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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 87 - (show annotations)
Sat Feb 24 21:41:21 2007 UTC (12 years, 5 months ago) by nigel
File MIME type: text/html
File size: 73924 byte(s)
Load pcre-6.5 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">MULTITHREADING</a>
19 <li><a name="TOC4" href="#SEC4">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
20 <li><a name="TOC5" href="#SEC5">CHECKING BUILD-TIME OPTIONS</a>
21 <li><a name="TOC6" href="#SEC6">COMPILING A PATTERN</a>
22 <li><a name="TOC7" href="#SEC7">COMPILATION ERROR CODES</a>
23 <li><a name="TOC8" href="#SEC8">STUDYING A PATTERN</a>
24 <li><a name="TOC9" href="#SEC9">LOCALE SUPPORT</a>
25 <li><a name="TOC10" href="#SEC10">INFORMATION ABOUT A PATTERN</a>
26 <li><a name="TOC11" href="#SEC11">OBSOLETE INFO FUNCTION</a>
27 <li><a name="TOC12" href="#SEC12">REFERENCE COUNTS</a>
28 <li><a name="TOC13" href="#SEC13">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
29 <li><a name="TOC14" href="#SEC14">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
30 <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
31 <li><a name="TOC16" href="#SEC16">FINDING ALL POSSIBLE MATCHES</a>
32 <li><a name="TOC17" href="#SEC17">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
33 </ul>
34 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
35 <P>
36 <b>#include &#60;pcre.h&#62;</b>
37 </P>
38 <P>
39 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
40 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
41 <b>const unsigned char *<i>tableptr</i>);</b>
42 </P>
43 <P>
44 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
45 <b>int *<i>errorcodeptr</i>,</b>
46 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
47 <b>const unsigned char *<i>tableptr</i>);</b>
48 </P>
49 <P>
50 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
51 <b>const char **<i>errptr</i>);</b>
52 </P>
53 <P>
54 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
55 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
56 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
57 </P>
58 <P>
59 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
60 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
61 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
62 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
63 </P>
64 <P>
65 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
66 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
67 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
68 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
69 </P>
70 <P>
71 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
72 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
73 <b>int <i>buffersize</i>);</b>
74 </P>
75 <P>
76 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
77 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
78 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
79 <b>const char **<i>stringptr</i>);</b>
80 </P>
81 <P>
82 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
83 <b>const char *<i>name</i>);</b>
84 </P>
85 <P>
86 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
87 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
88 <b>const char **<i>stringptr</i>);</b>
89 </P>
90 <P>
91 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
92 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
93 </P>
94 <P>
95 <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
96 </P>
97 <P>
98 <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
99 </P>
100 <P>
101 <b>const unsigned char *pcre_maketables(void);</b>
102 </P>
103 <P>
104 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
105 <b>int <i>what</i>, void *<i>where</i>);</b>
106 </P>
107 <P>
108 <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
109 <b>*<i>firstcharptr</i>);</b>
110 </P>
111 <P>
112 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
113 </P>
114 <P>
115 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
116 </P>
117 <P>
118 <b>char *pcre_version(void);</b>
119 </P>
120 <P>
121 <b>void *(*pcre_malloc)(size_t);</b>
122 </P>
123 <P>
124 <b>void (*pcre_free)(void *);</b>
125 </P>
126 <P>
127 <b>void *(*pcre_stack_malloc)(size_t);</b>
128 </P>
129 <P>
130 <b>void (*pcre_stack_free)(void *);</b>
131 </P>
132 <P>
133 <b>int (*pcre_callout)(pcre_callout_block *);</b>
134 </P>
135 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
136 <P>
137 PCRE has its own native API, which is described in this document. There is
138 also a set of wrapper functions that correspond to the POSIX regular expression
139 API. These are described in the
140 <a href="pcreposix.html"><b>pcreposix</b></a>
141 documentation. Both of these APIs define a set of C function calls. A C++
142 wrapper is distributed with PCRE. It is documented in the
143 <a href="pcrecpp.html"><b>pcrecpp</b></a>
144 page.
145 </P>
146 <P>
147 The native API C function prototypes are defined in the header file
148 <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
149 It can normally be accessed by adding <b>-lpcre</b> to the command for linking
150 an application that uses PCRE. The header file defines the macros PCRE_MAJOR
151 and PCRE_MINOR to contain the major and minor release numbers for the library.
152 Applications can use these to include support for different releases of PCRE.
153 </P>
154 <P>
155 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
156 and <b>pcre_exec()</b> are used for compiling and matching regular expressions
157 in a Perl-compatible manner. A sample program that demonstrates the simplest
158 way of using them is provided in the file called <i>pcredemo.c</i> in the source
159 distribution. The
160 <a href="pcresample.html"><b>pcresample</b></a>
161 documentation describes how to run it.
162 </P>
163 <P>
164 A second matching function, <b>pcre_dfa_exec()</b>, which is not
165 Perl-compatible, is also provided. This uses a different algorithm for the
166 matching. This allows it to find all possible matches (at a given point in the
167 subject), not just one. However, this algorithm does not return captured
168 substrings. A description of the two matching algorithms and their advantages
169 and disadvantages is given in the
170 <a href="pcrematching.html"><b>pcrematching</b></a>
171 documentation.
172 </P>
173 <P>
174 In addition to the main compiling and matching functions, there are convenience
175 functions for extracting captured substrings from a subject string that is
176 matched by <b>pcre_exec()</b>. They are:
177 <pre>
178 <b>pcre_copy_substring()</b>
179 <b>pcre_copy_named_substring()</b>
180 <b>pcre_get_substring()</b>
181 <b>pcre_get_named_substring()</b>
182 <b>pcre_get_substring_list()</b>
183 <b>pcre_get_stringnumber()</b>
184 </pre>
185 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
186 provided, to free the memory used for extracted strings.
