. If
1551 \fIstartoffset\fP contains a value that does not point to the start of a UTF-8
1552 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1553 returned.
1554 .P
1555 If you already know that your subject is valid, and you want to skip these
1556 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1557 calling \fBpcre_exec()\fP. You might want to do this for the second and
1558 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1559 all the matches in a single subject string. However, you should be sure that
1560 the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1561 end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1562 invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1563 undefined. Your program may crash.
1564 .sp
1565 PCRE_PARTIAL_HARD
1566 PCRE_PARTIAL_SOFT
1567 .sp
1568 These options turn on the partial matching feature. For backwards
1569 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1570 occurs if the end of the subject string is reached successfully, but there are
1571 not enough subject characters to complete the match. If this happens when
1572 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1573 testing any remaining alternatives. Only if no complete match can be found is
1574 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1575 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1576 but only if no complete match can be found.
1577 .P
1578 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1579 partial match is found, \fBpcre_exec()\fP immediately returns
1580 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1581 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1582 important that an alternative complete match.
1583 .P
1584 In both cases, the portion of the string that was inspected when the partial
1585 match was found is set as the first matching string. There is a more detailed
1586 discussion of partial and multi-segment matching, with examples, in the
1587 .\" HREF
1588 \fBpcrepartial\fP
1589 .\"
1590 documentation.
1591 .
1592 .
1593 .SS "The string to be matched by \fBpcre_exec()\fP"
1594 .rs
1595 .sp
1596 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1597 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1598 in \fIstartoffset\fP. If this is negative or greater than the length of the
1599 subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1600 offset is zero, the search for a match starts at the beginning of the subject,
1601 and this is by far the most common case. In UTF-8 mode, the byte offset must
1602 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1603 pattern string, the subject may contain binary zero bytes.
1604 .P
1605 A non-zero starting offset is useful when searching for another match in the
1606 same subject by calling \fBpcre_exec()\fP again after a previous success.
1607 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1608 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1609 lookbehind. For example, consider the pattern
1610 .sp
1611 \eBiss\eB
1612 .sp
1613 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1614 the current position in the subject is not a word boundary.) When applied to
1615 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1616 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1617 subject, namely "issipi", it does not match, because \eB is always false at the
1618 start of the subject, which is deemed to be a word boundary. However, if
1619 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1620 set to 4, it finds the second occurrence of "iss" because it is able to look
1621 behind the starting point to discover that it is preceded by a letter.
1622 .P
1623 Finding all the matches in a subject is tricky when the pattern can match an
1624 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1625 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1626 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1627 and trying an ordinary match again. There is some code that demonstrates how to
1628 do this in the
1629 .\" HREF
1630 \fBpcredemo\fP
1631 .\"
1632 sample program. In the most general case, you have to check to see if the
1633 newline convention recognizes CRLF as a newline, and if so, and the current
1634 character is CR followed by LF, advance the starting offset by two characters
1635 instead of one.
1636 .P
1637 If a non-zero starting offset is passed when the pattern is anchored, one
1638 attempt to match at the given offset is made. This can only succeed if the
1639 pattern does not require the match to be at the start of the subject.
1640 .
1641 .
1642 .SS "How \fBpcre_exec()\fP returns captured substrings"
1643 .rs
1644 .sp
1645 In general, a pattern matches a certain portion of the subject, and in
1646 addition, further substrings from the subject may be picked out by parts of the
1647 pattern. Following the usage in Jeffrey Friedl's book, this is called
1648 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1649 a fragment of a pattern that picks out a substring. PCRE supports several other
1650 kinds of parenthesized subpattern that do not cause substrings to be captured.
1651 .P
1652 Captured substrings are returned to the caller via a vector of integers whose
1653 address is passed in \fIovector\fP. The number of elements in the vector is
1654 passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1655 argument is NOT the size of \fIovector\fP in bytes.
1656 .P
1657 The first two-thirds of the vector is used to pass back captured substrings,
1658 each substring using a pair of integers. The remaining third of the vector is
1659 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1660 and is not available for passing back information. The number passed in
1661 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1662 rounded down.
