ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 91 - (show annotations)
Sat Feb 24 21:41:34 2007 UTC (14 years, 1 month ago) by nigel
File MIME type: text/html
File size: 79800 byte(s)
Error occurred while calculating annotation data.
Load pcre-6.7 into code/trunk.
1 <html>
2 <head>
3 <title>pcreapi specification</title>
4 </head>
5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 <h1>pcreapi man page</h1>
7 <p>
8 Return to the <a href="index.html">PCRE index page</a>.
9 </p>
10 <p>
11 This page is part of the PCRE HTML documentation. It was generated automatically
12 from the original man page. If there is any nonsense in it, please consult the
13 man page, in case the conversion went wrong.
14 <br>
15 <ul>
16 <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17 <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18 <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19 <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
21 <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22 <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23 <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24 <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25 <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26 <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27 <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28 <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
32 <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33 <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
35 </ul>
36 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
37 <P>
38 <b>#include &#60;pcre.h&#62;</b>
39 </P>
40 <P>
41 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
42 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
43 <b>const unsigned char *<i>tableptr</i>);</b>
44 </P>
45 <P>
46 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
47 <b>int *<i>errorcodeptr</i>,</b>
48 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
49 <b>const unsigned char *<i>tableptr</i>);</b>
50 </P>
51 <P>
52 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
53 <b>const char **<i>errptr</i>);</b>
54 </P>
55 <P>
56 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
57 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
58 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
59 </P>
60 <P>
61 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
62 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
63 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
64 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
65 </P>
66 <P>
67 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
68 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
69 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
70 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
71 </P>
72 <P>
73 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
74 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
75 <b>int <i>buffersize</i>);</b>
76 </P>
77 <P>
78 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
79 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
80 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
81 <b>const char **<i>stringptr</i>);</b>
82 </P>
83 <P>
84 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
85 <b>const char *<i>name</i>);</b>
86 </P>
87 <P>
88 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
89 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
90 </P>
91 <P>
92 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
93 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
94 <b>const char **<i>stringptr</i>);</b>
95 </P>
96 <P>
97 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
98 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
99 </P>
100 <P>
101 <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
102 </P>
103 <P>
104 <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
105 </P>
106 <P>
107 <b>const unsigned char *pcre_maketables(void);</b>
108 </P>
109 <P>
110 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
111 <b>int <i>what</i>, void *<i>where</i>);</b>
112 </P>
113 <P>
114 <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
115 <b>*<i>firstcharptr</i>);</b>
116 </P>
117 <P>
118 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
119 </P>
120 <P>
121 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
122 </P>
123 <P>
124 <b>char *pcre_version(void);</b>
125 </P>
126 <P>
127 <b>void *(*pcre_malloc)(size_t);</b>
128 </P>
129 <P>
130 <b>void (*pcre_free)(void *);</b>
131 </P>
132 <P>
133 <b>void *(*pcre_stack_malloc)(size_t);</b>
134 </P>
135 <P>
136 <b>void (*pcre_stack_free)(void *);</b>
137 </P>
138 <P>
139 <b>int (*pcre_callout)(pcre_callout_block *);</b>
140 </P>
141 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
142 <P>
143 PCRE has its own native API, which is described in this document. There is
144 also a set of wrapper functions that correspond to the POSIX regular expression
145 API. These are described in the
146 <a href="pcreposix.html"><b>pcreposix</b></a>
147 documentation. Both of these APIs define a set of C function calls. A C++
148 wrapper is distributed with PCRE. It is documented in the
149 <a href="pcrecpp.html"><b>pcrecpp</b></a>
150 page.
151 </P>
152 <P>
153 The native API C function prototypes are defined in the header file
154 <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
155 It can normally be accessed by adding <b>-lpcre</b> 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 <P>
161 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
162 and <b>pcre_exec()</b> are used for compiling and matching regular expressions
163 in a Perl-compatible manner. A sample program that demonstrates the simplest
164 way of using them is provided in the file called <i>pcredemo.c</i> in the source
165 distribution. The
166 <a href="pcresample.html"><b>pcresample</b></a>
167 documentation describes how to run it.
168 </P>
169 <P>
170 A second matching function, <b>pcre_dfa_exec()</b>, 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 <a href="pcrematching.html"><b>pcrematching</b></a>
177 documentation.
178 </P>
179 <P>
180 In addition to the main compiling and matching functions, there are convenience
181 functions for extracting captured substrings from a subject string that is
182 matched by <b>pcre_exec()</b>. They are:
183 <pre>
184 <b>pcre_copy_substring()</b>
185 <b>pcre_copy_named_substring()</b>
186 <b>pcre_get_substring()</b>
187 <b>pcre_get_named_substring()</b>
188 <b>pcre_get_substring_list()</b>
189 <b>pcre_get_stringnumber()</b>
190 <b>pcre_get_stringtable_entries()</b>
191 </pre>
192 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
193 provided, to free the memory used for extracted strings.
194 </P>
195 <P>
196 The function <b>pcre_maketables()</b> is used to build a set of character tables
197 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
198 or <b>pcre_dfa_exec()</b>. 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 <P>
203 The function <b>pcre_fullinfo()</b> is used to find out information about a
204 compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
205 some of the available information, but is retained for backwards compatibility.
206 The function <b>pcre_version()</b> returns a pointer to a string containing the
207 version of PCRE and its date of release.
208 </P>
209 <P>
210 The function <b>pcre_refcount()</b> maintains a reference count in a data block
211 containing a compiled pattern. This is provided for the benefit of
212 object-oriented applications.
213 </P>
214 <P>
215 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
216 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
217 respectively. PCRE calls the memory management functions via these variables,
218 so a calling program can replace them if it wishes to intercept the calls. This
219 should be done before calling any PCRE functions.
220 </P>
221 <P>
222 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
223 indirections to memory management functions. These special functions are used
224 only when PCRE is compiled to use the heap for remembering data, instead of
225 recursive function calls, when running the <b>pcre_exec()</b> function. See the
226 <a href="pcrebuild.html"><b>pcrebuild</b></a>
227 documentation for details of how to do this. It is a non-standard way of
228 building PCRE, for use in environments that have limited stacks. Because of the
229 greater use of memory management, it runs more slowly. Separate functions are
230 provided so that special-purpose external code can be used for this case. When
231 used, these functions are always called in a stack-like manner (last obtained,
232 first freed), and always for memory blocks of the same size. There is a
233 discussion about PCRE's stack usage in the
234 <a href="pcrestack.html"><b>pcrestack</b></a>
235 documentation.
236 </P>
237 <P>
238 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
239 by the caller to a "callout" function, which PCRE will then call at specified
240 points during a matching operation. Details are given in the
241 <a href="pcrecallout.html"><b>pcrecallout</b></a>
242 documentation.
243 </P>
244 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
245 <P>
246 PCRE supports three different conventions for indicating line breaks in
247 strings: a single CR character, a single LF character, or the two-character
248 sequence CRLF. All three are used as "standard" by different operating systems.
249 When PCRE is built, a default can be specified. The default default is LF,
250 which is the Unix standard. When PCRE is run, the default can be overridden,
251 either when a pattern is compiled, or when it is matched.
252 <br>
253 <br>
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 </P>
257 <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
258 <P>
259 The PCRE functions can be used in multi-threading applications, with the
260 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
261 <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
262 callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
263 </P>
264 <P>
265 The compiled form of a regular expression is not altered during matching, so
266 the same compiled pattern can safely be used by several threads at once.