187 </P>
188 <P>
189 The function <b>pcre_maketables()</b> is used to build a set of character tables
190 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
191 or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
192 specialist use. Most commonly, no special tables are passed, in which case
193 internal tables that are generated when PCRE is built are used.
194 </P>
195 <P>
196 The function <b>pcre_fullinfo()</b> is used to find out information about a
197 compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
198 some of the available information, but is retained for backwards compatibility.
199 The function <b>pcre_version()</b> returns a pointer to a string containing the
200 version of PCRE and its date of release.
201 </P>
202 <P>
203 The function <b>pcre_refcount()</b> maintains a reference count in a data block
204 containing a compiled pattern. This is provided for the benefit of
205 object-oriented applications.
206 </P>
207 <P>
208 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
209 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
210 respectively. PCRE calls the memory management functions via these variables,
211 so a calling program can replace them if it wishes to intercept the calls. This
212 should be done before calling any PCRE functions.
213 </P>
214 <P>
215 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
216 indirections to memory management functions. These special functions are used
217 only when PCRE is compiled to use the heap for remembering data, instead of
218 recursive function calls, when running the <b>pcre_exec()</b> function. This is
219 a non-standard way of building PCRE, for use in environments that have limited
220 stacks. Because of the greater use of memory management, it runs more slowly.
221 Separate functions are provided so that special-purpose external code can be
222 used for this case. When used, these functions are always called in a
223 stack-like manner (last obtained, first freed), and always for memory blocks of
224 the same size.
225 </P>
226 <P>
227 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
228 by the caller to a "callout" function, which PCRE will then call at specified
229 points during a matching operation. Details are given in the
230 <a href="pcrecallout.html"><b>pcrecallout</b></a>
231 documentation.
232 </P>
233 <br><a name="SEC3" href="#TOC1">MULTITHREADING</a><br>
234 <P>
235 The PCRE functions can be used in multi-threading applications, with the
236 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
237 <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
238 callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
239 </P>
240 <P>
241 The compiled form of a regular expression is not altered during matching, so
242 the same compiled pattern can safely be used by several threads at once.
243 </P>
244 <br><a name="SEC4" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
245 <P>
246 The compiled form of a regular expression can be saved and re-used at a later
247 time, possibly by a different program, and even on a host other than the one on
248 which it was compiled. Details are given in the
249 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
250 documentation.
251 </P>
252 <br><a name="SEC5" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
253 <P>
254 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
255 </P>
256 <P>
257 The function <b>pcre_config()</b> makes it possible for a PCRE client to
258 discover which optional features have been compiled into the PCRE library. The
259 <a href="pcrebuild.html"><b>pcrebuild</b></a>
260 documentation has more details about these optional features.
261 </P>
262 <P>
263 The first argument for <b>pcre_config()</b> is an integer, specifying which
264 information is required; the second argument is a pointer to a variable into
265 which the information is placed. The following information is available:
266 <pre>
267 PCRE_CONFIG_UTF8
268 </pre>
269 The output is an integer that is set to one if UTF-8 support is available;
270 otherwise it is set to zero.
271 <pre>
272 PCRE_CONFIG_UNICODE_PROPERTIES
273 </pre>
274 The output is an integer that is set to one if support for Unicode character
275 properties is available; otherwise it is set to zero.
276 <pre>
277 PCRE_CONFIG_NEWLINE
278 </pre>
279 The output is an integer that is set to the value of the code that is used for
280 the newline character. It is either linefeed (10) or carriage return (13), and
281 should normally be the standard character for your operating system.
282 <pre>
283 PCRE_CONFIG_LINK_SIZE
284 </pre>
285 The output is an integer that contains the number of bytes used for internal
286 linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
287 allow larger regular expressions to be compiled, at the expense of slower
288 matching. The default value of 2 is sufficient for all but the most massive
289 patterns, since it allows the compiled pattern to be up to 64K in size.
290 <pre>
291 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
292 </pre>
293 The output is an integer that contains the threshold above which the POSIX
294 interface uses <b>malloc()</b> for output vectors. Further details are given in
295 the
296 <a href="pcreposix.html"><b>pcreposix</b></a>
297 documentation.
298 <pre>
299 PCRE_CONFIG_MATCH_LIMIT
300 </pre>
301 The output is an integer that gives the default limit for the number of
302 internal matching function calls in a <b>pcre_exec()</b> execution. Further
303 details are given with <b>pcre_exec()</b> below.
304 <pre>
305 PCRE_CONFIG_MATCH_LIMIT_RECURSION
306 </pre>
307 The output is an integer that gives the default limit for the depth of
308 recursion when calling the internal matching function in a <b>pcre_exec()</b>
309 execution. Further details are given with <b>pcre_exec()</b> below.
310 <pre>
311 PCRE_CONFIG_STACKRECURSE
312 </pre>
313 The output is an integer that is set to one if internal recursion when running
314 <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
315 to remember their state. This is the usual way that PCRE is compiled. The
316 output is zero if PCRE was compiled to use blocks of data on the heap instead
317 of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
318 <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
319 avoiding the use of the stack.
320 </P>
321 <br><a name="SEC6" href="#TOC1">COMPILING A PATTERN</a><br>
322 <P>
323 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
324 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
325 <b>const unsigned char *<i>tableptr</i>);</b>
326 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
327 <b>int *<i>errorcodeptr</i>,</b>
328 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
329 <b>const unsigned char *<i>tableptr</i>);</b>
330 </P>
331 <P>
332 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
333 called to compile a pattern into an internal form. The only difference between
334 the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
335 <i>errorcodeptr</i>, via which a numerical error code can be returned.