1663 .P
1664 When a match is successful, information about captured substrings is returned
1665 in pairs of integers, starting at the beginning of \fIovector\fP, and
1666 continuing up to two-thirds of its length at the most. The first element of
1667 each pair is set to the byte offset of the first character in a substring, and
1668 the second is set to the byte offset of the first character after the end of a
1669 substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1670 mode. They are not character counts.
1671 .P
1672 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1673 portion of the subject string matched by the entire pattern. The next pair is
1674 used for the first capturing subpattern, and so on. The value returned by
1675 \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1676 For example, if two substrings have been captured, the returned value is 3. If
1677 there are no capturing subpatterns, the return value from a successful match is
1678 1, indicating that just the first pair of offsets has been set.
1679 .P
1680 If a capturing subpattern is matched repeatedly, it is the last portion of the
1681 string that it matched that is returned.
1682 .P
1683 If the vector is too small to hold all the captured substring offsets, it is
1684 used as far as possible (up to two-thirds of its length), and the function
1685 returns a value of zero. If the substring offsets are not of interest,
1686 \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1687 \fIovecsize\fP as zero. However, if the pattern contains back references and
1688 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1689 has to get additional memory for use during matching. Thus it is usually
1690 advisable to supply an \fIovector\fP.
1691 .P
1692 The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1693 subpatterns there are in a compiled pattern. The smallest size for
1694 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1695 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1696 .P
1697 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1698 the subject when subpattern \fIn\fP has not been used at all. For example, if
1699 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1700 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1701 happens, both values in the offset pairs corresponding to unused subpatterns
1702 are set to -1.
1703 .P
1704 Offset values that correspond to unused subpatterns at the end of the
1705 expression are also set to -1. For example, if the string "abc" is matched
1706 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1707 return from the function is 2, because the highest used capturing subpattern
1708 number is 1, and the offsets for for the second and third capturing subpatterns
1709 (assuming the vector is large enough, of course) are set to -1.
1710 .P
1711 \fBNote\fP: Elements of \fIovector\fP that do not correspond to capturing
1712 parentheses in the pattern are never changed. That is, if a pattern contains
1713 \fIn\fP capturing parentheses, no more than \fIovector[0]\fP to
1714 \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other elements retain
1715 whatever values they previously had.
1716 .P
1717 Some convenience functions are provided for extracting the captured substrings
1718 as separate strings. These are described below.
1719 .
1720 .\" HTML <a name="errorlist"></a>
1721 .SS "Error return values from \fBpcre_exec()\fP"
1722 .rs
1723 .sp
1724 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1725 defined in the header file:
1726 .sp
1727 PCRE_ERROR_NOMATCH (-1)
1728 .sp
1729 The subject string did not match the pattern.
1730 .sp
1731 PCRE_ERROR_NULL (-2)
1732 .sp
1733 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1734 NULL and \fIovecsize\fP was not zero.
1735 .sp
1736 PCRE_ERROR_BADOPTION (-3)
1737 .sp
1738 An unrecognized bit was set in the \fIoptions\fP argument.
1739 .sp
1740 PCRE_ERROR_BADMAGIC (-4)
1741 .sp
1742 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1743 the case when it is passed a junk pointer and to detect when a pattern that was
1744 compiled in an environment of one endianness is run in an environment with the
1745 other endianness. This is the error that PCRE gives when the magic number is
1746 not present.
1747 .sp
1748 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1749 .sp
1750 While running the pattern match, an unknown item was encountered in the
1751 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1752 of the compiled pattern.
1753 .sp
1754 PCRE_ERROR_NOMEMORY (-6)
1755 .sp
1756 If a pattern contains back references, but the \fIovector\fP that is passed to
1757 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1758 gets a block of memory at the start of matching to use for this purpose. If the
1759 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1760 automatically freed at the end of matching.
1761 .P
1762 This error is also given if \fBpcre_stack_malloc()\fP fails in
1763 \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1764 \fB--disable-stack-for-recursion\fP.
1765 .sp
1766 PCRE_ERROR_NOSUBSTRING (-7)
1767 .sp
1768 This error is used by the \fBpcre_copy_substring()\fP,
1769 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1770 below). It is never returned by \fBpcre_exec()\fP.
1771 .sp
1772 PCRE_ERROR_MATCHLIMIT (-8)
1773 .sp
1774 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1775 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1776 above.