267 </P>
268 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
269 <P>
270 The compiled form of a regular expression can be saved and re-used at a later
271 time, possibly by a different program, and even on a host other than the one on
272 which it was compiled. Details are given in the
273 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
274 documentation.
275 </P>
276 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
277 <P>
278 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
279 </P>
280 <P>
281 The function <b>pcre_config()</b> makes it possible for a PCRE client to
282 discover which optional features have been compiled into the PCRE library. The
283 <a href="pcrebuild.html"><b>pcrebuild</b></a>
284 documentation has more details about these optional features.
285 </P>
286 <P>
287 The first argument for <b>pcre_config()</b> is an integer, specifying which
288 information is required; the second argument is a pointer to a variable into
289 which the information is placed. The following information is available:
290 <pre>
292 </pre>
293 The output is an integer that is set to one if UTF-8 support is available;
294 otherwise it is set to zero.
295 <pre>
297 </pre>
298 The output is an integer that is set to one if support for Unicode character
299 properties is available; otherwise it is set to zero.
300 <pre>
302 </pre>
303 The output is an integer whose value specifies the default character sequence
304 that is recognized as meaning "newline". The three values that are supported
305 are: 10 for LF, 13 for CR, and 3338 for CRLF. The default should normally be
306 the standard sequence for your operating system.
307 <pre>
309 </pre>
310 The output is an integer that contains the number of bytes used for internal
311 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
312 allow larger regular expressions to be compiled, at the expense of slower
313 matching. The default value of 2 is sufficient for all but the most massive
314 patterns, since it allows the compiled pattern to be up to 64K in size.
315 <pre>
317 </pre>
318 The output is an integer that contains the threshold above which the POSIX
319 interface uses <b>malloc()</b> for output vectors. Further details are given in
320 the
321 <a href="pcreposix.html"><b>pcreposix</b></a>
322 documentation.
323 <pre>
325 </pre>
326 The output is an integer that gives the default limit for the number of
327 internal matching function calls in a <b>pcre_exec()</b> execution. Further
328 details are given with <b>pcre_exec()</b> below.
329 <pre>
331 </pre>
332 The output is an integer that gives the default limit for the depth of
333 recursion when calling the internal matching function in a <b>pcre_exec()</b>
334 execution. Further details are given with <b>pcre_exec()</b> below.
335 <pre>
337 </pre>
338 The output is an integer that is set to one if internal recursion when running
339 <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
340 to remember their state. This is the usual way that PCRE is compiled. The
341 output is zero if PCRE was compiled to use blocks of data on the heap instead
342 of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
343 <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
344 avoiding the use of the stack.
345 </P>
346 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
347 <P>
348 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
349 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
350 <b>const unsigned char *<i>tableptr</i>);</b>
351 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
352 <b>int *<i>errorcodeptr</i>,</b>
353 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
354 <b>const unsigned char *<i>tableptr</i>);</b>
355 </P>
356 <P>
357 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
358 called to compile a pattern into an internal form. The only difference between
359 the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
360 <i>errorcodeptr</i>, via which a numerical error code can be returned.
361 </P>
362 <P>
363 The pattern is a C string terminated by a binary zero, and is passed in the
364 <i>pattern</i> argument. A pointer to a single block of memory that is obtained
365 via <b>pcre_malloc</b> is returned. This contains the compiled code and related
366 data. The <b>pcre</b> type is defined for the returned block; this is a typedef
367 for a structure whose contents are not externally defined. It is up to the
368 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
369 </P>
370 <P>
371 Although the compiled code of a PCRE regex is relocatable, that is, it does not
372 depend on memory location, the complete <b>pcre</b> data block is not
373 fully relocatable, because it may contain a copy of the <i>tableptr</i>
374 argument, which is an address (see below).
375 </P>
376 <P>
377 The <i>options</i> argument contains independent bits that affect the
378 compilation. It should be zero if no options are required. The available
379 options are described below. Some of them, in particular, those that are
380 compatible with Perl, can also be set and unset from within the pattern (see
381 the detailed description in the
382 <a href="pcrepattern.html"><b>pcrepattern</b></a>
383 documentation). For these options, the contents of the <i>options</i> argument
384 specifies their initial settings at the start of compilation and execution. The
385 PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
386 matching as well as at compile time.
387 </P>
388 <P>
389 If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
390 Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
391 NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
392 error message. This is a static string that is part of the library. You must
393 not try to free it. The offset from the start of the pattern to the character
394 where the error was discovered is placed in the variable pointed to by
395 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
396 </P>
397 <P>
398 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
399 <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
400 returned via this argument in the event of an error. This is in addition to the
401 textual error message. Error codes and messages are listed below.
402 </P>
403 <P>
404 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
405 character tables that are built when PCRE is compiled, using the default C
406 locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
407 call to <b>pcre_maketables()</b>. This value is stored with the compiled
408 pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
409 passed to it. For more discussion, see the section on locale support below.
410 </P>
411 <P>
412 This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
413 <pre>
414 pcre *re;
415 const char *error;
416 int erroffset;
417 re = pcre_compile(
418 "^A.*Z", /* the pattern */
419 0, /* default options */
420 &error, /* for error message */
421 &erroffset, /* for error offset */
422 NULL); /* use default character tables */
423 </pre>
424 The following names for option bits are defined in the <b>pcre.h</b> header
425 file:
426 <pre>
428 </pre>
429 If this bit is set, the pattern is forced to be "anchored", that is, it is
430 constrained to match only at the first matching point in the string that is
431 being searched (the "subject string"). This effect can also be achieved by
432 appropriate constructs in the pattern itself, which is the only way to do it in
433 Perl.
434 <pre>
436 </pre>
437 If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
438 all with number 255, before each pattern item. For discussion of the callout
439 facility, see the
440 <a href="pcrecallout.html"><b>pcrecallout</b></a>
441 documentation.
442 <pre>
444 </pre>
445 If this bit is set, letters in the pattern match both upper and lower case
446 letters. It is equivalent to Perl's /i option, and it can be changed within a
447 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
448 concept of case for characters whose values are less than 128, so caseless
449 matching is always possible. For characters with higher values, the concept of
450 case is supported if PCRE is compiled with Unicode property support, but not
451 otherwise. If you want to use caseless matching for characters 128 and above,
452 you must ensure that PCRE is compiled with Unicode property support as well as
453 with UTF-8 support.
454 <pre>
456 </pre>
457 If this bit is set, a dollar metacharacter in the pattern matches only at the
458 end of the subject string. Without this option, a dollar also matches
459 immediately before a newline at the end of the string (but not before any other
460 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
461 There is no equivalent to this option in Perl, and no way to set it within a
462 pattern.
463 <pre>
465 </pre>
466 If this bit is set, a dot metacharater in the pattern matches all characters,
467 including those that indicate newline. Without it, a dot does not match when
468 the current position is at a newline. This option is equivalent to Perl's /s
469 option, and it can be changed within a pattern by a (?s) option setting. A
470 negative class such as [^a] always matches newlines, independent of the setting
471 of this option.