336 </P>
337 <P>
338 The pattern is a C string terminated by a binary zero, and is passed in the
339 <i>pattern</i> argument. A pointer to a single block of memory that is obtained
340 via <b>pcre_malloc</b> is returned. This contains the compiled code and related
341 data. The <b>pcre</b> type is defined for the returned block; this is a typedef
342 for a structure whose contents are not externally defined. It is up to the
343 caller to free the memory when it is no longer required.
344 </P>
345 <P>
346 Although the compiled code of a PCRE regex is relocatable, that is, it does not
347 depend on memory location, the complete <b>pcre</b> data block is not
348 fully relocatable, because it may contain a copy of the <i>tableptr</i>
349 argument, which is an address (see below).
350 </P>
351 <P>
352 The <i>options</i> argument contains independent bits that affect the
353 compilation. It should be zero if no options are required. The available
354 options are described below. Some of them, in particular, those that are
355 compatible with Perl, can also be set and unset from within the pattern (see
356 the detailed description in the
357 <a href="pcrepattern.html"><b>pcrepattern</b></a>
358 documentation). For these options, the contents of the <i>options</i> argument
359 specifies their initial settings at the start of compilation and execution. The
360 PCRE_ANCHORED option can be set at the time of matching as well as at compile
361 time.
362 </P>
363 <P>
364 If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
365 Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
366 NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
367 error message. This is a static string that is part of the library. You must
368 not try to free it. The offset from the start of the pattern to the character
369 where the error was discovered is placed in the variable pointed to by
370 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
371 </P>
372 <P>
373 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
374 <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
375 returned via this argument in the event of an error. This is in addition to the
376 textual error message. Error codes and messages are listed below.
377 </P>
378 <P>
379 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
380 character tables that are built when PCRE is compiled, using the default C
381 locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
382 call to <b>pcre_maketables()</b>. This value is stored with the compiled
383 pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
384 passed to it. For more discussion, see the section on locale support below.
385 </P>
386 <P>
387 This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
388 <pre>
389 pcre *re;
390 const char *error;
391 int erroffset;
392 re = pcre_compile(
393 "^A.*Z", /* the pattern */
394 0, /* default options */
395 &error, /* for error message */
396 &erroffset, /* for error offset */
397 NULL); /* use default character tables */
398 </pre>
399 The following names for option bits are defined in the <b>pcre.h</b> header
400 file:
401 <pre>
402 PCRE_ANCHORED
403 </pre>
404 If this bit is set, the pattern is forced to be "anchored", that is, it is
405 constrained to match only at the first matching point in the string that is
406 being searched (the "subject string"). This effect can also be achieved by
407 appropriate constructs in the pattern itself, which is the only way to do it in
408 Perl.
409 <pre>
410 PCRE_AUTO_CALLOUT
411 </pre>
412 If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
413 all with number 255, before each pattern item. For discussion of the callout
414 facility, see the
415 <a href="pcrecallout.html"><b>pcrecallout</b></a>
416 documentation.
417 <pre>
418 PCRE_CASELESS
419 </pre>
420 If this bit is set, letters in the pattern match both upper and lower case
421 letters. It is equivalent to Perl's /i option, and it can be changed within a
422 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
423 concept of case for characters whose values are less than 128, so caseless
424 matching is always possible. For characters with higher values, the concept of
425 case is supported if PCRE is compiled with Unicode property support, but not
426 otherwise. If you want to use caseless matching for characters 128 and above,
427 you must ensure that PCRE is compiled with Unicode property support as well as
428 with UTF-8 support.
429 <pre>
430 PCRE_DOLLAR_ENDONLY
431 </pre>
432 If this bit is set, a dollar metacharacter in the pattern matches only at the
433 end of the subject string. Without this option, a dollar also matches
434 immediately before the final character if it is a newline (but not before any
435 other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is
436 set. There is no equivalent to this option in Perl, and no way to set it within
437 a pattern.
438 <pre>
439 PCRE_DOTALL
440 </pre>
441 If this bit is set, a dot metacharater in the pattern matches all characters,
442 including newlines. Without it, newlines are excluded. This option is
443 equivalent to Perl's /s option, and it can be changed within a pattern by a
444 (?s) option setting. A negative class such as [^a] always matches a newline
445 character, independent of the setting of this option.
446 <pre>
447 PCRE_EXTENDED
448 </pre>
449 If this bit is set, whitespace data characters in the pattern are totally
450 ignored except when escaped or inside a character class. Whitespace does not
451 include the VT character (code 11). In addition, characters between an
452 unescaped # outside a character class and the next newline character,
453 inclusive, are also ignored. This is equivalent to Perl's /x option, and it can
454 be changed within a pattern by a (?x) option setting.
455 </P>
456 <P>
457 This option makes it possible to include comments inside complicated patterns.
458 Note, however, that this applies only to data characters. Whitespace characters
459 may never appear within special character sequences in a pattern, for example
460 within the sequence (?( which introduces a conditional subpattern.
461 <pre>
462 PCRE_EXTRA
463 </pre>
464 This option was invented in order to turn on additional functionality of PCRE
465 that is incompatible with Perl, but it is currently of very little use. When
466 set, any backslash in a pattern that is followed by a letter that has no
467 special meaning causes an error, thus reserving these combinations for future
468 expansion. By default, as in Perl, a backslash followed by a letter with no
469 special meaning is treated as a literal. There are at present no other features
470 controlled by this option. It can also be set by a (?X) option setting within a
471 pattern.