1777 .sp
1778 PCRE_ERROR_CALLOUT (-9)
1779 .sp
1780 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1781 use by callout functions that want to yield a distinctive error code. See the
1782 .\" HREF
1783 \fBpcrecallout\fP
1784 .\"
1785 documentation for details.
1786 .sp
1787 PCRE_ERROR_BADUTF8 (-10)
1788 .sp
1789 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1790 However, if PCRE_PARTIAL_HARD is set and the problem is a truncated UTF-8
1791 character at the end of the subject, PCRE_ERROR_SHORTUTF8 is used instead.
1792 .sp
1793 PCRE_ERROR_BADUTF8_OFFSET (-11)
1794 .sp
1795 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1796 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1797 end of the subject.
1798 .sp
1799 PCRE_ERROR_PARTIAL (-12)
1800 .sp
1801 The subject string did not match, but it did match partially. See the
1802 .\" HREF
1803 \fBpcrepartial\fP
1804 .\"
1805 documentation for details of partial matching.
1806 .sp
1807 PCRE_ERROR_BADPARTIAL (-13)
1808 .sp
1809 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1810 option was used with a compiled pattern containing items that were not
1811 supported for partial matching. From release 8.00 onwards, there are no
1812 restrictions on partial matching.
1813 .sp
1814 PCRE_ERROR_INTERNAL (-14)
1815 .sp
1816 An unexpected internal error has occurred. This error could be caused by a bug
1817 in PCRE or by overwriting of the compiled pattern.
1818 .sp
1819 PCRE_ERROR_BADCOUNT (-15)
1820 .sp
1821 This error is given if the value of the \fIovecsize\fP argument is negative.
1822 .sp
1823 PCRE_ERROR_RECURSIONLIMIT (-21)
1824 .sp
1825 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1826 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1827 description above.
1828 .sp
1829 PCRE_ERROR_BADNEWLINE (-23)
1830 .sp
1831 An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1832 .sp
1833 PCRE_ERROR_BADOFFSET (-24)
1834 .sp
1835 The value of \fIstartoffset\fP was negative or greater than the length of the
1836 subject, that is, the value in \fIlength\fP.
1837 .sp
1838 PCRE_ERROR_SHORTUTF8 (-25)
1839 .sp
1840 The subject string ended with an incomplete (truncated) UTF-8 character, and
1841 the PCRE_PARTIAL_HARD option was set. Without this option, PCRE_ERROR_BADUTF8
1842 is returned in this situation.
1843 .P
1844 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1845 .
1846 .
1847 .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1848 .rs
1849 .sp
1850 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1851 .ti +5n
1852 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1853 .ti +5n
1854 .B int \fIbuffersize\fP);
1855 .PP
1856 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1857 .ti +5n
1858 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1859 .ti +5n
1860 .B const char **\fIstringptr\fP);
1861 .PP
1862 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1863 .ti +5n
1864 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1865 .PP
1866 Captured substrings can be accessed directly by using the offsets returned by
1867 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1868 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1869 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1870 as new, separate, zero-terminated strings. These functions identify substrings
1871 by number. The next section describes functions for extracting named
1872 substrings.
1873 .P
1874 A substring that contains a binary zero is correctly extracted and has a
1875 further zero added on the end, but the result is not, of course, a C string.
1876 However, you can process such a string by referring to the length that is
1877 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1878 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1879 for handling strings containing binary zeros, because the end of the final
1880 string is not independently indicated.
1881 .P
1882 The first three arguments are the same for all three of these functions:
1883 \fIsubject\fP is the subject string that has just been successfully matched,
1884 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1885 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1886 captured by the match, including the substring that matched the entire regular
1887 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1888 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1889 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1890 number of elements in the vector divided by three.
1891 .P
1892 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1893 extract a single substring, whose number is given as \fIstringnumber\fP. A
1894 value of zero extracts the substring that matched the entire pattern, whereas
1895 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1896 the string is placed in \fIbuffer\fP, whose length is given by
1897 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1898 obtained via \fBpcre_malloc\fP, and its address is returned via
1899 \fIstringptr\fP. The yield of the function is the length of the string, not
1900 including the terminating zero, or one of these error codes:
1901 .sp
1902 PCRE_ERROR_NOMEMORY (-6)
1903 .sp
1904 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1905 memory failed for \fBpcre_get_substring()\fP.