472 <pre>
474 </pre>
475 If this bit is set, names used to identify capturing subpatterns need not be
476 unique. This can be helpful for certain types of pattern when it is known that
477 only one instance of the named subpattern can ever be matched. There are more
478 details of named subpatterns below; see also the
479 <a href="pcrepattern.html"><b>pcrepattern</b></a>
480 documentation.
481 <pre>
483 </pre>
484 If this bit is set, whitespace data characters in the pattern are totally
485 ignored except when escaped or inside a character class. Whitespace does not
486 include the VT character (code 11). In addition, characters between an
487 unescaped # outside a character class and the next newline, inclusive, are also
488 ignored. This is equivalent to Perl's /x option, and it can be changed within a
489 pattern by a (?x) option setting.
490 </P>
491 <P>
492 This option makes it possible to include comments inside complicated patterns.
493 Note, however, that this applies only to data characters. Whitespace characters
494 may never appear within special character sequences in a pattern, for example
495 within the sequence (?( which introduces a conditional subpattern.
496 <pre>
498 </pre>
499 This option was invented in order to turn on additional functionality of PCRE
500 that is incompatible with Perl, but it is currently of very little use. When
501 set, any backslash in a pattern that is followed by a letter that has no
502 special meaning causes an error, thus reserving these combinations for future
503 expansion. By default, as in Perl, a backslash followed by a letter with no
504 special meaning is treated as a literal. (Perl can, however, be persuaded to
505 give a warning for this.) There are at present no other features controlled by
506 this option. It can also be set by a (?X) option setting within a pattern.
507 <pre>
509 </pre>
510 If this option is set, an unanchored pattern is required to match before or at
511 the first newline in the subject string, though the matched text may continue
512 over the newline.
513 <pre>
515 </pre>
516 By default, PCRE treats the subject string as consisting of a single line of
517 characters (even if it actually contains newlines). The "start of line"
518 metacharacter (^) matches only at the start of the string, while the "end of
519 line" metacharacter ($) matches only at the end of the string, or before a
520 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
521 Perl.
522 </P>
523 <P>
524 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
525 match immediately following or immediately before internal newlines in the
526 subject string, respectively, as well as at the very start and end. This is
527 equivalent to Perl's /m option, and it can be changed within a pattern by a
528 (?m) option setting. If there are no newlines in a subject string, or no
529 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
530 <pre>
534 </pre>
535 These options override the default newline definition that was chosen when PCRE
536 was built. Setting the first or the second specifies that a newline is
537 indicated by a single character (CR or LF, respectively). Setting both of them
538 specifies that a newline is indicated by the two-character CRLF sequence. For
539 convenience, PCRE_NEWLINE_CRLF is defined to contain both bits. The only time
540 that a line break is relevant when compiling a pattern is if PCRE_EXTENDED is
541 set, and an unescaped # outside a character class is encountered. This
542 indicates a comment that lasts until after the next newline.
543 </P>
544 <P>
545 The newline option set at compile time becomes the default that is used for
546 <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
547 <pre>
549 </pre>
550 If this option is set, it disables the use of numbered capturing parentheses in
551 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
552 were followed by ?: but named parentheses can still be used for capturing (and
553 they acquire numbers in the usual way). There is no equivalent of this option
554 in Perl.
555 <pre>
557 </pre>
558 This option inverts the "greediness" of the quantifiers so that they are not
559 greedy by default, but become greedy if followed by "?". It is not compatible
560 with Perl. It can also be set by a (?U) option setting within the pattern.
561 <pre>
563 </pre>
564 This option causes PCRE to regard both the pattern and the subject as strings
565 of UTF-8 characters instead of single-byte character strings. However, it is
566 available only when PCRE is built to include UTF-8 support. If not, the use
567 of this option provokes an error. Details of how this option changes the
568 behaviour of PCRE are given in the
569 <a href="pcre.html#utf8support">section on UTF-8 support</a>
570 in the main
571 <a href="pcre.html"><b>pcre</b></a>
572 page.
573 <pre>
575 </pre>
576 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
577 automatically checked. If an invalid UTF-8 sequence of bytes is found,
578 <b>pcre_compile()</b> returns an error. If you already know that your pattern is
579 valid, and you want to skip this check for performance reasons, you can set the
580 PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
581 UTF-8 string as a pattern is undefined. It may cause your program to crash.
582 Note that this option can also be passed to <b>pcre_exec()</b> and
583 <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
584 strings.
585 </P>
586 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
587 <P>
588 The following table lists the error codes than may be returned by
589 <b>pcre_compile2()</b>, along with the error messages that may be returned by
590 both compiling functions.
591 <pre>
592 0 no error
593 1 \ at end of pattern
594 2 \c at end of pattern
595 3 unrecognized character follows \
596 4 numbers out of order in {} quantifier
597 5 number too big in {} quantifier
598 6 missing terminating ] for character class
599 7 invalid escape sequence in character class
600 8 range out of order in character class
601 9 nothing to repeat
602 10 operand of unlimited repeat could match the empty string
603 11 internal error: unexpected repeat
604 12 unrecognized character after (?
605 13 POSIX named classes are supported only within a class
606 14 missing )
607 15 reference to non-existent subpattern
608 16 erroffset passed as NULL
609 17 unknown option bit(s) set
610 18 missing ) after comment
611 19 parentheses nested too deeply
612 20 regular expression too large
613 21 failed to get memory
614 22 unmatched parentheses
615 23 internal error: code overflow
616 24 unrecognized character after (?&#60;
617 25 lookbehind assertion is not fixed length
618 26 malformed number or name after (?(
619 27 conditional group contains more than two branches
620 28 assertion expected after (?(
621 29 (?R or (?digits must be followed by )
622 30 unknown POSIX class name
623 31 POSIX collating elements are not supported
624 32 this version of PCRE is not compiled with PCRE_UTF8 support
625 33 spare error
626 34 character value in \x{...} sequence is too large
627 35 invalid condition (?(0)
628 36 \C not allowed in lookbehind assertion
629 37 PCRE does not support \L, \l, \N, \U, or \u
630 38 number after (?C is &#62; 255
631 39 closing ) for (?C expected
632 40 recursive call could loop indefinitely
633 41 unrecognized character after (?P
634 42 syntax error after (?P
635 43 two named subpatterns have the same name
636 44 invalid UTF-8 string
637 45 support for \P, \p, and \X has not been compiled
638 46 malformed \P or \p sequence
639 47 unknown property name after \P or \p
640 48 subpattern name is too long (maximum 32 characters)
641 49 too many named subpatterns (maximum 10,000)
642 50 repeated subpattern is too long
643 51 octal value is greater than \377 (not in UTF-8 mode)
644 </PRE>
645 </P>
646 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
647 <P>
648 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
649 <b>const char **<i>errptr</i>);</b>
650 </P>
651 <P>
652 If a compiled pattern is going to be used several times, it is worth spending
653 more time analyzing it in order to speed up the time taken for matching. The
654 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
655 argument. If studying the pattern produces additional information that will
656 help speed up matching, <b>pcre_study()</b> returns a pointer to a
657 <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
658 results of the study.
659 </P>
660 <P>
661 The returned value from <b>pcre_study()</b> can be passed directly to
662 <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
663 fields that can be set by the caller before the block is passed; these are
664 described
665 <a href="#extradata">below</a>
666 in the section on matching a pattern.
667 </P>
668 <P>
669 If studying the pattern does not produce any additional information
670 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
671 wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
672 own <b>pcre_extra</b> block.