472 <pre>
473 PCRE_FIRSTLINE
474 </pre>
475 If this option is set, an unanchored pattern is required to match before or at
476 the first newline character in the subject string, though the matched text may
477 continue over the newline.
478 <pre>
479 PCRE_MULTILINE
480 </pre>
481 By default, PCRE treats the subject string as consisting of a single line of
482 characters (even if it actually contains newlines). The "start of line"
483 metacharacter (^) matches only at the start of the string, while the "end of
484 line" metacharacter ($) matches only at the end of the string, or before a
485 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
486 Perl.
487 </P>
488 <P>
489 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
490 match immediately following or immediately before any newline in the subject
491 string, respectively, as well as at the very start and end. This is equivalent
492 to Perl's /m option, and it can be changed within a pattern by a (?m) option
493 setting. If there are no "\n" characters in a subject string, or no
494 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
495 <pre>
496 PCRE_NO_AUTO_CAPTURE
497 </pre>
498 If this option is set, it disables the use of numbered capturing parentheses in
499 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
500 were followed by ?: but named parentheses can still be used for capturing (and
501 they acquire numbers in the usual way). There is no equivalent of this option
502 in Perl.
503 <pre>
504 PCRE_UNGREEDY
505 </pre>
506 This option inverts the "greediness" of the quantifiers so that they are not
507 greedy by default, but become greedy if followed by "?". It is not compatible
508 with Perl. It can also be set by a (?U) option setting within the pattern.
509 <pre>
510 PCRE_UTF8
511 </pre>
512 This option causes PCRE to regard both the pattern and the subject as strings
513 of UTF-8 characters instead of single-byte character strings. However, it is
514 available only when PCRE is built to include UTF-8 support. If not, the use
515 of this option provokes an error. Details of how this option changes the
516 behaviour of PCRE are given in the
517 <a href="pcre.html#utf8support">section on UTF-8 support</a>
518 in the main
519 <a href="pcre.html"><b>pcre</b></a>
520 page.
521 <pre>
522 PCRE_NO_UTF8_CHECK
523 </pre>
524 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
525 automatically checked. If an invalid UTF-8 sequence of bytes is found,
526 <b>pcre_compile()</b> returns an error. If you already know that your pattern is
527 valid, and you want to skip this check for performance reasons, you can set the
528 PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
529 UTF-8 string as a pattern is undefined. It may cause your program to crash.
530 Note that this option can also be passed to <b>pcre_exec()</b> and
531 <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
532 strings.
533 </P>
534 <br><a name="SEC7" href="#TOC1">COMPILATION ERROR CODES</a><br>
535 <P>
536 The following table lists the error codes than may be returned by
537 <b>pcre_compile2()</b>, along with the error messages that may be returned by
538 both compiling functions.
539 <pre>
540 0 no error
541 1 \ at end of pattern
542 2 \c at end of pattern
543 3 unrecognized character follows \
544 4 numbers out of order in {} quantifier
545 5 number too big in {} quantifier
546 6 missing terminating ] for character class
547 7 invalid escape sequence in character class
548 8 range out of order in character class
549 9 nothing to repeat
550 10 operand of unlimited repeat could match the empty string
551 11 internal error: unexpected repeat
552 12 unrecognized character after (?
553 13 POSIX named classes are supported only within a class
554 14 missing )
555 15 reference to non-existent subpattern
556 16 erroffset passed as NULL
557 17 unknown option bit(s) set
558 18 missing ) after comment
559 19 parentheses nested too deeply
560 20 regular expression too large
561 21 failed to get memory
562 22 unmatched parentheses
563 23 internal error: code overflow
564 24 unrecognized character after (?&#60;
565 25 lookbehind assertion is not fixed length
566 26 malformed number after (?(
567 27 conditional group contains more than two branches
568 28 assertion expected after (?(
569 29 (?R or (?digits must be followed by )
570 30 unknown POSIX class name
571 31 POSIX collating elements are not supported
572 32 this version of PCRE is not compiled with PCRE_UTF8 support
573 33 spare error
574 34 character value in \x{...} sequence is too large
575 35 invalid condition (?(0)
576 36 \C not allowed in lookbehind assertion
577 37 PCRE does not support \L, \l, \N, \U, or \u
578 38 number after (?C is &#62; 255
579 39 closing ) for (?C expected
580 40 recursive call could loop indefinitely
581 41 unrecognized character after (?P
582 42 syntax error after (?P
583 43 two named groups have the same name
584 44 invalid UTF-8 string
585 45 support for \P, \p, and \X has not been compiled
586 46 malformed \P or \p sequence
587 47 unknown property name after \P or \p
588 </PRE>
589 </P>
590 <br><a name="SEC8" href="#TOC1">STUDYING A PATTERN</a><br>
591 <P>
592 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
593 <b>const char **<i>errptr</i>);</b>
594 </P>
595 <P>
596 If a compiled pattern is going to be used several times, it is worth spending
597 more time analyzing it in order to speed up the time taken for matching. The
598 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
599 argument. If studying the pattern produces additional information that will
600 help speed up matching, <b>pcre_study()</b> returns a pointer to a
601 <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
602 results of the study.
603 </P>
604 <P>
605 The returned value from <b>pcre_study()</b> can be passed directly to
606 <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
607 fields that can be set by the caller before the block is passed; these are
608 described
609 <a href="#extradata">below</a>
610 in the section on matching a pattern.
611 </P>
612 <P>
613 If studying the pattern does not produce any additional information
614 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
615 wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
616 own <b>pcre_extra</b> block.