1906 .sp
1907 PCRE_ERROR_NOSUBSTRING (-7)
1908 .sp
1909 There is no substring whose number is \fIstringnumber\fP.
1910 .P
1911 The \fBpcre_get_substring_list()\fP function extracts all available substrings
1912 and builds a list of pointers to them. All this is done in a single block of
1913 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1914 is returned via \fIlistptr\fP, which is also the start of the list of string
1915 pointers. The end of the list is marked by a NULL pointer. The yield of the
1916 function is zero if all went well, or the error code
1917 .sp
1918 PCRE_ERROR_NOMEMORY (-6)
1919 .sp
1920 if the attempt to get the memory block failed.
1921 .P
1922 When any of these functions encounter a substring that is unset, which can
1923 happen when capturing subpattern number \fIn+1\fP matches some part of the
1924 subject, but subpattern \fIn\fP has not been used at all, they return an empty
1925 string. This can be distinguished from a genuine zero-length substring by
1926 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1927 substrings.
1928 .P
1929 The two convenience functions \fBpcre_free_substring()\fP and
1930 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1931 a previous call of \fBpcre_get_substring()\fP or
1932 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1933 the function pointed to by \fBpcre_free\fP, which of course could be called
1934 directly from a C program. However, PCRE is used in some situations where it is
1935 linked via a special interface to another programming language that cannot use
1936 \fBpcre_free\fP directly; it is for these cases that the functions are
1937 provided.
1938 .
1939 .
1940 .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1941 .rs
1942 .sp
1943 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1944 .ti +5n
1945 .B const char *\fIname\fP);
1946 .PP
1947 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1948 .ti +5n
1949 .B const char *\fIsubject\fP, int *\fIovector\fP,
1950 .ti +5n
1951 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1952 .ti +5n
1953 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1954 .PP
1955 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1956 .ti +5n
1957 .B const char *\fIsubject\fP, int *\fIovector\fP,
1958 .ti +5n
1959 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1960 .ti +5n
1961 .B const char **\fIstringptr\fP);
1962 .PP
1963 To extract a substring by name, you first have to find associated number.
1964 For example, for this pattern
1965 .sp
1966 (a+)b(?<xxx>\ed+)...
1967 .sp
1968 the number of the subpattern called "xxx" is 2. If the name is known to be
1969 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1970 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1971 pattern, and the second is the name. The yield of the function is the
1972 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1973 that name.
1974 .P
1975 Given the number, you can extract the substring directly, or use one of the
1976 functions described in the previous section. For convenience, there are also
1977 two functions that do the whole job.
1978 .P
1979 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1980 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1981 functions that extract by number. As these are described in the previous
1982 section, they are not re-described here. There are just two differences:
1983 .P
1984 First, instead of a substring number, a substring name is given. Second, there
1985 is an extra argument, given at the start, which is a pointer to the compiled
1986 pattern. This is needed in order to gain access to the name-to-number
1987 translation table.
1988 .P
1989 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1990 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1991 appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1992 the behaviour may not be what you want (see the next section).
1993 .P
1994 \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
1995 subpatterns with the same number, as described in the
1996 .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1997 .\" </a>
1998 section on duplicate subpattern numbers
1999 .\"
2000 in the
2001 .\" HREF
2002 \fBpcrepattern\fP
2003 .\"
2004 page, you cannot use names to distinguish the different subpatterns, because
2005 names are not included in the compiled code. The matching process uses only
2006 numbers. For this reason, the use of different names for subpatterns of the
2007 same number causes an error at compile time.
2008 .
2009 .SH "DUPLICATE SUBPATTERN NAMES"
2010 .rs
2011 .sp
2012 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2013 .ti +5n
2014 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2015 .PP
2016 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2017 are not required to be unique. (Duplicate names are always allowed for
2018 subpatterns with the same number, created by using the (?| feature. Indeed, if
2019 such subpatterns are named, they are required to use the same names.)