673 </P>
674 <P>
675 The second argument of <b>pcre_study()</b> contains option bits. At present, no
676 options are defined, and this argument should always be zero.
677 </P>
678 <P>
679 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
680 studying succeeds (even if no data is returned), the variable it points to is
681 set to NULL. Otherwise it is set to point to a textual error message. This is a
682 static string that is part of the library. You must not try to free it. You
683 should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
684 sure that it has run successfully.
685 </P>
686 <P>
687 This is a typical call to <b>pcre_study</b>():
688 <pre>
689 pcre_extra *pe;
690 pe = pcre_study(
691 re, /* result of pcre_compile() */
692 0, /* no options exist */
693 &error); /* set to NULL or points to a message */
694 </pre>
695 At present, studying a pattern is useful only for non-anchored patterns that do
696 not have a single fixed starting character. A bitmap of possible starting
697 bytes is created.
698 <a name="localesupport"></a></P>
699 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
700 <P>
701 PCRE handles caseless matching, and determines whether characters are letters
702 digits, or whatever, by reference to a set of tables, indexed by character
703 value. When running in UTF-8 mode, this applies only to characters with codes
704 less than 128. Higher-valued codes never match escapes such as \w or \d, but
705 can be tested with \p if PCRE is built with Unicode character property
706 support. The use of locales with Unicode is discouraged.
707 </P>
708 <P>
709 An internal set of tables is created in the default C locale when PCRE is
710 built. This is used when the final argument of <b>pcre_compile()</b> is NULL,
711 and is sufficient for many applications. An alternative set of tables can,
712 however, be supplied. These may be created in a different locale from the
713 default. As more and more applications change to using Unicode, the need for
714 this locale support is expected to die away.
715 </P>
716 <P>
717 External tables are built by calling the <b>pcre_maketables()</b> function,
718 which has no arguments, in the relevant locale. The result can then be passed
719 to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
720 example, to build and use tables that are appropriate for the French locale
721 (where accented characters with values greater than 128 are treated as letters),
722 the following code could be used:
723 <pre>
724 setlocale(LC_CTYPE, "fr_FR");
725 tables = pcre_maketables();
726 re = pcre_compile(..., tables);
727 </pre>
728 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
729 obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
730 that the memory containing the tables remains available for as long as it is
731 needed.
732 </P>
733 <P>
734 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
735 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
736 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
737 pattern, compilation, studying and matching all happen in the same locale, but
738 different patterns can be compiled in different locales.
739 </P>
740 <P>
741 It is possible to pass a table pointer or NULL (indicating the use of the
742 internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
743 this facility could be used to match a pattern in a different locale from the
744 one in which it was compiled. Passing table pointers at run time is discussed
745 below in the section on matching a pattern.
746 </P>
747 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
748 <P>
749 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
750 <b>int <i>what</i>, void *<i>where</i>);</b>
751 </P>
752 <P>
753 The <b>pcre_fullinfo()</b> function returns information about a compiled
754 pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
755 nevertheless retained for backwards compability (and is documented below).
756 </P>
757 <P>
758 The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
759 pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
760 the pattern was not studied. The third argument specifies which piece of
761 information is required, and the fourth argument is a pointer to a variable
762 to receive the data. The yield of the function is zero for success, or one of
763 the following negative numbers:
764 <pre>
765 PCRE_ERROR_NULL the argument <i>code</i> was NULL
766 the argument <i>where</i> was NULL
767 PCRE_ERROR_BADMAGIC the "magic number" was not found
768 PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
769 </pre>
770 The "magic number" is placed at the start of each compiled pattern as an simple
771 check against passing an arbitrary memory pointer. Here is a typical call of
772 <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
773 <pre>
774 int rc;
775 size_t length;
776 rc = pcre_fullinfo(
777 re, /* result of pcre_compile() */
778 pe, /* result of pcre_study(), or NULL */
779 PCRE_INFO_SIZE, /* what is required */
780 &length); /* where to put the data */
781 </pre>
782 The possible values for the third argument are defined in <b>pcre.h</b>, and are
783 as follows:
784 <pre>
786 </pre>
787 Return the number of the highest back reference in the pattern. The fourth
788 argument should point to an <b>int</b> variable. Zero is returned if there are
789 no back references.
790 <pre>
792 </pre>
793 Return the number of capturing subpatterns in the pattern. The fourth argument
794 should point to an <b>int</b> variable.
795 <pre>
797 </pre>
798 Return a pointer to the internal default character tables within PCRE. The
799 fourth argument should point to an <b>unsigned char *</b> variable. This
800 information call is provided for internal use by the <b>pcre_study()</b>
801 function. External callers can cause PCRE to use its internal tables by passing
802 a NULL table pointer.
803 <pre>
805 </pre>
806 Return information about the first byte of any matched string, for a
807 non-anchored pattern. The fourth argument should point to an <b>int</b>
808 variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
809 still recognized for backwards compatibility.)
810 </P>
811 <P>
812 If there is a fixed first byte, for example, from a pattern such as
813 (cat|cow|coyote). Otherwise, if either
814 <br>
815 <br>
816 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
817 starts with "^", or
818 <br>
819 <br>
820 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
821 (if it were set, the pattern would be anchored),
822 <br>
823 <br>
824 -1 is returned, indicating that the pattern matches only at the start of a
825 subject string or after any newline within the string. Otherwise -2 is
826 returned. For anchored patterns, -2 is returned.
827 <pre>
829 </pre>
830 If the pattern was studied, and this resulted in the construction of a 256-bit
831 table indicating a fixed set of bytes for the first byte in any matching
832 string, a pointer to the table is returned. Otherwise NULL is returned. The
833 fourth argument should point to an <b>unsigned char *</b> variable.
834 <pre>
836 </pre>
837 Return the value of the rightmost literal byte that must exist in any matched
838 string, other than at its start, if such a byte has been recorded. The fourth
839 argument should point to an <b>int</b> variable. If there is no such byte, -1 is
840 returned. For anchored patterns, a last literal byte is recorded only if it
841 follows something of variable length. For example, for the pattern
842 /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
843 is -1.
844 <pre>
848 </pre>
849 PCRE supports the use of named as well as numbered capturing parentheses. The
850 names are just an additional way of identifying the parentheses, which still
851 acquire numbers. Several convenience functions such as
852 <b>pcre_get_named_substring()</b> are provided for extracting captured
853 substrings by name. It is also possible to extract the data directly, by first
854 converting the name to a number in order to access the correct pointers in the
855 output vector (described with <b>pcre_exec()</b> below). To do the conversion,
856 you need to use the name-to-number map, which is described by these three
857 values.
858 </P>
859 <P>
860 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
861 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
862 entry; both of these return an <b>int</b> value. The entry size depends on the
863 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
864 entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
865 are the number of the capturing parenthesis, most significant byte first. The
866 rest of the entry is the corresponding name, zero terminated. The names are in
867 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
868 their parentheses numbers. For example, consider the following pattern (assume
869 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
870 <pre>
871 (?P&#60;date&#62; (?P&#60;year&#62;(\d\d)?\d\d) - (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )
872 </pre>
873 There are four named subpatterns, so the table has four entries, and each entry
874 in the table is eight bytes long. The table is as follows, with non-printing
875 bytes shows in hexadecimal, and undefined bytes shown as ??:
876 <pre>
877 00 01 d a t e 00 ??