617 </P>
618 <P>
619 The second argument of <b>pcre_study()</b> contains option bits. At present, no
620 options are defined, and this argument should always be zero.
621 </P>
622 <P>
623 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
624 studying succeeds (even if no data is returned), the variable it points to is
625 set to NULL. Otherwise it is set to point to a textual error message. This is a
626 static string that is part of the library. You must not try to free it. You
627 should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
628 sure that it has run successfully.
629 </P>
630 <P>
631 This is a typical call to <b>pcre_study</b>():
632 <pre>
633 pcre_extra *pe;
634 pe = pcre_study(
635 re, /* result of pcre_compile() */
636 0, /* no options exist */
637 &error); /* set to NULL or points to a message */
638 </pre>
639 At present, studying a pattern is useful only for non-anchored patterns that do
640 not have a single fixed starting character. A bitmap of possible starting
641 bytes is created.
642 <a name="localesupport"></a></P>
643 <br><a name="SEC9" href="#TOC1">LOCALE SUPPORT</a><br>
644 <P>
645 PCRE handles caseless matching, and determines whether characters are letters
646 digits, or whatever, by reference to a set of tables, indexed by character
647 value. When running in UTF-8 mode, this applies only to characters with codes
648 less than 128. Higher-valued codes never match escapes such as \w or \d, but
649 can be tested with \p if PCRE is built with Unicode character property
650 support. The use of locales with Unicode is discouraged.
651 </P>
652 <P>
653 An internal set of tables is created in the default C locale when PCRE is
654 built. This is used when the final argument of <b>pcre_compile()</b> is NULL,
655 and is sufficient for many applications. An alternative set of tables can,
656 however, be supplied. These may be created in a different locale from the
657 default. As more and more applications change to using Unicode, the need for
658 this locale support is expected to die away.
659 </P>
660 <P>
661 External tables are built by calling the <b>pcre_maketables()</b> function,
662 which has no arguments, in the relevant locale. The result can then be passed
663 to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
664 example, to build and use tables that are appropriate for the French locale
665 (where accented characters with values greater than 128 are treated as letters),
666 the following code could be used:
667 <pre>
668 setlocale(LC_CTYPE, "fr_FR");
669 tables = pcre_maketables();
670 re = pcre_compile(..., tables);
671 </pre>
672 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
673 obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
674 that the memory containing the tables remains available for as long as it is
675 needed.
676 </P>
677 <P>
678 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
679 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
680 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
681 pattern, compilation, studying and matching all happen in the same locale, but
682 different patterns can be compiled in different locales.
683 </P>
684 <P>
685 It is possible to pass a table pointer or NULL (indicating the use of the
686 internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
687 this facility could be used to match a pattern in a different locale from the
688 one in which it was compiled. Passing table pointers at run time is discussed
689 below in the section on matching a pattern.
690 </P>
691 <br><a name="SEC10" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
692 <P>
693 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
694 <b>int <i>what</i>, void *<i>where</i>);</b>
695 </P>
696 <P>
697 The <b>pcre_fullinfo()</b> function returns information about a compiled
698 pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
699 nevertheless retained for backwards compability (and is documented below).
700 </P>
701 <P>
702 The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
703 pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
704 the pattern was not studied. The third argument specifies which piece of
705 information is required, and the fourth argument is a pointer to a variable
706 to receive the data. The yield of the function is zero for success, or one of
707 the following negative numbers:
708 <pre>
709 PCRE_ERROR_NULL the argument <i>code</i> was NULL
710 the argument <i>where</i> was NULL
711 PCRE_ERROR_BADMAGIC the "magic number" was not found
712 PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
713 </pre>
714 The "magic number" is placed at the start of each compiled pattern as an simple
715 check against passing an arbitrary memory pointer. Here is a typical call of
716 <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
717 <pre>
718 int rc;
719 unsigned long int length;
720 rc = pcre_fullinfo(
721 re, /* result of pcre_compile() */
722 pe, /* result of pcre_study(), or NULL */
723 PCRE_INFO_SIZE, /* what is required */
724 &length); /* where to put the data */
725 </pre>
726 The possible values for the third argument are defined in <b>pcre.h</b>, and are
727 as follows:
728 <pre>
729 PCRE_INFO_BACKREFMAX
730 </pre>
731 Return the number of the highest back reference in the pattern. The fourth
732 argument should point to an <b>int</b> variable. Zero is returned if there are
733 no back references.
734 <pre>
735 PCRE_INFO_CAPTURECOUNT
736 </pre>
737 Return the number of capturing subpatterns in the pattern. The fourth argument
738 should point to an <b>int</b> variable.
739 <pre>
740 PCRE_INFO_DEFAULT_TABLES
741 </pre>
742 Return a pointer to the internal default character tables within PCRE. The
743 fourth argument should point to an <b>unsigned char *</b> variable. This
744 information call is provided for internal use by the <b>pcre_study()</b>
745 function. External callers can cause PCRE to use its internal tables by passing
746 a NULL table pointer.
747 <pre>
748 PCRE_INFO_FIRSTBYTE
749 </pre>
750 Return information about the first byte of any matched string, for a
751 non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the
752 old name is still recognized for backwards compatibility.)
753 </P>
754 <P>
755 If there is a fixed first byte, for example, from a pattern such as
756 (cat|cow|coyote), it is returned in the integer pointed to by <i>where</i>.