2020 .P
2021 Normally, patterns with duplicate names are such that in any one match, only
2022 one of the named subpatterns participates. An example is shown in the
2023 .\" HREF
2024 \fBpcrepattern\fP
2025 .\"
2026 documentation.
2027 .P
2028 When duplicates are present, \fBpcre_copy_named_substring()\fP and
2029 \fBpcre_get_named_substring()\fP return the first substring corresponding to
2030 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2031 returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2032 returns one of the numbers that are associated with the name, but it is not
2033 defined which it is.
2034 .P
2035 If you want to get full details of all captured substrings for a given name,
2036 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2037 argument is the compiled pattern, and the second is the name. The third and
2038 fourth are pointers to variables which are updated by the function. After it
2039 has run, they point to the first and last entries in the name-to-number table
2040 for the given name. The function itself returns the length of each entry, or
2041 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2042 described above in the section entitled \fIInformation about a pattern\fP.
2043 Given all the relevant entries for the name, you can extract each of their
2044 numbers, and hence the captured data, if any.
2045 .
2046 .
2047 .SH "FINDING ALL POSSIBLE MATCHES"
2048 .rs
2049 .sp
2050 The traditional matching function uses a similar algorithm to Perl, which stops
2051 when it finds the first match, starting at a given point in the subject. If you
2052 want to find all possible matches, or the longest possible match, consider
2053 using the alternative matching function (see below) instead. If you cannot use
2054 the alternative function, but still need to find all possible matches, you
2055 can kludge it up by making use of the callout facility, which is described in
2056 the
2057 .\" HREF
2058 \fBpcrecallout\fP
2059 .\"
2060 documentation.
2061 .P
2062 What you have to do is to insert a callout right at the end of the pattern.
2063 When your callout function is called, extract and save the current matched
2064 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2065 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2066 will yield PCRE_ERROR_NOMATCH.
2067 .
2068 .
2069 .\" HTML <a name="dfamatch"></a>
2070 .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2071 .rs
2072 .sp
2073 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2074 .ti +5n
2075 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2076 .ti +5n
2077 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2078 .ti +5n
2079 .B int *\fIworkspace\fP, int \fIwscount\fP);
2080 .P
2081 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2082 a compiled pattern, using a matching algorithm that scans the subject string
2083 just once, and does not backtrack. This has different characteristics to the
2084 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2085 patterns are not supported. Nevertheless, there are times when this kind of
2086 matching can be useful. For a discussion of the two matching algorithms, and a
2087 list of features that \fBpcre_dfa_exec()\fP does not support, see the
2088 .\" HREF
2089 \fBpcrematching\fP
2090 .\"
2091 documentation.
2092 .P
2093 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2094 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2095 different way, and this is described below. The other common arguments are used
2096 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2097 here.
2098 .P
2099 The two additional arguments provide workspace for the function. The workspace
2100 vector should contain at least 20 elements. It is used for keeping track of
2101 multiple paths through the pattern tree. More workspace will be needed for
2102 patterns and subjects where there are a lot of potential matches.
2103 .P
2104 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2105 .sp
2106 int rc;
2107 int ovector[10];
2108 int wspace[20];
2109 rc = pcre_dfa_exec(
2110 re, /* result of pcre_compile() */
2111 NULL, /* we didn't study the pattern */
2112 "some string", /* the subject string */
2113 11, /* the length of the subject string */
2114 0, /* start at offset 0 in the subject */
2115 0, /* default options */
2116 ovector, /* vector of integers for substring information */
2117 10, /* number of elements (NOT size in bytes) */
2118 wspace, /* working space vector */
2119 20); /* number of elements (NOT size in bytes) */
2120 .
2121 .SS "Option bits for \fBpcre_dfa_exec()\fP"
2122 .rs
2123 .sp
2124 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2125 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2126 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2127 PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2128 PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2129 All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2130 so their description is not repeated here.
2131 .sp
2132 PCRE_PARTIAL_HARD
2133 PCRE_PARTIAL_SOFT
2134 .sp
2135 These have the same general effect as they do for \fBpcre_exec()\fP, but the
2136 details are slightly different. When PCRE_PARTIAL_HARD is set for
2137 \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2138 is reached and there is still at least one matching possibility that requires
2139 additional characters. This happens even if some complete matches have also
2140 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2141 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2142 there have been no complete matches, but there is still at least one matching
2143 possibility. The portion of the string that was inspected when the longest
2144 partial match was found is set as the first matching string in both cases.