878 00 05 d a y 00 ?? ??
879 00 04 m o n t h 00
880 00 02 y e a r 00 ??
881 </pre>
882 When writing code to extract data from named subpatterns using the
883 name-to-number map, remember that the length of the entries is likely to be
884 different for each compiled pattern.
885 <pre>
887 </pre>
888 Return a copy of the options with which the pattern was compiled. The fourth
889 argument should point to an <b>unsigned long int</b> variable. These option bits
890 are those specified in the call to <b>pcre_compile()</b>, modified by any
891 top-level option settings within the pattern itself.
892 </P>
893 <P>
894 A pattern is automatically anchored by PCRE if all of its top-level
895 alternatives begin with one of the following:
896 <pre>
897 ^ unless PCRE_MULTILINE is set
898 \A always
899 \G always
900 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
901 </pre>
902 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
903 <b>pcre_fullinfo()</b>.
904 <pre>
906 </pre>
907 Return the size of the compiled pattern, that is, the value that was passed as
908 the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
909 place the compiled data. The fourth argument should point to a <b>size_t</b>
910 variable.
911 <pre>
913 </pre>
914 Return the size of the data block pointed to by the <i>study_data</i> field in
915 a <b>pcre_extra</b> block. That is, it is the value that was passed to
916 <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
917 created by <b>pcre_study()</b>. The fourth argument should point to a
918 <b>size_t</b> variable.
919 </P>
920 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
921 <P>
922 <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
923 <b>*<i>firstcharptr</i>);</b>
924 </P>
925 <P>
926 The <b>pcre_info()</b> function is now obsolete because its interface is too
927 restrictive to return all the available data about a compiled pattern. New
928 programs should use <b>pcre_fullinfo()</b> instead. The yield of
929 <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
930 following negative numbers:
931 <pre>
932 PCRE_ERROR_NULL the argument <i>code</i> was NULL
933 PCRE_ERROR_BADMAGIC the "magic number" was not found
934 </pre>
935 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
936 pattern was compiled is placed in the integer it points to (see
938 </P>
939 <P>
940 If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
941 it is used to pass back information about the first character of any matched
942 string (see PCRE_INFO_FIRSTBYTE above).
943 </P>
944 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
945 <P>
946 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
947 </P>
948 <P>
949 The <b>pcre_refcount()</b> function is used to maintain a reference count in the
950 data block that contains a compiled pattern. It is provided for the benefit of
951 applications that operate in an object-oriented manner, where different parts
952 of the application may be using the same compiled pattern, but you want to free
953 the block when they are all done.
954 </P>
955 <P>
956 When a pattern is compiled, the reference count field is initialized to zero.
957 It is changed only by calling this function, whose action is to add the
958 <i>adjust</i> value (which may be positive or negative) to it. The yield of the
959 function is the new value. However, the value of the count is constrained to
960 lie between 0 and 65535, inclusive. If the new value is outside these limits,
961 it is forced to the appropriate limit value.
962 </P>
963 <P>
964 Except when it is zero, the reference count is not correctly preserved if a
965 pattern is compiled on one host and then transferred to a host whose byte-order
966 is different. (This seems a highly unlikely scenario.)
967 </P>
968 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
969 <P>
970 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
971 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
972 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
973 </P>
974 <P>
975 The function <b>pcre_exec()</b> is called to match a subject string against a
976 compiled pattern, which is passed in the <i>code</i> argument. If the
977 pattern has been studied, the result of the study should be passed in the
978 <i>extra</i> argument. This function is the main matching facility of the
979 library, and it operates in a Perl-like manner. For specialist use there is
980 also an alternative matching function, which is described
981 <a href="#dfamatch">below</a>
982 in the section about the <b>pcre_dfa_exec()</b> function.
983 </P>
984 <P>
985 In most applications, the pattern will have been compiled (and optionally
986 studied) in the same process that calls <b>pcre_exec()</b>. However, it is
987 possible to save compiled patterns and study data, and then use them later
988 in different processes, possibly even on different hosts. For a discussion
989 about this, see the
990 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
991 documentation.
992 </P>
993 <P>
994 Here is an example of a simple call to <b>pcre_exec()</b>:
995 <pre>
996 int rc;
997 int ovector[30];
998 rc = pcre_exec(
999 re, /* result of pcre_compile() */
1000 NULL, /* we didn't study the pattern */
1001 "some string", /* the subject string */
1002 11, /* the length of the subject string */
1003 0, /* start at offset 0 in the subject */
1004 0, /* default options */
1005 ovector, /* vector of integers for substring information */
1006 30); /* number of elements (NOT size in bytes) */
1007 <a name="extradata"></a></PRE>
1008 </P>
1009 <br><b>
1010 Extra data for <b>pcre_exec()</b>
1011 </b><br>
1012 <P>
1013 If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1014 data block. The <b>pcre_study()</b> function returns such a block (when it
1015 doesn't return NULL), but you can also create one for yourself, and pass
1016 additional information in it. The <b>pcre_extra</b> block contains the following
1017 fields (not necessarily in this order):
1018 <pre>
1019 unsigned long int <i>flags</i>;
1020 void *<i>study_data</i>;
1021 unsigned long int <i>match_limit</i>;
1022 unsigned long int <i>match_limit_recursion</i>;
1023 void *<i>callout_data</i>;
1024 const unsigned char *<i>tables</i>;
1025 </pre>
1026 The <i>flags</i> field is a bitmap that specifies which of the other fields
1027 are set. The flag bits are:
1028 <pre>
1034 </pre>
1035 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1036 <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1037 the appropriate flag bit. You should not set this yourself, but you may add to
1038 the block by setting the other fields and their corresponding flag bits.
1039 </P>
1040 <P>
1041 The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1042 vast amount of resources when running patterns that are not going to match,
1043 but which have a very large number of possibilities in their search trees. The
1044 classic example is the use of nested unlimited repeats.
1045 </P>
1046 <P>
1047 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1048 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1049 number of times this function is called during a match, which has the effect of
1050 limiting the amount of backtracking that can take place. For patterns that are
1051 not anchored, the count restarts from zero for each position in the subject
1052 string.
1053 </P>
1054 <P>
1055 The default value for the limit can be set when PCRE is built; the default
1056 default is 10 million, which handles all but the most extreme cases. You can
1057 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1058 block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1059 the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1061 </P>
1062 <P>
1063 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1064 instead of limiting the total number of times that <b>match()</b> is called, it
1065 limits the depth of recursion. The recursion depth is a smaller number than the
1066 total number of calls, because not all calls to <b>match()</b> are recursive.
1067 This limit is of use only if it is set smaller than <i>match_limit</i>.
1068 </P>
1069 <P>
1070 Limiting the recursion depth limits the amount of stack that can be used, or,
1071 when PCRE has been compiled to use memory on the heap instead of the stack, the
1072 amount of heap memory that can be used.
1073 </P>
1074 <P>
1075 The default value for <i>match_limit_recursion</i> can be set when PCRE is
1076 built; the default default is the same value as the default for
1077 <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1078 with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1079 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1080 is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1081 </P>
1082 <P>
1083 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1084 which is described in the
1085 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1086 documentation.
