ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 91 - (show annotations)
Sat Feb 24 21:41:34 2007 UTC (14 years, 2 months ago) by nigel
File size: 75554 byte(s)
Load pcre-6.7 into code/trunk.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 .B #include <pcre.h>
8 .PP
9 .SM
10 .br
11 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
12 .ti +5n
13 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
14 .ti +5n
15 .B const unsigned char *\fItableptr\fP);
16 .PP
17 .br
18 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19 .ti +5n
20 .B int *\fIerrorcodeptr\fP,
21 .ti +5n
22 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
23 .ti +5n
24 .B const unsigned char *\fItableptr\fP);
25 .PP
26 .br
27 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
28 .ti +5n
29 .B const char **\fIerrptr\fP);
30 .PP
31 .br
32 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
33 .ti +5n
34 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
35 .ti +5n
36 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37 .PP
38 .br
39 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
40 .ti +5n
41 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
42 .ti +5n
43 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
44 .ti +5n
45 .B int *\fIworkspace\fP, int \fIwscount\fP);
46 .PP
47 .br
48 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
49 .ti +5n
50 .B const char *\fIsubject\fP, int *\fIovector\fP,
51 .ti +5n
52 .B int \fIstringcount\fP, const char *\fIstringname\fP,
53 .ti +5n
54 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
55 .PP
56 .br
57 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
58 .ti +5n
59 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
60 .ti +5n
61 .B int \fIbuffersize\fP);
62 .PP
63 .br
64 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
65 .ti +5n
66 .B const char *\fIsubject\fP, int *\fIovector\fP,
67 .ti +5n
68 .B int \fIstringcount\fP, const char *\fIstringname\fP,
69 .ti +5n
70 .B const char **\fIstringptr\fP);
71 .PP
72 .br
73 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
74 .ti +5n
75 .B const char *\fIname\fP);
76 .PP
77 .br
78 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
79 .ti +5n
80 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
81 .PP
82 .br
83 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
84 .ti +5n
85 .B int \fIstringcount\fP, int \fIstringnumber\fP,
86 .ti +5n
87 .B const char **\fIstringptr\fP);
88 .PP
89 .br
90 .B int pcre_get_substring_list(const char *\fIsubject\fP,
91 .ti +5n
92 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
93 .PP
94 .br
95 .B void pcre_free_substring(const char *\fIstringptr\fP);
96 .PP
97 .br
98 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
99 .PP
100 .br
101 .B const unsigned char *pcre_maketables(void);
102 .PP
103 .br
104 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
105 .ti +5n
106 .B int \fIwhat\fP, void *\fIwhere\fP);
107 .PP
108 .br
109 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
110 .B *\fIfirstcharptr\fP);
111 .PP
112 .br
113 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
114 .PP
115 .br
116 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
117 .PP
118 .br
119 .B char *pcre_version(void);
120 .PP
121 .br
122 .B void *(*pcre_malloc)(size_t);
123 .PP
124 .br
125 .B void (*pcre_free)(void *);
126 .PP
127 .br
128 .B void *(*pcre_stack_malloc)(size_t);
129 .PP
130 .br
131 .B void (*pcre_stack_free)(void *);
132 .PP
133 .br
134 .B int (*pcre_callout)(pcre_callout_block *);
135 .
136 .
138 .rs
139 .sp
140 PCRE has its own native API, which is described in this document. There is
141 also a set of wrapper functions that correspond to the POSIX regular expression
142 API. These are described in the
143 .\" HREF
144 \fBpcreposix\fP
145 .\"
146 documentation. Both of these APIs define a set of C function calls. A C++
147 wrapper is distributed with PCRE. It is documented in the
148 .\" HREF
149 \fBpcrecpp\fP
150 .\"
151 page.
152 .P
153 The native API C function prototypes are defined in the header file
154 \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
155 It can normally be accessed by adding \fB-lpcre\fP to the command for linking
156 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
157 and PCRE_MINOR to contain the major and minor release numbers for the library.
158 Applications can use these to include support for different releases of PCRE.
159 .P
160 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
161 and \fBpcre_exec()\fP are used for compiling and matching regular expressions
162 in a Perl-compatible manner. A sample program that demonstrates the simplest
163 way of using them is provided in the file called \fIpcredemo.c\fP in the source
164 distribution. The
165 .\" HREF
166 \fBpcresample\fP
167 .\"
168 documentation describes how to run it.
169 .P
170 A second matching function, \fBpcre_dfa_exec()\fP, which is not
171 Perl-compatible, is also provided. This uses a different algorithm for the
172 matching. The alternative algorithm finds all possible matches (at a given
173 point in the subject). However, this algorithm does not return captured
174 substrings. A description of the two matching algorithms and their advantages
175 and disadvantages is given in the
176 .\" HREF
177 \fBpcrematching\fP
178 .\"
179 documentation.
180 .P
181 In addition to the main compiling and matching functions, there are convenience
182 functions for extracting captured substrings from a subject string that is
183 matched by \fBpcre_exec()\fP. They are:
184 .sp
185 \fBpcre_copy_substring()\fP
186 \fBpcre_copy_named_substring()\fP
187 \fBpcre_get_substring()\fP
188 \fBpcre_get_named_substring()\fP
189 \fBpcre_get_substring_list()\fP
190 \fBpcre_get_stringnumber()\fP
191 \fBpcre_get_stringtable_entries()\fP
192 .sp
193 \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
194 provided, to free the memory used for extracted strings.
195 .P
196 The function \fBpcre_maketables()\fP is used to build a set of character tables
197 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
198 or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
199 specialist use. Most commonly, no special tables are passed, in which case
200 internal tables that are generated when PCRE is built are used.
201 .P
202 The function \fBpcre_fullinfo()\fP is used to find out information about a
203 compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
204 some of the available information, but is retained for backwards compatibility.
205 The function \fBpcre_version()\fP returns a pointer to a string containing the
206 version of PCRE and its date of release.
207 .P
208 The function \fBpcre_refcount()\fP maintains a reference count in a data block
209 containing a compiled pattern. This is provided for the benefit of
210 object-oriented applications.
211 .P
212 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
213 the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
214 respectively. PCRE calls the memory management functions via these variables,
215 so a calling program can replace them if it wishes to intercept the calls. This
216 should be done before calling any PCRE functions.
217 .P
218 The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
219 indirections to memory management functions. These special functions are used
220 only when PCRE is compiled to use the heap for remembering data, instead of
221 recursive function calls, when running the \fBpcre_exec()\fP function. See the
222 .\" HREF
223 \fBpcrebuild\fP
224 .\"
225 documentation for details of how to do this. It is a non-standard way of
226 building PCRE, for use in environments that have limited stacks. Because of the
227 greater use of memory management, it runs more slowly. Separate functions are
228 provided so that special-purpose external code can be used for this case. When
229 used, these functions are always called in a stack-like manner (last obtained,
230 first freed), and always for memory blocks of the same size. There is a
231 discussion about PCRE's stack usage in the
232 .\" HREF
233 \fBpcrestack\fP
234 .\"
235 documentation.
236 .P
237 The global variable \fBpcre_callout\fP initially contains NULL. It can be set
238 by the caller to a "callout" function, which PCRE will then call at specified
239 points during a matching operation. Details are given in the
240 .\" HREF
241 \fBpcrecallout\fP
242 .\"
243 documentation.
244 .
245 .