757 Otherwise, if either
758 <br>
759 <br>
760 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
761 starts with "^", or
762 <br>
763 <br>
764 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
765 (if it were set, the pattern would be anchored),
766 <br>
767 <br>
768 -1 is returned, indicating that the pattern matches only at the start of a
769 subject string or after any newline within the string. Otherwise -2 is
770 returned. For anchored patterns, -2 is returned.
771 <pre>
772 PCRE_INFO_FIRSTTABLE
773 </pre>
774 If the pattern was studied, and this resulted in the construction of a 256-bit
775 table indicating a fixed set of bytes for the first byte in any matching
776 string, a pointer to the table is returned. Otherwise NULL is returned. The
777 fourth argument should point to an <b>unsigned char *</b> variable.
778 <pre>
779 PCRE_INFO_LASTLITERAL
780 </pre>
781 Return the value of the rightmost literal byte that must exist in any matched
782 string, other than at its start, if such a byte has been recorded. The fourth
783 argument should point to an <b>int</b> variable. If there is no such byte, -1 is
784 returned. For anchored patterns, a last literal byte is recorded only if it
785 follows something of variable length. For example, for the pattern
786 /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
787 is -1.
788 <pre>
789 PCRE_INFO_NAMECOUNT
790 PCRE_INFO_NAMEENTRYSIZE
791 PCRE_INFO_NAMETABLE
792 </pre>
793 PCRE supports the use of named as well as numbered capturing parentheses. The
794 names are just an additional way of identifying the parentheses, which still
795 acquire numbers. A convenience function called <b>pcre_get_named_substring()</b>
796 is provided for extracting an individual captured substring by name. It is also
797 possible to extract the data directly, by first converting the name to a number
798 in order to access the correct pointers in the output vector (described with
799 <b>pcre_exec()</b> below). To do the conversion, you need to use the
800 name-to-number map, which is described by these three values.
801 </P>
802 <P>
803 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
804 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
805 entry; both of these return an <b>int</b> value. The entry size depends on the
806 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
807 entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
808 are the number of the capturing parenthesis, most significant byte first. The
809 rest of the entry is the corresponding name, zero terminated. The names are in
810 alphabetical order. For example, consider the following pattern (assume
811 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
812 <pre>
813 (?P&#60;date&#62; (?P&#60;year&#62;(\d\d)?\d\d) - (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )
814 </pre>
815 There are four named subpatterns, so the table has four entries, and each entry
816 in the table is eight bytes long. The table is as follows, with non-printing
817 bytes shows in hexadecimal, and undefined bytes shown as ??:
818 <pre>
819 00 01 d a t e 00 ??
820 00 05 d a y 00 ?? ??
821 00 04 m o n t h 00
822 00 02 y e a r 00 ??
823 </pre>
824 When writing code to extract data from named subpatterns using the
825 name-to-number map, remember that the length of each entry is likely to be
826 different for each compiled pattern.
827 <pre>
828 PCRE_INFO_OPTIONS
829 </pre>
830 Return a copy of the options with which the pattern was compiled. The fourth
831 argument should point to an <b>unsigned long int</b> variable. These option bits
832 are those specified in the call to <b>pcre_compile()</b>, modified by any
833 top-level option settings within the pattern itself.
834 </P>
835 <P>
836 A pattern is automatically anchored by PCRE if all of its top-level
837 alternatives begin with one of the following:
838 <pre>
839 ^ unless PCRE_MULTILINE is set
840 \A always
841 \G always
842 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
843 </pre>
844 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
845 <b>pcre_fullinfo()</b>.
846 <pre>
847 PCRE_INFO_SIZE
848 </pre>
849 Return the size of the compiled pattern, that is, the value that was passed as
850 the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
851 place the compiled data. The fourth argument should point to a <b>size_t</b>
852 variable.
853 <pre>
854 PCRE_INFO_STUDYSIZE
855 </pre>
856 Return the size of the data block pointed to by the <i>study_data</i> field in
857 a <b>pcre_extra</b> block. That is, it is the value that was passed to
858 <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
859 created by <b>pcre_study()</b>. The fourth argument should point to a
860 <b>size_t</b> variable.
861 </P>
862 <br><a name="SEC11" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
863 <P>
864 <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
865 <b>*<i>firstcharptr</i>);</b>
866 </P>
867 <P>
868 The <b>pcre_info()</b> function is now obsolete because its interface is too
869 restrictive to return all the available data about a compiled pattern. New
870 programs should use <b>pcre_fullinfo()</b> instead. The yield of
871 <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
872 following negative numbers:
873 <pre>
874 PCRE_ERROR_NULL the argument <i>code</i> was NULL
875 PCRE_ERROR_BADMAGIC the "magic number" was not found
876 </pre>
877 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
878 pattern was compiled is placed in the integer it points to (see
879 PCRE_INFO_OPTIONS above).
880 </P>
881 <P>
882 If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
883 it is used to pass back information about the first character of any matched
884 string (see PCRE_INFO_FIRSTBYTE above).
885 </P>
886 <br><a name="SEC12" href="#TOC1">REFERENCE COUNTS</a><br>
887 <P>
888 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
889 </P>
890 <P>
891 The <b>pcre_refcount()</b> function is used to maintain a reference count in the
892 data block that contains a compiled pattern. It is provided for the benefit of
893 applications that operate in an object-oriented manner, where different parts
894 of the application may be using the same compiled pattern, but you want to free
895 the block when they are all done.
896 </P>
897 <P>
898 When a pattern is compiled, the reference count field is initialized to zero.
899 It is changed only by calling this function, whose action is to add the
900 <i>adjust</i> value (which may be positive or negative) to it. The yield of the
901 function is the new value. However, the value of the count is constrained to
902 lie between 0 and 65535, inclusive. If the new value is outside these limits,
903 it is forced to the appropriate limit value.