1087 </P>
1088 <P>
1089 The <i>tables</i> field is used to pass a character tables pointer to
1090 <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1091 pattern. A non-NULL value is stored with the compiled pattern only if custom
1092 tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1093 If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1094 internal tables to be used. This facility is helpful when re-using patterns
1095 that have been saved after compiling with an external set of tables, because
1096 the external tables might be at a different address when <b>pcre_exec()</b> is
1097 called. See the
1098 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1099 documentation for a discussion of saving compiled patterns for later use.
1100 </P>
1101 <br><b>
1102 Option bits for <b>pcre_exec()</b>
1103 </b><br>
1104 <P>
1105 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1106 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1108 <pre>
1110 </pre>
1111 The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1112 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1113 to be anchored by virtue of its contents, it cannot be made unachored at
1114 matching time.
1115 <pre>
1119 </pre>
1120 These options override the newline definition that was chosen or defaulted when
1121 the pattern was compiled. For details, see the description <b>pcre_compile()</b>
1122 above. During matching, the newline choice affects the behaviour of the dot,
1123 circumflex, and dollar metacharacters.
1124 <pre>
1126 </pre>
1127 This option specifies that first character of the subject string is not the
1128 beginning of a line, so the circumflex metacharacter should not match before
1129 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1130 never to match. This option affects only the behaviour of the circumflex
1131 metacharacter. It does not affect \A.
1132 <pre>
1134 </pre>
1135 This option specifies that the end of the subject string is not the end of a
1136 line, so the dollar metacharacter should not match it nor (except in multiline
1137 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1138 compile time) causes dollar never to match. This option affects only the
1139 behaviour of the dollar metacharacter. It does not affect \Z or \z.
1140 <pre>
1142 </pre>
1143 An empty string is not considered to be a valid match if this option is set. If
1144 there are alternatives in the pattern, they are tried. If all the alternatives
1145 match the empty string, the entire match fails. For example, if the pattern
1146 <pre>
1147 a?b?
1148 </pre>
1149 is applied to a string not beginning with "a" or "b", it matches the empty
1150 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1151 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1152 </P>
1153 <P>
1154 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1155 of a pattern match of the empty string within its <b>split()</b> function, and
1156 when using the /g modifier. It is possible to emulate Perl's behaviour after
1157 matching a null string by first trying the match again at the same offset with
1158 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1159 starting offset (see below) and trying an ordinary match again. There is some
1160 code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1161 <pre>
1163 </pre>
1164 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1165 string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1166 The value of <i>startoffset</i> is also checked to ensure that it points to the
1167 start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1168 <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1169 contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1170 </P>
1171 <P>
1172 If you already know that your subject is valid, and you want to skip these
1173 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1174 calling <b>pcre_exec()</b>. You might want to do this for the second and
1175 subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1176 all the matches in a single subject string. However, you should be sure that
1177 the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1178 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1179 subject, or a value of <i>startoffset</i> that does not point to the start of a
1180 UTF-8 character, is undefined. Your program may crash.
1181 <pre>
1183 </pre>
1184 This option turns on the partial matching feature. If the subject string fails
1185 to match the pattern, but at some point during the matching process the end of
1186 the subject was reached (that is, the subject partially matches the pattern and
1187 the failure to match occurred only because there were not enough subject
1188 characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1189 PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1190 may appear in the pattern. These are discussed in the
1191 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1192 documentation.
1193 </P>
1194 <br><b>
1195 The string to be matched by <b>pcre_exec()</b>
1196 </b><br>
1197 <P>
1198 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1199 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1200 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1201 UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1202 bytes. When the starting offset is zero, the search for a match starts at the
1203 beginning of the subject, and this is by far the most common case.
1204 </P>
1205 <P>
1206 A non-zero starting offset is useful when searching for another match in the
1207 same subject by calling <b>pcre_exec()</b> again after a previous success.
1208 Setting <i>startoffset</i> differs from just passing over a shortened string and
1209 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1210 lookbehind. For example, consider the pattern
1211 <pre>
1212 \Biss\B
1213 </pre>
1214 which finds occurrences of "iss" in the middle of words. (\B matches only if
1215 the current position in the subject is not a word boundary.) When applied to
1216 the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1217 occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1218 subject, namely "issipi", it does not match, because \B is always false at the
1219 start of the subject, which is deemed to be a word boundary. However, if
1220 <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1221 set to 4, it finds the second occurrence of "iss" because it is able to look
1222 behind the starting point to discover that it is preceded by a letter.
1223 </P>
1224 <P>
1225 If a non-zero starting offset is passed when the pattern is anchored, one
1226 attempt to match at the given offset is made. This can only succeed if the
1227 pattern does not require the match to be at the start of the subject.
1228 </P>
1229 <br><b>
1230 How <b>pcre_exec()</b> returns captured substrings
1231 </b><br>
1232 <P>
1233 In general, a pattern matches a certain portion of the subject, and in
1234 addition, further substrings from the subject may be picked out by parts of the
1235 pattern. Following the usage in Jeffrey Friedl's book, this is called
1236 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1237 a fragment of a pattern that picks out a substring. PCRE supports several other
1238 kinds of parenthesized subpattern that do not cause substrings to be captured.
1239 </P>
1240 <P>
1241 Captured substrings are returned to the caller via a vector of integer offsets
1242 whose address is passed in <i>ovector</i>. The number of elements in the vector
1243 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1244 this argument is NOT the size of <i>ovector</i> in bytes.
1245 </P>
1246 <P>
1247 The first two-thirds of the vector is used to pass back captured substrings,
1248 each substring using a pair of integers. The remaining third of the vector is
1249 used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1250 and is not available for passing back information. The length passed in
1251 <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1252 rounded down.
1253 </P>
1254 <P>
1255 When a match is successful, information about captured substrings is returned
1256 in pairs of integers, starting at the beginning of <i>ovector</i>, and
1257 continuing up to two-thirds of its length at the most. The first element of a
1258 pair is set to the offset of the first character in a substring, and the second
1259 is set to the offset of the first character after the end of a substring. The
1260 first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1261 subject string matched by the entire pattern. The next pair is used for the
1262 first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1263 is one more than the highest numbered pair that has been set. For example, if
1264 two substrings have been captured, the returned value is 3. If there are no
1265 capturing subpatterns, the return value from a successful match is 1,
1266 indicating that just the first pair of offsets has been set.
1267 </P>
1268 <P>
1269 If a capturing subpattern is matched repeatedly, it is the last portion of the
1270 string that it matched that is returned.
1271 </P>
1272 <P>
1273 If the vector is too small to hold all the captured substring offsets, it is
1274 used as far as possible (up to two-thirds of its length), and the function
1275 returns a value of zero. In particular, if the substring offsets are not of
1276 interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1277 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1278 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1279 has to get additional memory for use during matching. Thus it is usually
1280 advisable to supply an <i>ovector</i>.
1281 </P>
1282 <P>
1283 The <b>pcre_info()</b> function can be used to find out how many capturing
1284 subpatterns there are in a compiled pattern. The smallest size for
1285 <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1286 the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1287 </P>
1288 <P>
1289 It is possible for capturing subpattern number <i>n+1</i> to match some part of
1290 the subject when subpattern <i>n</i> has not been used at all. For example, if
1291 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1292 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1293 happens, both values in the offset pairs corresponding to unused subpatterns
1294 are set to -1.