247 PCRE supports three different conventions for indicating line breaks in
248 strings: a single CR character, a single LF character, or the two-character
249 sequence CRLF. All three are used as "standard" by different operating systems.
250 When PCRE is built, a default can be specified. The default default is LF,
251 which is the Unix standard. When PCRE is run, the default can be overridden,
252 either when a pattern is compiled, or when it is matched.
253 .sp
254 In the PCRE documentation the word "newline" is used to mean "the character or
255 pair of characters that indicate a line break".
256 .
257 .
259 .rs
260 .sp
261 The PCRE functions can be used in multi-threading applications, with the
262 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
263 \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
264 callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
265 .P
266 The compiled form of a regular expression is not altered during matching, so
267 the same compiled pattern can safely be used by several threads at once.
268 .
269 .
271 .rs
272 .sp
273 The compiled form of a regular expression can be saved and re-used at a later
274 time, possibly by a different program, and even on a host other than the one on
275 which it was compiled. Details are given in the
276 .\" HREF
277 \fBpcreprecompile\fP
278 .\"
279 documentation.
280 .
281 .
283 .rs
284 .sp
285 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
286 .PP
287 The function \fBpcre_config()\fP makes it possible for a PCRE client to
288 discover which optional features have been compiled into the PCRE library. The
289 .\" HREF
290 \fBpcrebuild\fP
291 .\"
292 documentation has more details about these optional features.
293 .P
294 The first argument for \fBpcre_config()\fP is an integer, specifying which
295 information is required; the second argument is a pointer to a variable into
296 which the information is placed. The following information is available:
297 .sp
299 .sp
300 The output is an integer that is set to one if UTF-8 support is available;
301 otherwise it is set to zero.
302 .sp
304 .sp
305 The output is an integer that is set to one if support for Unicode character
306 properties is available; otherwise it is set to zero.
307 .sp
309 .sp
310 The output is an integer whose value specifies the default character sequence
311 that is recognized as meaning "newline". The three values that are supported
312 are: 10 for LF, 13 for CR, and 3338 for CRLF. The default should normally be
313 the standard sequence for your operating system.
314 .sp
316 .sp
317 The output is an integer that contains the number of bytes used for internal
318 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
319 allow larger regular expressions to be compiled, at the expense of slower
320 matching. The default value of 2 is sufficient for all but the most massive
321 patterns, since it allows the compiled pattern to be up to 64K in size.
322 .sp
324 .sp
325 The output is an integer that contains the threshold above which the POSIX
326 interface uses \fBmalloc()\fP for output vectors. Further details are given in
327 the
328 .\" HREF
329 \fBpcreposix\fP
330 .\"
331 documentation.
332 .sp
334 .sp
335 The output is an integer that gives the default limit for the number of
336 internal matching function calls in a \fBpcre_exec()\fP execution. Further
337 details are given with \fBpcre_exec()\fP below.
338 .sp
340 .sp
341 The output is an integer that gives the default limit for the depth of
342 recursion when calling the internal matching function in a \fBpcre_exec()\fP
343 execution. Further details are given with \fBpcre_exec()\fP below.
344 .sp
346 .sp
347 The output is an integer that is set to one if internal recursion when running
348 \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
349 to remember their state. This is the usual way that PCRE is compiled. The
350 output is zero if PCRE was compiled to use blocks of data on the heap instead
351 of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
352 \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
353 avoiding the use of the stack.
354 .
355 .
357 .rs
358 .sp
359 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
360 .ti +5n
361 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
362 .ti +5n
363 .B const unsigned char *\fItableptr\fP);
364 .sp
365 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
366 .ti +5n
367 .B int *\fIerrorcodeptr\fP,
368 .ti +5n
369 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
370 .ti +5n
371 .B const unsigned char *\fItableptr\fP);
372 .P
373 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
374 called to compile a pattern into an internal form. The only difference between
375 the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
376 \fIerrorcodeptr\fP, via which a numerical error code can be returned.
377 .P
378 The pattern is a C string terminated by a binary zero, and is passed in the
379 \fIpattern\fP argument. A pointer to a single block of memory that is obtained
380 via \fBpcre_malloc\fP is returned. This contains the compiled code and related
381 data. The \fBpcre\fP type is defined for the returned block; this is a typedef
382 for a structure whose contents are not externally defined. It is up to the
383 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
384 .P
385 Although the compiled code of a PCRE regex is relocatable, that is, it does not
386 depend on memory location, the complete \fBpcre\fP data block is not
387 fully relocatable, because it may contain a copy of the \fItableptr\fP
388 argument, which is an address (see below).
389 .P
390 The \fIoptions\fP argument contains independent bits that affect the
391 compilation. It should be zero if no options are required. The available
392 options are described below. Some of them, in particular, those that are
393 compatible with Perl, can also be set and unset from within the pattern (see
394 the detailed description in the
395 .\" HREF
396 \fBpcrepattern\fP
397 .\"
398 documentation). For these options, the contents of the \fIoptions\fP argument
399 specifies their initial settings at the start of compilation and execution. The
400 PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
401 matching as well as at compile time.
402 .P
403 If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
404 Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
405 NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
406 error message. This is a static string that is part of the library. You must
407 not try to free it. The offset from the start of the pattern to the character
408 where the error was discovered is placed in the variable pointed to by
409 \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
410 .P
411 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
412 \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
413 returned via this argument in the event of an error. This is in addition to the
414 textual error message. Error codes and messages are listed below.
415 .P
416 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
417 character tables that are built when PCRE is compiled, using the default C
418 locale. Otherwise, \fItableptr\fP must be an address that is the result of a
419 call to \fBpcre_maketables()\fP. This value is stored with the compiled
420 pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
421 passed to it. For more discussion, see the section on locale support below.
422 .P
423 This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
424 .sp
425 pcre *re;
426 const char *error;
427 int erroffset;
428 re = pcre_compile(
429 "^A.*Z", /* the pattern */
430 0, /* default options */
431 &error, /* for error message */
432 &erroffset, /* for error offset */
433 NULL); /* use default character tables */
434 .sp
435 The following names for option bits are defined in the \fBpcre.h\fP header
436 file:
437 .sp
439 .sp
440 If this bit is set, the pattern is forced to be "anchored", that is, it is
441 constrained to match only at the first matching point in the string that is
442 being searched (the "subject string"). This effect can also be achieved by
443 appropriate constructs in the pattern itself, which is the only way to do it in
444 Perl.
445 .sp
447 .sp
448 If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
449 all with number 255, before each pattern item. For discussion of the callout
450 facility, see the
451 .\" HREF
452 \fBpcrecallout\fP
453 .\"
454 documentation.
455 .sp
457 .sp
458 If this bit is set, letters in the pattern match both upper and lower case
459 letters. It is equivalent to Perl's /i option, and it can be changed within a
460 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
461 concept of case for characters whose values are less than 128, so caseless
462 matching is always possible. For characters with higher values, the concept of
463 case is supported if PCRE is compiled with Unicode property support, but not
464 otherwise. If you want to use caseless matching for characters 128 and above,
465 you must ensure that PCRE is compiled with Unicode property support as well as
466 with UTF-8 support.
467 .sp
469 .sp
470 If this bit is set, a dollar metacharacter in the pattern matches only at the
471 end of the subject string. Without this option, a dollar also matches
472 immediately before a newline at the end of the string (but not before any other
473 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
474 There is no equivalent to this option in Perl, and no way to set it within a
475 pattern.