2145 There is a more detailed discussion of partial and multi-segment matching, with
2146 examples, in the
2147 .\" HREF
2148 \fBpcrepartial\fP
2149 .\"
2150 documentation.
2151 .sp
2152 PCRE_DFA_SHORTEST
2153 .sp
2154 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2155 soon as it has found one match. Because of the way the alternative algorithm
2156 works, this is necessarily the shortest possible match at the first possible
2157 matching point in the subject string.
2158 .sp
2159 PCRE_DFA_RESTART
2160 .sp
2161 When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2162 again, with additional subject characters, and have it continue with the same
2163 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2164 \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2165 before because data about the match so far is left in them after a partial
2166 match. There is more discussion of this facility in the
2167 .\" HREF
2168 \fBpcrepartial\fP
2169 .\"
2170 documentation.
2171 .
2172 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2173 .rs
2174 .sp
2175 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2176 substring in the subject. Note, however, that all the matches from one run of
2177 the function start at the same point in the subject. The shorter matches are
2178 all initial substrings of the longer matches. For example, if the pattern
2179 .sp
2180 <.*>
2181 .sp
2182 is matched against the string
2183 .sp
2184 This is <something> <something else> <something further> no more
2185 .sp
2186 the three matched strings are
2187 .sp
2188 <something>
2189 <something> <something else>
2190 <something> <something else> <something further>
2191 .sp
2192 On success, the yield of the function is a number greater than zero, which is
2193 the number of matched substrings. The substrings themselves are returned in
2194 \fIovector\fP. Each string uses two elements; the first is the offset to the
2195 start, and the second is the offset to the end. In fact, all the strings have
2196 the same start offset. (Space could have been saved by giving this only once,
2197 but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2198 returns data, even though the meaning of the strings is different.)
2199 .P
2200 The strings are returned in reverse order of length; that is, the longest
2201 matching string is given first. If there were too many matches to fit into
2202 \fIovector\fP, the yield of the function is zero, and the vector is filled with
2203 the longest matches.
2204 .
2205 .SS "Error returns from \fBpcre_dfa_exec()\fP"
2206 .rs
2207 .sp
2208 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2209 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2210 described
2211 .\" HTML <a href="#errorlist">
2212 .\" </a>
2213 above.
2214 .\"
2215 There are in addition the following errors that are specific to
2216 \fBpcre_dfa_exec()\fP:
2217 .sp
2218 PCRE_ERROR_DFA_UITEM (-16)
2219 .sp
2220 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2221 that it does not support, for instance, the use of \eC or a back reference.
2222 .sp
2223 PCRE_ERROR_DFA_UCOND (-17)
2224 .sp
2225 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2226 uses a back reference for the condition, or a test for recursion in a specific
2227 group. These are not supported.
2228 .sp
2229 PCRE_ERROR_DFA_UMLIMIT (-18)
2230 .sp
2231 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2232 block that contains a setting of the \fImatch_limit\fP field. This is not
2233 supported (it is meaningless).
2234 .sp
2235 PCRE_ERROR_DFA_WSSIZE (-19)
2236 .sp
2237 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2238 \fIworkspace\fP vector.
2239 .sp
2240 PCRE_ERROR_DFA_RECURSE (-20)
2241 .sp
2242 When a recursive subpattern is processed, the matching function calls itself
2243 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2244 error is given if the output vector is not large enough. This should be
2245 extremely rare, as a vector of size 1000 is used.
2246 .
2247 .
2248 .SH "SEE ALSO"
2249 .rs
2250 .sp
2251 \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2252 \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2253 \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2254 .
2255 .
2256 .SH AUTHOR
2257 .rs
2258 .sp
2259 .nf
2260 Philip Hazel
2261 University Computing Service
2262 Cambridge CB2 3QH, England.
2263 .fi
2264 .
2265 .
2266 .SH REVISION
2267 .rs
2268 .sp
2269 .nf
2270 Last updated: 21 November 2010
2271 Copyright (c) 1997-2010 University of Cambridge.
2272 .fi

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5