1295 </P>
1296 <P>
1297 Offset values that correspond to unused subpatterns at the end of the
1298 expression are also set to -1. For example, if the string "abc" is matched
1299 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1300 return from the function is 2, because the highest used capturing subpattern
1301 number is 1. However, you can refer to the offsets for the second and third
1302 capturing subpatterns if you wish (assuming the vector is large enough, of
1303 course).
1304 </P>
1305 <P>
1306 Some convenience functions are provided for extracting the captured substrings
1307 as separate strings. These are described below.
1308 <a name="errorlist"></a></P>
1309 <br><b>
1310 Error return values from <b>pcre_exec()</b>
1311 </b><br>
1312 <P>
1313 If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1314 defined in the header file:
1315 <pre>
1317 </pre>
1318 The subject string did not match the pattern.
1319 <pre>
1321 </pre>
1322 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1323 NULL and <i>ovecsize</i> was not zero.
1324 <pre>
1326 </pre>
1327 An unrecognized bit was set in the <i>options</i> argument.
1328 <pre>
1330 </pre>
1331 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1332 the case when it is passed a junk pointer and to detect when a pattern that was
1333 compiled in an environment of one endianness is run in an environment with the
1334 other endianness. This is the error that PCRE gives when the magic number is
1335 not present.
1336 <pre>
1338 </pre>
1339 While running the pattern match, an unknown item was encountered in the
1340 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1341 of the compiled pattern.
1342 <pre>
1344 </pre>
1345 If a pattern contains back references, but the <i>ovector</i> that is passed to
1346 <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1347 gets a block of memory at the start of matching to use for this purpose. If the
1348 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1349 automatically freed at the end of matching.
1350 <pre>
1352 </pre>
1353 This error is used by the <b>pcre_copy_substring()</b>,
1354 <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1355 below). It is never returned by <b>pcre_exec()</b>.
1356 <pre>
1358 </pre>
1359 The backtracking limit, as specified by the <i>match_limit</i> field in a
1360 <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1361 above.
1362 <pre>
1364 </pre>
1365 The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1366 field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1367 description above.
1368 <pre>
1370 </pre>
1371 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1372 use by callout functions that want to yield a distinctive error code. See the
1373 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1374 documentation for details.
1375 <pre>
1377 </pre>
1378 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1379 <pre>
1381 </pre>
1382 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1383 of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1384 <pre>
1386 </pre>
1387 The subject string did not match, but it did match partially. See the
1388 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1389 documentation for details of partial matching.
1390 <pre>
1392 </pre>
1393 The PCRE_PARTIAL option was used with a compiled pattern containing items that
1394 are not supported for partial matching. See the
1395 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1396 documentation for details of partial matching.
1397 <pre>
1399 </pre>
1400 An unexpected internal error has occurred. This error could be caused by a bug
1401 in PCRE or by overwriting of the compiled pattern.
1402 <pre>
1404 </pre>
1405 This error is given if the value of the <i>ovecsize</i> argument is negative.
1406 </P>
1407 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1408 <P>
1409 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1410 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1411 <b>int <i>buffersize</i>);</b>
1412 </P>
1413 <P>
1414 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1415 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1416 <b>const char **<i>stringptr</i>);</b>
1417 </P>
1418 <P>
1419 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1420 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1421 </P>
1422 <P>
1423 Captured substrings can be accessed directly by using the offsets returned by
1424 <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1425 <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1426 <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1427 as new, separate, zero-terminated strings. These functions identify substrings
1428 by number. The next section describes functions for extracting named
1429 substrings.
1430 </P>
1431 <P>
1432 A substring that contains a binary zero is correctly extracted and has a
1433 further zero added on the end, but the result is not, of course, a C string.
1434 However, you can process such a string by referring to the length that is
1435 returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1436 Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1437 for handling strings containing binary zeros, because the end of the final
1438 string is not independently indicated.
1439 </P>
1440 <P>
1441 The first three arguments are the same for all three of these functions:
1442 <i>subject</i> is the subject string that has just been successfully matched,
1443 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1444 <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1445 captured by the match, including the substring that matched the entire regular
1446 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1447 than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1448 space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1449 number of elements in the vector divided by three.
1450 </P>
1451 <P>
1452 The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1453 extract a single substring, whose number is given as <i>stringnumber</i>. A
1454 value of zero extracts the substring that matched the entire pattern, whereas
1455 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1456 the string is placed in <i>buffer</i>, whose length is given by
1457 <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1458 obtained via <b>pcre_malloc</b>, and its address is returned via
1459 <i>stringptr</i>. The yield of the function is the length of the string, not
1460 including the terminating zero, or one of
1461 <pre>
1463 </pre>
1464 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1465 memory failed for <b>pcre_get_substring()</b>.
1466 <pre>
1468 </pre>
1469 There is no substring whose number is <i>stringnumber</i>.
1470 </P>
1471 <P>
1472 The <b>pcre_get_substring_list()</b> function extracts all available substrings
1473 and builds a list of pointers to them. All this is done in a single block of
1474 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1475 is returned via <i>listptr</i>, which is also the start of the list of string
1476 pointers. The end of the list is marked by a NULL pointer. The yield of the
1477 function is zero if all went well, or
1478 <pre>
1480 </pre>
1481 if the attempt to get the memory block failed.
1482 </P>
1483 <P>
1484 When any of these functions encounter a substring that is unset, which can
1485 happen when capturing subpattern number <i>n+1</i> matches some part of the
1486 subject, but subpattern <i>n</i> has not been used at all, they return an empty
1487 string. This can be distinguished from a genuine zero-length substring by
1488 inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1489 substrings.
1490 </P>
1491 <P>
1492 The two convenience functions <b>pcre_free_substring()</b> and
1493 <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1494 a previous call of <b>pcre_get_substring()</b> or
1495 <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1496 the function pointed to by <b>pcre_free</b>, which of course could be called
1497 directly from a C program. However, PCRE is used in some situations where it is
1498 linked via a special interface to another programming language that cannot use
1499 <b>pcre_free</b> directly; it is for these cases that the functions are
1500 provided.
1501 </P>
1502 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1503 <P>
1504 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1505 <b>const char *<i>name</i>);</b>
1506 </P>
1507 <P>
1508 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1509 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1510 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1511 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1512 </P>
1513 <P>
1514 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1515 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1516 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1517 <b>const char **<i>stringptr</i>);</b>
1518 </P>
1519 <P>
1520 To extract a substring by name, you first have to find associated number.
1521 For example, for this pattern
1522 <pre>
1523 (a+)b(?P&#60;xxx&#62;\d+)...
1524 </pre>
1525 the number of the subpattern called "xxx" is 2. If the name is known to be
1526 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1527 calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1528 pattern, and the second is the name. The yield of the function is the
1529 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1530 that name.
1531 </P>
1532 <P>
1533 Given the number, you can extract the substring directly, or use one of the
1534 functions described in the previous section. For convenience, there are also
1535 two functions that do the whole job.
1536 </P>
1537 <P>
1538 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1539 <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1540 functions that extract by number. As these are described in the previous
1541 section, they are not re-described here. There are just two differences:
1542 </P>
1543 <P>
1544 First, instead of a substring number, a substring name is given. Second, there
1545 is an extra argument, given at the start, which is a pointer to the compiled
1546 pattern. This is needed in order to gain access to the name-to-number
1547 translation table.
1548 </P>
1549 <P>
1550 These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1551 then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as
1552 appropriate.