476 .sp
478 .sp
479 If this bit is set, a dot metacharater in the pattern matches all characters,
480 including those that indicate newline. Without it, a dot does not match when
481 the current position is at a newline. This option is equivalent to Perl's /s
482 option, and it can be changed within a pattern by a (?s) option setting. A
483 negative class such as [^a] always matches newlines, independent of the setting
484 of this option.
485 .sp
487 .sp
488 If this bit is set, names used to identify capturing subpatterns need not be
489 unique. This can be helpful for certain types of pattern when it is known that
490 only one instance of the named subpattern can ever be matched. There are more
491 details of named subpatterns below; see also the
492 .\" HREF
493 \fBpcrepattern\fP
494 .\"
495 documentation.
496 .sp
498 .sp
499 If this bit is set, whitespace data characters in the pattern are totally
500 ignored except when escaped or inside a character class. Whitespace does not
501 include the VT character (code 11). In addition, characters between an
502 unescaped # outside a character class and the next newline, inclusive, are also
503 ignored. This is equivalent to Perl's /x option, and it can be changed within a
504 pattern by a (?x) option setting.
505 .P
506 This option makes it possible to include comments inside complicated patterns.
507 Note, however, that this applies only to data characters. Whitespace characters
508 may never appear within special character sequences in a pattern, for example
509 within the sequence (?( which introduces a conditional subpattern.
510 .sp
512 .sp
513 This option was invented in order to turn on additional functionality of PCRE
514 that is incompatible with Perl, but it is currently of very little use. When
515 set, any backslash in a pattern that is followed by a letter that has no
516 special meaning causes an error, thus reserving these combinations for future
517 expansion. By default, as in Perl, a backslash followed by a letter with no
518 special meaning is treated as a literal. (Perl can, however, be persuaded to
519 give a warning for this.) There are at present no other features controlled by
520 this option. It can also be set by a (?X) option setting within a pattern.
521 .sp
523 .sp
524 If this option is set, an unanchored pattern is required to match before or at
525 the first newline in the subject string, though the matched text may continue
526 over the newline.
527 .sp
529 .sp
530 By default, PCRE treats the subject string as consisting of a single line of
531 characters (even if it actually contains newlines). The "start of line"
532 metacharacter (^) matches only at the start of the string, while the "end of
533 line" metacharacter ($) matches only at the end of the string, or before a
534 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
535 Perl.
536 .P
537 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
538 match immediately following or immediately before internal newlines in the
539 subject string, respectively, as well as at the very start and end. This is
540 equivalent to Perl's /m option, and it can be changed within a pattern by a
541 (?m) option setting. If there are no newlines in a subject string, or no
542 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
543 .sp
547 .sp
548 These options override the default newline definition that was chosen when PCRE
549 was built. Setting the first or the second specifies that a newline is
550 indicated by a single character (CR or LF, respectively). Setting both of them
551 specifies that a newline is indicated by the two-character CRLF sequence. For
552 convenience, PCRE_NEWLINE_CRLF is defined to contain both bits. The only time
553 that a line break is relevant when compiling a pattern is if PCRE_EXTENDED is
554 set, and an unescaped # outside a character class is encountered. This
555 indicates a comment that lasts until after the next newline.
556 .P
557 The newline option set at compile time becomes the default that is used for
558 \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
559 .sp
561 .sp
562 If this option is set, it disables the use of numbered capturing parentheses in
563 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
564 were followed by ?: but named parentheses can still be used for capturing (and
565 they acquire numbers in the usual way). There is no equivalent of this option
566 in Perl.
567 .sp
569 .sp
570 This option inverts the "greediness" of the quantifiers so that they are not
571 greedy by default, but become greedy if followed by "?". It is not compatible
572 with Perl. It can also be set by a (?U) option setting within the pattern.
573 .sp
575 .sp
576 This option causes PCRE to regard both the pattern and the subject as strings
577 of UTF-8 characters instead of single-byte character strings. However, it is
578 available only when PCRE is built to include UTF-8 support. If not, the use
579 of this option provokes an error. Details of how this option changes the
580 behaviour of PCRE are given in the
581 .\" HTML <a href="pcre.html#utf8support">
582 .\" </a>
583 section on UTF-8 support
584 .\"
585 in the main
586 .\" HREF
587 \fBpcre\fP
588 .\"
589 page.
590 .sp
592 .sp
593 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
594 automatically checked. If an invalid UTF-8 sequence of bytes is found,
595 \fBpcre_compile()\fP returns an error. If you already know that your pattern is
596 valid, and you want to skip this check for performance reasons, you can set the
597 PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
598 UTF-8 string as a pattern is undefined. It may cause your program to crash.
599 Note that this option can also be passed to \fBpcre_exec()\fP and
600 \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject
601 strings.
602 .
603 .
605 .rs
606 .sp
607 The following table lists the error codes than may be returned by
608 \fBpcre_compile2()\fP, along with the error messages that may be returned by
609 both compiling functions.
610 .sp
611 0 no error
612 1 \e at end of pattern
613 2 \ec at end of pattern
614 3 unrecognized character follows \e
615 4 numbers out of order in {} quantifier
616 5 number too big in {} quantifier
617 6 missing terminating ] for character class
618 7 invalid escape sequence in character class
619 8 range out of order in character class
620 9 nothing to repeat
621 10 operand of unlimited repeat could match the empty string
622 11 internal error: unexpected repeat
623 12 unrecognized character after (?
624 13 POSIX named classes are supported only within a class
625 14 missing )
626 15 reference to non-existent subpattern
627 16 erroffset passed as NULL
628 17 unknown option bit(s) set
629 18 missing ) after comment
630 19 parentheses nested too deeply
631 20 regular expression too large
632 21 failed to get memory
633 22 unmatched parentheses
634 23 internal error: code overflow
635 24 unrecognized character after (?<
636 25 lookbehind assertion is not fixed length
637 26 malformed number or name after (?(
638 27 conditional group contains more than two branches
639 28 assertion expected after (?(
640 29 (?R or (?digits must be followed by )
641 30 unknown POSIX class name
642 31 POSIX collating elements are not supported
643 32 this version of PCRE is not compiled with PCRE_UTF8 support
644 33 spare error
645 34 character value in \ex{...} sequence is too large
646 35 invalid condition (?(0)
647 36 \eC not allowed in lookbehind assertion
648 37 PCRE does not support \eL, \el, \eN, \eU, or \eu
649 38 number after (?C is > 255
650 39 closing ) for (?C expected
651 40 recursive call could loop indefinitely
652 41 unrecognized character after (?P
653 42 syntax error after (?P
654 43 two named subpatterns have the same name
655 44 invalid UTF-8 string
656 45 support for \eP, \ep, and \eX has not been compiled
657 46 malformed \eP or \ep sequence
658 47 unknown property name after \eP or \ep
659 48 subpattern name is too long (maximum 32 characters)
660 49 too many named subpatterns (maximum 10,000)
661 50 repeated subpattern is too long
662 51 octal value is greater than \e377 (not in UTF-8 mode)
663 .
664 .
666 .rs
667 .sp
668 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
669 .ti +5n
670 .B const char **\fIerrptr\fP);
671 .PP
672 If a compiled pattern is going to be used several times, it is worth spending
673 more time analyzing it in order to speed up the time taken for matching. The
674 function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
675 argument. If studying the pattern produces additional information that will
676 help speed up matching, \fBpcre_study()\fP returns a pointer to a
677 \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
678 results of the study.