904 </P>
905 <P>
906 Except when it is zero, the reference count is not correctly preserved if a
907 pattern is compiled on one host and then transferred to a host whose byte-order
908 is different. (This seems a highly unlikely scenario.)
909 </P>
910 <br><a name="SEC13" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
911 <P>
912 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
913 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
914 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
915 </P>
916 <P>
917 The function <b>pcre_exec()</b> is called to match a subject string against a
918 compiled pattern, which is passed in the <i>code</i> argument. If the
919 pattern has been studied, the result of the study should be passed in the
920 <i>extra</i> argument. This function is the main matching facility of the
921 library, and it operates in a Perl-like manner. For specialist use there is
922 also an alternative matching function, which is described
923 <a href="#dfamatch">below</a>
924 in the section about the <b>pcre_dfa_exec()</b> function.
925 </P>
926 <P>
927 In most applications, the pattern will have been compiled (and optionally
928 studied) in the same process that calls <b>pcre_exec()</b>. However, it is
929 possible to save compiled patterns and study data, and then use them later
930 in different processes, possibly even on different hosts. For a discussion
931 about this, see the
932 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
933 documentation.
934 </P>
935 <P>
936 Here is an example of a simple call to <b>pcre_exec()</b>:
937 <pre>
938 int rc;
939 int ovector[30];
940 rc = pcre_exec(
941 re, /* result of pcre_compile() */
942 NULL, /* we didn't study the pattern */
943 "some string", /* the subject string */
944 11, /* the length of the subject string */
945 0, /* start at offset 0 in the subject */
946 0, /* default options */
947 ovector, /* vector of integers for substring information */
948 30); /* number of elements (NOT size in bytes) */
949 <a name="extradata"></a></PRE>
950 </P>
951 <br><b>
952 Extra data for <b>pcre_exec()</b>
953 </b><br>
954 <P>
955 If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
956 data block. The <b>pcre_study()</b> function returns such a block (when it
957 doesn't return NULL), but you can also create one for yourself, and pass
958 additional information in it. The <b>pcre_extra</b> block contains the following
959 fields (not necessarily in this order):
960 <pre>
961 unsigned long int <i>flags</i>;
962 void *<i>study_data</i>;
963 unsigned long int <i>match_limit</i>;
964 unsigned long int <i>match_limit_recursion</i>;
965 void *<i>callout_data</i>;
966 const unsigned char *<i>tables</i>;
967 </pre>
968 The <i>flags</i> field is a bitmap that specifies which of the other fields
969 are set. The flag bits are:
970 <pre>
971 PCRE_EXTRA_STUDY_DATA
972 PCRE_EXTRA_MATCH_LIMIT
973 PCRE_EXTRA_MATCH_LIMIT_RECURSION
974 PCRE_EXTRA_CALLOUT_DATA
975 PCRE_EXTRA_TABLES
976 </pre>
977 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
978 <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
979 the appropriate flag bit. You should not set this yourself, but you may add to
980 the block by setting the other fields and their corresponding flag bits.
981 </P>
982 <P>
983 The <i>match_limit</i> field provides a means of preventing PCRE from using up a
984 vast amount of resources when running patterns that are not going to match,
985 but which have a very large number of possibilities in their search trees. The
986 classic example is the use of nested unlimited repeats.
987 </P>
988 <P>
989 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
990 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
991 number of times this function is called during a match, which has the effect of
992 limiting the amount of backtracking that can take place. For patterns that are
993 not anchored, the count restarts from zero for each position in the subject
994 string.
995 </P>
996 <P>
997 The default value for the limit can be set when PCRE is built; the default
998 default is 10 million, which handles all but the most extreme cases. You can
999 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1000 block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1001 the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1002 PCRE_ERROR_MATCHLIMIT.
1003 </P>
1004 <P>
1005 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1006 instead of limiting the total number of times that <b>match()</b> is called, it
1007 limits the depth of recursion. The recursion depth is a smaller number than the
1008 total number of calls, because not all calls to <b>match()</b> are recursive.
1009 This limit is of use only if it is set smaller than <i>match_limit</i>.
1010 </P>
1011 <P>
1012 Limiting the recursion depth limits the amount of stack that can be used, or,
1013 when PCRE has been compiled to use memory on the heap instead of the stack, the
1014 amount of heap memory that can be used.
1015 </P>
1016 <P>
1017 The default value for <i>match_limit_recursion</i> can be set when PCRE is
1018 built; the default default is the same value as the default for
1019 <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1020 with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1021 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1022 is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1023 </P>
1024 <P>
1025 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1026 which is described in the
1027 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1028 documentation.
1029 </P>
1030 <P>
1031 The <i>tables</i> field is used to pass a character tables pointer to
1032 <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1033 pattern. A non-NULL value is stored with the compiled pattern only if custom
1034 tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1035 If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1036 internal tables to be used. This facility is helpful when re-using patterns
1037 that have been saved after compiling with an external set of tables, because
1038 the external tables might be at a different address when <b>pcre_exec()</b> is
1039 called. See the
1040 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1041 documentation for a discussion of saving compiled patterns for later use.
1042 </P>
1043 <br><b>
1044 Option bits for <b>pcre_exec()</b>
1045 </b><br>
1046 <P>
1047 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1048 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,
1049 PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1050 <pre>
1051 PCRE_ANCHORED
1052 </pre>
1053 The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1054 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1055 to be anchored by virtue of its contents, it cannot be made unachored at
1056 matching time.