1553 </P>
1554 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1555 <P>
1556 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1557 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1558 </P>
1559 <P>
1560 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1561 are not required to be unique. Normally, patterns with duplicate names are such
1562 that in any one match, only one of the named subpatterns participates. An
1563 example is shown in the
1564 <a href="pcrepattern.html"><b>pcrepattern</b></a>
1565 documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1566 and <b>pcre_get_named_substring()</b> return the first substring corresponding
1567 to the given name that is set. If none are set, an empty string is returned.
1568 The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1569 associated with the name, but it is not defined which it is.
1570 <br>
1571 <br>
1572 If you want to get full details of all captured substrings for a given name,
1573 you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1574 argument is the compiled pattern, and the second is the name. The third and
1575 fourth are pointers to variables which are updated by the function. After it
1576 has run, they point to the first and last entries in the name-to-number table
1577 for the given name. The function itself returns the length of each entry, or
1578 PCRE_ERROR_NOSUBSTRING if there are none. The format of the table is described
1579 above in the section entitled <i>Information about a pattern</i>. Given all the
1580 relevant entries for the name, you can extract each of their numbers, and hence
1581 the captured data, if any.
1582 </P>
1583 <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1584 <P>
1585 The traditional matching function uses a similar algorithm to Perl, which stops
1586 when it finds the first match, starting at a given point in the subject. If you
1587 want to find all possible matches, or the longest possible match, consider
1588 using the alternative matching function (see below) instead. If you cannot use
1589 the alternative function, but still need to find all possible matches, you
1590 can kludge it up by making use of the callout facility, which is described in
1591 the
1592 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1593 documentation.
1594 </P>
1595 <P>
1596 What you have to do is to insert a callout right at the end of the pattern.
1597 When your callout function is called, extract and save the current matched
1598 substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1599 other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1600 will yield PCRE_ERROR_NOMATCH.
1601 <a name="dfamatch"></a></P>
1602 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1603 <P>
1604 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1605 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1606 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1607 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1608 </P>
1609 <P>
1610 The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1611 a compiled pattern, using a "DFA" matching algorithm. This has different
1612 characteristics to the normal algorithm, and is not compatible with Perl. Some
1613 of the features of PCRE patterns are not supported. Nevertheless, there are
1614 times when this kind of matching can be useful. For a discussion of the two
1615 matching algorithms, see the
1616 <a href="pcrematching.html"><b>pcrematching</b></a>
1617 documentation.
1618 </P>
1619 <P>
1620 The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1621 <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1622 different way, and this is described below. The other common arguments are used
1623 in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1624 here.
1625 </P>
1626 <P>
1627 The two additional arguments provide workspace for the function. The workspace
1628 vector should contain at least 20 elements. It is used for keeping track of
1629 multiple paths through the pattern tree. More workspace will be needed for
1630 patterns and subjects where there are a lot of potential matches.
1631 </P>
1632 <P>
1633 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1634 <pre>
1635 int rc;
1636 int ovector[10];
1637 int wspace[20];
1638 rc = pcre_dfa_exec(
1639 re, /* result of pcre_compile() */
1640 NULL, /* we didn't study the pattern */
1641 "some string", /* the subject string */
1642 11, /* the length of the subject string */
1643 0, /* start at offset 0 in the subject */
1644 0, /* default options */
1645 ovector, /* vector of integers for substring information */
1646 10, /* number of elements (NOT size in bytes) */
1647 wspace, /* working space vector */
1648 20); /* number of elements (NOT size in bytes) */
1649 </PRE>
1650 </P>
1651 <br><b>
1652 Option bits for <b>pcre_dfa_exec()</b>
1653 </b><br>
1654 <P>
1655 The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1656 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1658 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1659 the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1660 <pre>
1662 </pre>
1663 This has the same general effect as it does for <b>pcre_exec()</b>, but the
1664 details are slightly different. When PCRE_PARTIAL is set for
1665 <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1666 PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1667 complete matches, but there is still at least one matching possibility. The
1668 portion of the string that provided the partial match is set as the first
1669 matching string.
1670 <pre>
1672 </pre>
1673 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1674 soon as it has found one match. Because of the way the DFA algorithm works,
1675 this is necessarily the shortest possible match at the first possible matching
1676 point in the subject string.
1677 <pre>
1679 </pre>
1680 When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1681 a partial match, it is possible to call it again, with additional subject
1682 characters, and have it continue with the same match. The PCRE_DFA_RESTART
1683 option requests this action; when it is set, the <i>workspace</i> and
1684 <i>wscount</i> options must reference the same vector as before because data
1685 about the match so far is left in them after a partial match. There is more
1686 discussion of this facility in the
1687 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1688 documentation.
1689 </P>
1690 <br><b>
1691 Successful returns from <b>pcre_dfa_exec()</b>
1692 </b><br>
1693 <P>
1694 When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1695 substring in the subject. Note, however, that all the matches from one run of
1696 the function start at the same point in the subject. The shorter matches are
1697 all initial substrings of the longer matches. For example, if the pattern
1698 <pre>
1699 &#60;.*&#62;
1700 </pre>
1701 is matched against the string
1702 <pre>
1703 This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1704 </pre>
1705 the three matched strings are
1706 <pre>
1707 &#60;something&#62;
1708 &#60;something&#62; &#60;something else&#62;
1709 &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1710 </pre>
1711 On success, the yield of the function is a number greater than zero, which is
1712 the number of matched substrings. The substrings themselves are returned in
1713 <i>ovector</i>. Each string uses two elements; the first is the offset to the
1714 start, and the second is the offset to the end. All the strings have the same
1715 start offset. (Space could have been saved by giving this only once, but it was
1716 decided to retain some compatibility with the way <b>pcre_exec()</b> returns
1717 data, even though the meaning of the strings is different.)
1718 </P>
1719 <P>
1720 The strings are returned in reverse order of length; that is, the longest
1721 matching string is given first. If there were too many matches to fit into
1722 <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1723 the longest matches.
1724 </P>
1725 <br><b>
1726 Error returns from <b>pcre_dfa_exec()</b>
1727 </b><br>
1728 <P>
1729 The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1730 Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1731 described
1732 <a href="#errorlist">above.</a>
1733 There are in addition the following errors that are specific to
1734 <b>pcre_dfa_exec()</b>:
1735 <pre>
1737 </pre>
1738 This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1739 that it does not support, for instance, the use of \C or a back reference.
1740 <pre>
1742 </pre>
1743 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item in a
1744 pattern that uses a back reference for the condition. This is not supported.
1745 <pre>
1747 </pre>
1748 This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1749 block that contains a setting of the <i>match_limit</i> field. This is not
1750 supported (it is meaningless).
1751 <pre>
1753 </pre>
1754 This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1755 <i>workspace</i> vector.
1756 <pre>
1758 </pre>
1759 When a recursive subpattern is processed, the matching function calls itself
1760 recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1761 error is given if the output vector is not large enough. This should be
1762 extremely rare, as a vector of size 1000 is used.
1763 </P>
1764 <P>
1765 Last updated: 08 June 2006
1766 <br>
1767 Copyright &copy; 1997-2006 University of Cambridge.
1768 <p>
1769 Return to the <a href="index.html">PCRE index page</a>.
1770 </p>

  ViewVC Help
Powered by ViewVC 1.1.5