679 .P
680 The returned value from \fBpcre_study()\fP can be passed directly to
681 \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
682 fields that can be set by the caller before the block is passed; these are
683 described
684 .\" HTML <a href="#extradata">
685 .\" </a>
686 below
687 .\"
688 in the section on matching a pattern.
689 .P
690 If studying the pattern does not produce any additional information
691 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
692 wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
693 own \fBpcre_extra\fP block.
694 .P
695 The second argument of \fBpcre_study()\fP contains option bits. At present, no
696 options are defined, and this argument should always be zero.
697 .P
698 The third argument for \fBpcre_study()\fP is a pointer for an error message. If
699 studying succeeds (even if no data is returned), the variable it points to is
700 set to NULL. Otherwise it is set to point to a textual error message. This is a
701 static string that is part of the library. You must not try to free it. You
702 should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
703 sure that it has run successfully.
704 .P
705 This is a typical call to \fBpcre_study\fP():
706 .sp
707 pcre_extra *pe;
708 pe = pcre_study(
709 re, /* result of pcre_compile() */
710 0, /* no options exist */
711 &error); /* set to NULL or points to a message */
712 .sp
713 At present, studying a pattern is useful only for non-anchored patterns that do
714 not have a single fixed starting character. A bitmap of possible starting
715 bytes is created.
716 .
717 .
718 .\" HTML <a name="localesupport"></a>
720 .rs
721 .sp
722 PCRE handles caseless matching, and determines whether characters are letters
723 digits, or whatever, by reference to a set of tables, indexed by character
724 value. When running in UTF-8 mode, this applies only to characters with codes
725 less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
726 can be tested with \ep if PCRE is built with Unicode character property
727 support. The use of locales with Unicode is discouraged.
728 .P
729 An internal set of tables is created in the default C locale when PCRE is
730 built. This is used when the final argument of \fBpcre_compile()\fP is NULL,
731 and is sufficient for many applications. An alternative set of tables can,
732 however, be supplied. These may be created in a different locale from the
733 default. As more and more applications change to using Unicode, the need for
734 this locale support is expected to die away.
735 .P
736 External tables are built by calling the \fBpcre_maketables()\fP function,
737 which has no arguments, in the relevant locale. The result can then be passed
738 to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
739 example, to build and use tables that are appropriate for the French locale
740 (where accented characters with values greater than 128 are treated as letters),
741 the following code could be used:
742 .sp
743 setlocale(LC_CTYPE, "fr_FR");
744 tables = pcre_maketables();
745 re = pcre_compile(..., tables);
746 .sp
747 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
748 obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
749 that the memory containing the tables remains available for as long as it is
750 needed.
751 .P
752 The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
753 pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
754 and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
755 pattern, compilation, studying and matching all happen in the same locale, but
756 different patterns can be compiled in different locales.
757 .P
758 It is possible to pass a table pointer or NULL (indicating the use of the
759 internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
760 this facility could be used to match a pattern in a different locale from the
761 one in which it was compiled. Passing table pointers at run time is discussed
762 below in the section on matching a pattern.
763 .
764 .
766 .rs
767 .sp
768 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
769 .ti +5n
770 .B int \fIwhat\fP, void *\fIwhere\fP);
771 .PP
772 The \fBpcre_fullinfo()\fP function returns information about a compiled
773 pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
774 nevertheless retained for backwards compability (and is documented below).
775 .P
776 The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
777 pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
778 the pattern was not studied. The third argument specifies which piece of
779 information is required, and the fourth argument is a pointer to a variable
780 to receive the data. The yield of the function is zero for success, or one of
781 the following negative numbers:
782 .sp
783 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
784 the argument \fIwhere\fP was NULL
785 PCRE_ERROR_BADMAGIC the "magic number" was not found
786 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
787 .sp
788 The "magic number" is placed at the start of each compiled pattern as an simple
789 check against passing an arbitrary memory pointer. Here is a typical call of
790 \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
791 .sp
792 int rc;
793 size_t length;
794 rc = pcre_fullinfo(
795 re, /* result of pcre_compile() */
796 pe, /* result of pcre_study(), or NULL */
797 PCRE_INFO_SIZE, /* what is required */
798 &length); /* where to put the data */
799 .sp
800 The possible values for the third argument are defined in \fBpcre.h\fP, and are
801 as follows:
802 .sp
804 .sp
805 Return the number of the highest back reference in the pattern. The fourth
806 argument should point to an \fBint\fP variable. Zero is returned if there are
807 no back references.
808 .sp
810 .sp
811 Return the number of capturing subpatterns in the pattern. The fourth argument
812 should point to an \fBint\fP variable.
813 .sp
815 .sp
816 Return a pointer to the internal default character tables within PCRE. The
817 fourth argument should point to an \fBunsigned char *\fP variable. This
818 information call is provided for internal use by the \fBpcre_study()\fP
819 function. External callers can cause PCRE to use its internal tables by passing
820 a NULL table pointer.
821 .sp
823 .sp
824 Return information about the first byte of any matched string, for a
825 non-anchored pattern. The fourth argument should point to an \fBint\fP
826 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
827 still recognized for backwards compatibility.)
828 .P
829 If there is a fixed first byte, for example, from a pattern such as
830 (cat|cow|coyote). Otherwise, if either
831 .sp
832 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
833 starts with "^", or
834 .sp
835 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
836 (if it were set, the pattern would be anchored),
837 .sp
838 -1 is returned, indicating that the pattern matches only at the start of a
839 subject string or after any newline within the string. Otherwise -2 is
840 returned. For anchored patterns, -2 is returned.
841 .sp
843 .sp
844 If the pattern was studied, and this resulted in the construction of a 256-bit
845 table indicating a fixed set of bytes for the first byte in any matching
846 string, a pointer to the table is returned. Otherwise NULL is returned. The
847 fourth argument should point to an \fBunsigned char *\fP variable.
848 .sp
850 .sp
851 Return the value of the rightmost literal byte that must exist in any matched
852 string, other than at its start, if such a byte has been recorded. The fourth
853 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
854 returned. For anchored patterns, a last literal byte is recorded only if it
855 follows something of variable length. For example, for the pattern
856 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
857 is -1.
858 .sp
862 .sp
863 PCRE supports the use of named as well as numbered capturing parentheses. The
864 names are just an additional way of identifying the parentheses, which still
865 acquire numbers. Several convenience functions such as
866 \fBpcre_get_named_substring()\fP are provided for extracting captured
867 substrings by name. It is also possible to extract the data directly, by first
868 converting the name to a number in order to access the correct pointers in the
869 output vector (described with \fBpcre_exec()\fP below). To do the conversion,
870 you need to use the name-to-number map, which is described by these three
871 values.
872 .P
873 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
874 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
875 entry; both of these return an \fBint\fP value. The entry size depends on the
876 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
877 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
878 are the number of the capturing parenthesis, most significant byte first. The
879 rest of the entry is the corresponding name, zero terminated. The names are in
880 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
881 their parentheses numbers. For example, consider the following pattern (assume
882 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
883 .sp
884 .\" JOIN
885 (?P<date> (?P<year>(\ed\ed)?\ed\ed) -
886 (?P<month>\ed\ed) - (?P<day>\ed\ed) )
887 .sp
888 There are four named subpatterns, so the table has four entries, and each entry
889 in the table is eight bytes long. The table is as follows, with non-printing
890 bytes shows in hexadecimal, and undefined bytes shown as ??:
891 .sp
892 00 01 d a t e 00 ??
893 00 05 d a y 00 ?? ??
894 00 04 m o n t h 00
895 00 02 y e a r 00 ??
896 .sp
897 When writing code to extract data from named subpatterns using the
898 name-to-number map, remember that the length of the entries is likely to be
899 different for each compiled pattern.
900 .sp
902 .sp
903 Return a copy of the options with which the pattern was compiled. The fourth
904 argument should point to an \fBunsigned long int\fP variable. These option bits
905 are those specified in the call to \fBpcre_compile()\fP, modified by any
906 top-level option settings within the pattern itself.
907 .P
908 A pattern is automatically anchored by PCRE if all of its top-level
909 alternatives begin with one of the following:
910 .sp
911 ^ unless PCRE_MULTILINE is set
912 \eA always
913 \eG always
914 .\" JOIN
915 .* if PCRE_DOTALL is set and there are no back
916 references to the subpattern in which .* appears
917 .sp
918 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
919 \fBpcre_fullinfo()\fP.
920 .sp
922 .sp
923 Return the size of the compiled pattern, that is, the value that was passed as
924 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
925 place the compiled data. The fourth argument should point to a \fBsize_t\fP
926 variable.
927 .sp
929 .sp
930 Return the size of the data block pointed to by the \fIstudy_data\fP field in
931 a \fBpcre_extra\fP block. That is, it is the value that was passed to
932 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
933 created by \fBpcre_study()\fP. The fourth argument should point to a
934 \fBsize_t\fP variable.
935 .
936 .
938 .rs
939 .sp
940 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
941 .B *\fIfirstcharptr\fP);
942 .PP
943 The \fBpcre_info()\fP function is now obsolete because its interface is too
944 restrictive to return all the available data about a compiled pattern. New
945 programs should use \fBpcre_fullinfo()\fP instead. The yield of
946 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
947 following negative numbers:
948 .sp
949 PCRE_ERROR_NULL the argument \fIcode\fP was NULL
950 PCRE_ERROR_BADMAGIC the "magic number" was not found
951 .sp
952 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
953 pattern was compiled is placed in the integer it points to (see
955 .P
956 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
957 it is used to pass back information about the first character of any matched
958 string (see PCRE_INFO_FIRSTBYTE above).
959 .
960 .
962 .rs
963 .sp
964 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
965 .PP
966 The \fBpcre_refcount()\fP function is used to maintain a reference count in the
967 data block that contains a compiled pattern. It is provided for the benefit of
968 applications that operate in an object-oriented manner, where different parts
969 of the application may be using the same compiled pattern, but you want to free
970 the block when they are all done.
971 .P
972 When a pattern is compiled, the reference count field is initialized to zero.
973 It is changed only by calling this function, whose action is to add the
974 \fIadjust\fP value (which may be positive or negative) to it. The yield of the
975 function is the new value. However, the value of the count is constrained to
976 lie between 0 and 65535, inclusive. If the new value is outside these limits,
977 it is forced to the appropriate limit value.
978 .P
979 Except when it is zero, the reference count is not correctly preserved if a
980 pattern is compiled on one host and then transferred to a host whose byte-order
981 is different. (This seems a highly unlikely scenario.)
982 .
983 .
985 .rs
986 .sp
987 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
988 .ti +5n
989 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
990 .ti +5n
991 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
992 .P
993 The function \fBpcre_exec()\fP is called to match a subject string against a
994 compiled pattern, which is passed in the \fIcode\fP argument. If the
995 pattern has been studied, the result of the study should be passed in the
996 \fIextra\fP argument. This function is the main matching facility of the
997 library, and it operates in a Perl-like manner. For specialist use there is
998 also an alternative matching function, which is described
999 .\" HTML <a href="#dfamatch">
1000 .\" </a>
1001 below
1002 .\"
1003 in the section about the \fBpcre_dfa_exec()\fP function.
1004 .P
1005 In most applications, the pattern will have been compiled (and optionally
1006 studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1007 possible to save compiled patterns and study data, and then use them later
1008 in different processes, possibly even on different hosts. For a discussion
1009 about this, see the
1010 .\" HREF
1011 \fBpcreprecompile\fP
1012 .\"
1013 documentation.
1014 .P
1015 Here is an example of a simple call to \fBpcre_exec()\fP:
1016 .sp
1017 int rc;
1018 int ovector[30];
1019 rc = pcre_exec(
1020 re, /* result of pcre_compile() */
1021 NULL, /* we didn't study the pattern */
1022 "some string", /* the subject string */
1023 11, /* the length of the subject string */
1024 0, /* start at offset 0 in the subject */
1025 0, /* default options */
1026 ovector, /* vector of integers for substring information */
1027 30); /* number of elements (NOT size in bytes) */
1028 .
1029 .\" HTML <a name="extradata"></a>
1030 .SS "Extra data for \fBpcre_exec()\fR"
1031 .rs
1032 .sp
1033 If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1034 data block. The \fBpcre_study()\fP function returns such a block (when it
1035 doesn't return NULL), but you can also create one for yourself, and pass
1036 additional information in it. The \fBpcre_extra\fP block contains the following
1037 fields (not necessarily in this order):
1038 .sp
1039 unsigned long int \fIflags\fP;
1040 void *\fIstudy_data\fP;
1041 unsigned long int \fImatch_limit\fP;
1042 unsigned long int \fImatch_limit_recursion\fP;
1043 void *\fIcallout_data\fP;
1044 const unsigned char *\fItables\fP;
1045 .sp
1046 The \fIflags\fP field is a bitmap that specifies which of the other fields
1047 are set. The flag bits are:
1048 .sp
1054 .sp
1055 Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1056 \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1057 the appropriate flag bit. You should not set this yourself, but you may add to
1058 the block by setting the other fields and their corresponding flag bits.
1059 .P
1060 The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1061 vast amount of resources when running patterns that are not going to match,
1062 but which have a very large number of possibilities in their search trees. The
1063 classic example is the use of nested unlimited repeats.
1064 .P
1065 Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1066 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1067 number of times this function is called during a match, which has the effect of
1068 limiting the amount of backtracking that can take place. For patterns that are
1069 not anchored, the count restarts from zero for each position in the subject
1070 string.
1071 .P
1072 The default value for the limit can be set when PCRE is built; the default
1073 default is 10 million, which handles all but the most extreme cases. You can
1074 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1075 block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1076 the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1078 .P
1079 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1080 instead of limiting the total number of times that \fBmatch()\fP is called, it
1081 limits the depth of recursion. The recursion depth is a smaller number than the
1082 total number of calls, because not all calls to \fBmatch()\fP are recursive.
1083 This limit is of use only if it is set smaller than \fImatch_limit\fP.
1084 .P
1085 Limiting the recursion depth limits the amount of stack that can be used, or,
1086 when PCRE has been compiled to use memory on the heap instead of the stack, the
1087 amount of heap memory that can be used.
1088 .P
1089 The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1090 built; the default default is the same value as the default for
1091 \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1092 with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1093 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1094 is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1095 .P
1096 The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1097 which is described in the
1098 .\" HREF
1099 \fBpcrecallout\fP
1100 .\"
1101 documentation.
1102 .P
1103 The \fItables\fP field is used to pass a character tables pointer to
1104 \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1105 pattern. A non-NULL value is stored with the compiled pattern only if custom
1106 tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1107 If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1108 internal tables to be used. This facility is helpful when re-using patterns
1109 that have been saved after compiling with an external set of tables, because
1110 the external tables might be at a different address when \fBpcre_exec()\fP is
1111 called. See the
1112 .\" HREF
1113 \fBpcreprecompile\fP
1114 .\"
1115 documentation for a discussion of saving compiled patterns for later use.
1116 .
1117 .SS "Option bits for \fBpcre_exec()\fP"
1118 .rs
1119 .sp
1120 The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1121 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1123 .sp
1125 .sp
1126 The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1127 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1128 to be anchored by virtue of its contents, it cannot be made unachored at
1129 matching time.
1130 .sp
1134 .sp
1135 These options override the newline definition that was chosen or defaulted when
1136 the pattern was compiled. For details, see the description \fBpcre_compile()\fP
1137 above. During matching, the newline choice affects the behaviour of the dot,
1138 circumflex, and dollar metacharacters.
1139 .sp
1141 .sp
1142 This option specifies that first character of the subject string is not the
1143 beginning of a line, so the circumflex metacharacter should not match before
1144 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1145 never to match. This option affects only the behaviour of the circumflex
1146 metacharacter. It does not affect \eA.
1147 .sp
1149 .sp
1150 This option specifies that the end of the subject string is not the end of a
1151 line, so the dollar metacharacter should not match it nor (except in multiline
1152 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1153 compile time) causes dollar never to match. This option affects only the
1154 behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1155 .sp
1157 .sp
1158 An empty string is not considered to be a valid match if this option is set. If
1159 there are alternatives in the pattern, they are tried. If all the alternatives
1160 match the empty string, the entire match fails. For example, if the pattern
1161 .sp
1162 a?b?
1163 .sp
1164 is applied to a string not beginning with "a" or "b", it matches the empty
1165 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1166 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1167 .P
1168 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1169 of a pattern match of the empty string within its \fBsplit()\fP function, and
1170 when using the /g modifier. It is possible to emulate Perl's behaviour after
1171 matching a null string by first trying the match again at the same offset with
1172 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1173 starting offset (see below) and trying an ordinary match again. There is some
1174 code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1175 .sp
1177 .sp
1178 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1179 string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1180 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1181 start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1182 \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP
1183 contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1184 .P
1185 If you already know that your subject is valid, and you want to skip these
1186 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1187 calling \fBpcre_exec()\fP. You might want to do this for the second and
1188 subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1189 all the matches in a single subject string. However, you should be sure that
1190 the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1191 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1192 subject, or a value of \fIstartoffset\fP that does not point to the start of a
1193 UTF-8 character, is undefined. Your program may crash.
1194 .sp
1196 .sp
1197 This option turns on the partial matching feature. If the subject string fails
1198 to match the pattern, but at some point during the matching process the end of
1199 the subject was reached (that is, the subject partially matches the pattern and
1200 the failure to match occurred only because there were not enough subject
1201 characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1202 PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1203 may appear in the pattern. These are discussed in the
1204 .\" HREF
1205 \fBpcrepartial\fP
1206 .\"
1207 documentation.
1208 .
1209 .SS "The string to be matched by \fBpcre_exec()\fP"
1210 .rs
1211 .sp
1212 The subject string is passed to \fBpcre_exec()\fP as a pointer in
1213 \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1214 \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1215 UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1216 bytes. When the starting offset is zero, the search for a match starts at the
1217 beginning of the subject, and this is by far the most common case.
1218 .P
1219 A non-zero starting offset is useful when searching for another match in the
1220 same subject by calling \fBpcre_exec()\fP again after a previous success.
1221 Setting \fIstartoffset\fP differs from just passing over a shortened string and
1222 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1223 lookbehind. For example, consider the pattern
1224 .sp
1225 \eBiss\eB
1226 .sp
1227 which finds occurrences of "iss" in the middle of words. (\eB matches only if
1228 the current position in the subject is not a word boundary.) When applied to
1229 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1230 occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1231 subject, namely "issipi", it does not match, because \eB is always false at the
1232 start of the subject, which is deemed to be a word boundary. However, if
1233 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1234 set to 4, it finds the second occurrence of "iss" because it is able to look
1235 behind the starting point to discover that it is preceded by a letter.
1236 .P
1237 If a non-zero starting offset is passed when the pattern is anchored, one
1238 attempt to match at the given offset is made. This can only succeed if the
1239 pattern does not require the match to be at the start of the subject.
1240 .
1241 .SS "How \fBpcre_exec()\fP returns captured substrings"
1242 .rs
1243 .sp
1244 In general, a pattern matches a certain portion of the subject, and in
1245 addition, further substrings from the subject may be picked out by parts of the
1246 pattern. Following the usage in Jeffrey Friedl's book, this is called
1247 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1248 a fragment of a pattern that picks out a substring. PCRE supports several other
1249 kinds of parenthesized subpattern that do not cause substrings to be captured.
1250 .P
1251 Captured substrings are returned to the caller via a vector of integer offsets
1252 whose address is passed in \fIovector\fP. The number of elements in the vector
1253 is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:
1254 this argument is NOT the size of \fIovector\fP in bytes.
1255 .P
1256 The first two-thirds of the vector is used to pass back captured substrings,
1257 each substring using a pair of integers. The remaining third of the vector is
1258 used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1259 and is not available for passing back information. The length passed in
1260 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1261 rounded down.
1262 .P
1263 When a match is successful, information about captured substrings is returned
1264 in pairs of integers, starting at the beginning of \fIovector\fP, and
1265 continuing up to two-thirds of its length at the most. The first element of a
1266 pair is set to the offset of the first character in a substring, and the second
1267 is set to the offset of the first character after the end of a substring. The
1268 first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1269 subject string matched by the entire pattern. The next pair is used for the
1270 first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1271 is one more than the highest numbered pair that has been set. For example, if
1272 two substrings have been captured, the returned value is 3. If there are no
1273 capturing subpatterns, the return value from a successful match is 1,
1274 indicating that just the first pair of offsets has been set.
1275 .P
1276 If a capturing subpattern is matched repeatedly, it is the last portion of the
1277 string that it matched that is returned.
1278 .P
1279 If the vector is too small to hold all the captured substring offsets, it is
1280 used as far as possible (up to two-thirds of its length), and the function
1281 returns a value of zero. In particular, if the substring offsets are not of
1282 interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1283 \fIovecsize\fP as zero. However, if the pattern contains back references and
1284 the \fIovector\fP is not big enough to remember the related substrings, PCRE
1285 has to get additional memory for use during matching. Thus it is usually
1286 advisable to supply an \fIovector\fP.
1287 .P
1288 The \fBpcre_info()\fP function can be used to find out how many capturing
1289 subpatterns there are in a compiled pattern. The smallest size for
1290 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1291 the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1292 .P
1293 It is possible for capturing subpattern number \fIn+1\fP to match some part of
1294 the subject when subpattern \fIn\fP has not been used at all. For example, if
1295 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1296 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1297 happens, both values in the offset pairs corresponding to unused subpatterns
1298 are set to -1.
1299 .P
1300 Offset values that correspond to unused subpatterns at the end of the
1301 expression are also set to -1. For example, if the string "abc" is matched
1302 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1303 return from the function is 2, because the highest used capturing subpattern
1304 number is 1. However, you can refer to the offsets for the second and third
1305 capturing subpatterns if you wish (assuming the vector is large enough, of
1306 course).
1307 .P
1308 Some convenience functions are provided for extracting the captured substrings
1309 as separate strings. These are described below.
1310 .
1311 .\" HTML <a name="errorlist"></a>
1312 .SS "Error return values from \fBpcre_exec()\fP"
1313 .rs
1314 .sp
1315 If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1316 defined in the header file:
1317 .sp
1319 .sp
1320 The subject string did not match the pattern.
1321 .sp
1323 .sp
1324 Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1325 NULL and \fIovecsize\fP was not zero.
1326 .sp
1328 .sp
1329 An unrecognized bit was set in the \fIoptions\fP argument.
1330 .sp
1332 .sp
1333 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1334 the case when it is passed a junk pointer and to detect when a pattern that was
1335 compiled in an environment of one endianness is run in an environment with the
1336 other endianness. This is the error that PCRE gives when the magic number is
1337 not present.
1338 .sp
1340 .sp
1341 While running the pattern match, an unknown item was encountered in the
1342 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1343 of the compiled pattern.
1344 .sp
1346 .sp
1347 If a pattern contains back references, but the \fIovector\fP that is passed to
1348 \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1349 gets a block of memory at the start of matching to use for this purpose. If the
1350 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1351 automatically freed at the end of matching.
1352 .sp
1354 .sp
1355 This error is used by the \fBpcre_copy_substring()\fP,
1356 \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1357 below). It is never returned by \fBpcre_exec()\fP.
1358 .sp
1360 .sp
1361 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1362 \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1363 above.
1364 .sp
1366 .sp
1367 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1368 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1369 description above.
1370 .sp
1372 .sp
1373 This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1374 use by callout functions that want to yield a distinctive error code. See the
1375 .\" HREF
1376 \fBpcrecallout\fP
1377 .\"
1378 documentation for details.
1379 .sp
1381 .sp
1382 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1383 .sp
1385 .sp
1386 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1387 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1388 .sp
1390 .sp
1391 The subject string did not match, but it did match partially. See the
1392 .\" HREF
1393 \fBpcrepartial\fP
1394 .\"
1395 documentation for details of partial matching.
1396 .sp
1398 .sp
1399 The PCRE_PARTIAL option was used with a compiled pattern containing items that
1400 are not supported for partial matching. See the
1401 .\" HREF
1402 \fBpcrepartial\fP
1403 .\"
1404 documentation for details of partial matching.
1405 .sp
1407 .sp
1408 An unexpected internal error has occurred. This error could be caused by a bug
1409 in PCRE or by overwriting of the compiled pattern.
1410 .sp
1412 .sp
1413 This error is given if the value of the \fIovecsize\fP argument is negative.
1414 .
1415 .
1417 .rs
1418 .sp
1419 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1420 .ti +5n
1421 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1422 .ti +5n
1423 .B int \fIbuffersize\fP);
1424 .PP
1425 .br
1426 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1427 .ti +5n
1428 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1429 .ti +5n
1430 .B const char **\fIstringptr\fP);
1431 .PP
1432 .br
1433 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1434 .ti +5n
1435 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1436 .PP
1437 Captured substrings can be accessed directly by using the offsets returned by
1438 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1439 \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1440 \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1441 as new, separate, zero-terminated strings. These functions identify substrings
1442 by number. The next section describes functions for extracting named
1443 substrings.
1444 .P
1445 A substring that contains a binary zero is correctly extracted and has a
1446 further zero added on the end, but the result is not, of course, a C string.
1447 However, you can process such a string by referring to the length that is
1448 returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1449 Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1450 for handling strings containing binary zeros, because the end of the final
1451 string is not independently indicated.
1452 .P
1453 The first three arguments are the same for all three of these functions:
1454 \fIsubject\fP is the subject string that has just been successfully matched,
1455 \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1456 \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1457 captured by the match, including the substring that matched the entire regular
1458 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1459 than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1460 space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1461 number of elements in the vector divided by three.
1462 .P
1463 The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1464 extract a single substring, whose number is given as \fIstringnumber\fP. A
1465 value of zero extracts the substring that matched the entire pattern, whereas
1466 higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1467 the string is placed in \fIbuffer\fP, whose length is given by
1468 \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1469 obtained via \fBpcre_malloc\fP, and its address is returned via
1470 \fIstringptr\fP. The yield of the function is the length of the string, not
1471 including the terminating zero, or one of
1472 .sp
1474 .sp
1475 The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1476 memory failed for \fBpcre_get_substring()\fP.
1477 .sp
1479 .sp
1480 There is no substring whose number is \fIstringnumber\fP.
1481 .P
1482 The \fBpcre_get_substring_list()\fP function extracts all available substrings
1483 and builds a list of pointers to them. All this is done in a single block of
1484 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1485 is returned via \fIlistptr\fP, which is also the start of the list of string
1486 pointers. The end of the list is marked by a NULL pointer. The yield of the
1487 function is zero if all went well, or
1488 .sp
1490 .sp
1491 if the attempt to get the memory block failed.
1492 .P
1493 When any of these functions encounter a substring that is unset, which can
1494 happen when capturing subpattern number \fIn+1\fP matches some part of the
1495 subject, but subpattern \fIn\fP has not been used at all, they return an empty
1496 string. This can be distinguished from a genuine zero-length substring by
1497 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1498 substrings.
1499 .P
1500 The two convenience functions \fBpcre_free_substring()\fP and
1501 \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1502 a previous call of \fBpcre_get_substring()\fP or
1503 \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1504 the function pointed to by \fBpcre_free\fP, which of course could be called
1505 directly from a C program. However, PCRE is used in some situations where it is
1506 linked via a special interface to another programming language that cannot use
1507 \fBpcre_free\fP directly; it is for these cases that the functions are
1508 provided.
1509 .
1510 .
1512 .rs
1513 .sp
1514 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1515 .ti +5n
1516 .B const char *\fIname\fP);
1517 .PP
1518 .br
1519 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1520 .ti +5n
1521 .B const char *\fIsubject\fP, int *\fIovector\fP,
1522 .ti +5n
1523 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1524 .ti +5n
1525 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1526 .PP
1527 .br
1528 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1529 .ti +5n
1530 .B const char *\fIsubject\fP, int *\fIovector\fP,
1531 .ti +5n
1532 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1533 .ti +5n
1534 .B const char **\fIstringptr\fP);
1535 .PP
1536 To extract a substring by name, you first have to find associated number.
1537 For example, for this pattern
1538 .sp
1539 (a+)b(?P<xxx>\ed+)...
1540 .sp
1541 the number of the subpattern called "xxx" is 2. If the name is known to be
1542 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1543 calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1544 pattern, and the second is the name. The yield of the function is the
1545 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1546 that name.
1547 .P
1548 Given the number, you can extract the substring directly, or use one of the
1549 functions described in the previous section. For convenience, there are also
1550 two functions that do the whole job.
1551 .P
1552 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1553 \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1554 functions that extract by number. As these are described in the previous
1555 section, they are not re-described here. There are just two differences:
1556 .P
1557 First, instead of a substring number, a substring name is given. Second, there
1558 is an extra argument, given at the start, which is a pointer to the compiled
1559 pattern. This is needed in order to gain access to the name-to-number
1560 translation table.
1561 .P
1562 These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1563 then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as
1564 appropriate.
1565 .
1566 .
1568 .rs
1569 .sp
1570 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1571 .ti +5n
1572 .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1573 .PP
1574 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1575 are not required to be unique. Normally, patterns with duplicate names are such
1576 that in any one match, only one of the named subpatterns participates. An
1577 example is shown in the
1578 .\" HREF
1579 \fBpcrepattern\fP
1580 .\"
1581 documentation. When duplicates are present, \fBpcre_copy_named_substring()\fP
1582 and \fBpcre_get_named_substring()\fP return the first substring corresponding
1583 to the given name that is set. If none are set, an empty string is returned.
1584 The \fBpcre_get_stringnumber()\fP function returns one of the numbers that are
1585 associated with the name, but it is not defined which it is.
1586 .sp
1587 If you want to get full details of all captured substrings for a given name,
1588 you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1589 argument is the compiled pattern, and the second is the name. The third and
1590 fourth are pointers to variables which are updated by the function. After it
1591 has run, they point to the first and last entries in the name-to-number table
1592 for the given name. The function itself returns the length of each entry, or
1593 PCRE_ERROR_NOSUBSTRING if there are none. The format of the table is described
1594 above in the section entitled \fIInformation about a pattern\fP. Given all the
1595 relevant entries for the name, you can extract each of their numbers, and hence
1596 the captured data, if any.
1597 .
1598 .
1600 .rs
1601 .sp
1602 The traditional matching function uses a similar algorithm to Perl, which stops
1603 when it finds the first match, starting at a given point in the subject. If you
1604 want to find all possible matches, or the longest possible match, consider
1605 using the alternative matching function (see below) instead. If you cannot use
1606 the alternative function, but still need to find all possible matches, you
1607 can kludge it up by making use of the callout facility, which is described in
1608 the
1609 .\" HREF
1610 \fBpcrecallout\fP
1611 .\"
1612 documentation.
1613 .P
1614 What you have to do is to insert a callout right at the end of the pattern.
1615 When your callout function is called, extract and save the current matched
1616 substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1617 other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1618 will yield PCRE_ERROR_NOMATCH.
1619 .
1620 .
1621 .\" HTML <a name="dfamatch"></a>
1623 .rs
1624 .sp
1625 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1626 .ti +5n
1627 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1628 .ti +5n
1629 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1630 .ti +5n
1631 .B int *\fIworkspace\fP, int \fIwscount\fP);
1632 .P
1633 The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1634 a compiled pattern, using a "DFA" matching algorithm. This has different
1635 characteristics to the normal algorithm, and is not compatible with Perl. Some
1636 of the features of PCRE patterns are not supported. Nevertheless, there are
1637 times when this kind of matching can be useful. For a discussion of the two
1638 matching algorithms, see the
1639 .\" HREF
1640 \fBpcrematching\fP
1641 .\"
1642 documentation.
1643 .P
1644 The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1645 \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1646 different way, and this is described below. The other common arguments are used
1647 in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1648 here.
1649 .P
1650 The two additional arguments provide workspace for the function. The workspace
1651 vector should contain at least 20 elements. It is used for keeping track of
1652 multiple paths through the pattern tree. More workspace will be needed for
1653 patterns and subjects where there are a lot of potential matches.
1654 .P
1655 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1656 .sp
1657 int rc;
1658 int ovector[10];
1659 int wspace[20];
1660 rc = pcre_dfa_exec(
1661 re, /* result of pcre_compile() */
1662 NULL, /* we didn't study the pattern */
1663 "some string", /* the subject string */
1664 11, /* the length of the subject string */
1665 0, /* start at offset 0 in the subject */
1666 0, /* default options */
1667 ovector, /* vector of integers for substring information */
1668 10, /* number of elements (NOT size in bytes) */
1669 wspace, /* working space vector */
1670 20); /* number of elements (NOT size in bytes) */
1671 .
1672 .SS "Option bits for \fBpcre_dfa_exec()\fP"
1673 .rs
1674 .sp
1675 The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1676 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1678 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1679 the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1680 .sp
1682 .sp
1683 This has the same general effect as it does for \fBpcre_exec()\fP, but the
1684 details are slightly different. When PCRE_PARTIAL is set for
1685 \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1686 PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1687 complete matches, but there is still at least one matching possibility. The
1688 portion of the string that provided the partial match is set as the first
1689 matching string.
1690 .sp
1692 .sp
1693 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1694 soon as it has found one match. Because of the way the DFA algorithm works,
1695 this is necessarily the shortest possible match at the first possible matching
1696 point in the subject string.
1697 .sp
1699 .sp
1700 When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1701 a partial match, it is possible to call it again, with additional subject
1702 characters, and have it continue with the same match. The PCRE_DFA_RESTART
1703 option requests this action; when it is set, the \fIworkspace\fP and
1704 \fIwscount\fP options must reference the same vector as before because data
1705 about the match so far is left in them after a partial match. There is more
1706 discussion of this facility in the
1707 .\" HREF
1708 \fBpcrepartial\fP
1709 .\"
1710 documentation.
1711 .
1712 .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1713 .rs
1714 .sp
1715 When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1716 substring in the subject. Note, however, that all the matches from one run of
1717 the function start at the same point in the subject. The shorter matches are
1718 all initial substrings of the longer matches. For example, if the pattern
1719 .sp
1720 <.*>
1721 .sp
1722 is matched against the string
1723 .sp
1724 This is <something> <something else> <something further> no more
1725 .sp
1726 the three matched strings are
1727 .sp
1728 <something>
1729 <something> <something else>
1730 <something> <something else> <something further>
1731 .sp
1732 On success, the yield of the function is a number greater than zero, which is
1733 the number of matched substrings. The substrings themselves are returned in
1734 \fIovector\fP. Each string uses two elements; the first is the offset to the
1735 start, and the second is the offset to the end. All the strings have the same
1736 start offset. (Space could have been saved by giving this only once, but it was
1737 decided to retain some compatibility with the way \fBpcre_exec()\fP returns
1738 data, even though the meaning of the strings is different.)
1739 .P
1740 The strings are returned in reverse order of length; that is, the longest
1741 matching string is given first. If there were too many matches to fit into
1742 \fIovector\fP, the yield of the function is zero, and the vector is filled with
1743 the longest matches.
1744 .
1745 .SS "Error returns from \fBpcre_dfa_exec()\fP"
1746 .rs
1747 .sp
1748 The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1749 Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1750 described
1751 .\" HTML <a href="#errorlist">
1752 .\" </a>
1753 above.
1754 .\"
1755 There are in addition the following errors that are specific to
1756 \fBpcre_dfa_exec()\fP:
1757 .sp
1759 .sp
1760 This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1761 that it does not support, for instance, the use of \eC or a back reference.
1762 .sp
1764 .sp
1765 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item in a
1766 pattern that uses a back reference for the condition. This is not supported.
1767 .sp
1769 .sp
1770 This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1771 block that contains a setting of the \fImatch_limit\fP field. This is not
1772 supported (it is meaningless).
1773 .sp
1775 .sp
1776 This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1777 \fIworkspace\fP vector.
1778 .sp
1780 .sp
1781 When a recursive subpattern is processed, the matching function calls itself
1782 recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1783 error is given if the output vector is not large enough. This should be
1784 extremely rare, as a vector of size 1000 is used.
1785 .P
1786 .in 0
1787 Last updated: 08 June 2006
1788 .br
1789 Copyright (c) 1997-2006 University of Cambridge.

  ViewVC Help
Powered by ViewVC 1.1.5