1057 <pre>
1058 PCRE_NOTBOL
1059 </pre>
1060 This option specifies that first character of the subject string is not the
1061 beginning of a line, so the circumflex metacharacter should not match before
1062 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1063 never to match. This option affects only the behaviour of the circumflex
1064 metacharacter. It does not affect \A.
1065 <pre>
1066 PCRE_NOTEOL
1067 </pre>
1068 This option specifies that the end of the subject string is not the end of a
1069 line, so the dollar metacharacter should not match it nor (except in multiline
1070 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1071 compile time) causes dollar never to match. This option affects only the
1072 behaviour of the dollar metacharacter. It does not affect \Z or \z.
1073 <pre>
1074 PCRE_NOTEMPTY
1075 </pre>
1076 An empty string is not considered to be a valid match if this option is set. If
1077 there are alternatives in the pattern, they are tried. If all the alternatives
1078 match the empty string, the entire match fails. For example, if the pattern
1079 <pre>
1080 a?b?
1081 </pre>
1082 is applied to a string not beginning with "a" or "b", it matches the empty
1083 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1084 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1085 </P>
1086 <P>
1087 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1088 of a pattern match of the empty string within its <b>split()</b> function, and
1089 when using the /g modifier. It is possible to emulate Perl's behaviour after
1090 matching a null string by first trying the match again at the same offset with
1091 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1092 starting offset (see below) and trying an ordinary match again. There is some
1093 code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1094 <pre>
1095 PCRE_NO_UTF8_CHECK
1096 </pre>
1097 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1098 string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1099 The value of <i>startoffset</i> is also checked to ensure that it points to the
1100 start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1101 <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1102 contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1103 </P>
1104 <P>
1105 If you already know that your subject is valid, and you want to skip these
1106 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1107 calling <b>pcre_exec()</b>. You might want to do this for the second and
1108 subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1109 all the matches in a single subject string. However, you should be sure that
1110 the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1111 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1112 subject, or a value of <i>startoffset</i> that does not point to the start of a
1113 UTF-8 character, is undefined. Your program may crash.
1114 <pre>
1115 PCRE_PARTIAL
1116 </pre>
1117 This option turns on the partial matching feature. If the subject string fails
1118 to match the pattern, but at some point during the matching process the end of
1119 the subject was reached (that is, the subject partially matches the pattern and
1120 the failure to match occurred only because there were not enough subject
1121 characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1122 PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1123 may appear in the pattern. These are discussed in the
1124 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1125 documentation.
1126 </P>
1127 <br><b>
1128 The string to be matched by <b>pcre_exec()</b>
1129 </b><br>
1130 <P>
1131 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1132 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1133 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1134 UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1135 bytes. When the starting offset is zero, the search for a match starts at the
1136 beginning of the subject, and this is by far the most common case.
1137 </P>
1138 <P>
1139 A non-zero starting offset is useful when searching for another match in the
1140 same subject by calling <b>pcre_exec()</b> again after a previous success.
1141 Setting <i>startoffset</i> differs from just passing over a shortened string and
1142 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1143 lookbehind. For example, consider the pattern
1144 <pre>
1145 \Biss\B
1146 </pre>
1147 which finds occurrences of "iss" in the middle of words. (\B matches only if
1148 the current position in the subject is not a word boundary.) When applied to
1149 the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1150 occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1151 subject, namely "issipi", it does not match, because \B is always false at the
1152 start of the subject, which is deemed to be a word boundary. However, if
1153 <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1154 set to 4, it finds the second occurrence of "iss" because it is able to look
1155 behind the starting point to discover that it is preceded by a letter.
1156 </P>
1157 <P>
1158 If a non-zero starting offset is passed when the pattern is anchored, one
1159 attempt to match at the given offset is made. This can only succeed if the
1160 pattern does not require the match to be at the start of the subject.
1161 </P>
1162 <br><b>
1163 How <b>pcre_exec()</b> returns captured substrings
1164 </b><br>
1165 <P>
1166 In general, a pattern matches a certain portion of the subject, and in
1167 addition, further substrings from the subject may be picked out by parts of the
1168 pattern. Following the usage in Jeffrey Friedl's book, this is called
1169 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1170 a fragment of a pattern that picks out a substring. PCRE supports several other
1171 kinds of parenthesized subpattern that do not cause substrings to be captured.
1172 </P>
1173 <P>
1174 Captured substrings are returned to the caller via a vector of integer offsets
1175 whose address is passed in <i>ovector</i>. The number of elements in the vector
1176 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1177 this argument is NOT the size of <i>ovector</i> in bytes.
1178 </P>
1179 <P>
1180 The first two-thirds of the vector is used to pass back captured substrings,
1181 each substring using a pair of integers. The remaining third of the vector is
1182 used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1183 and is not available for passing back information. The length passed in
1184 <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1185 rounded down.
1186 </P>
1187 <P>
1188 When a match is successful, information about captured substrings is returned
1189 in pairs of integers, starting at the beginning of <i>ovector</i>, and
1190 continuing up to two-thirds of its length at the most. The first element of a
1191 pair is set to the offset of the first character in a substring, and the second
1192 is set to the offset of the first character after the end of a substring. The
1193 first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1194 subject string matched by the entire pattern. The next pair is used for the
1195 first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1196 is the number of pairs that have been set. If there are no capturing
1197 subpatterns, the return value from a successful match is 1, indicating that
1198 just the first pair of offsets has been set.
1199 </P>
1200 <P>
1201 Some convenience functions are provided for extracting the captured substrings
1202 as separate strings. These are described in the following section.
1203 </P>
1204 <P>
1205 It is possible for an capturing subpattern number <i>n+1</i> to match some
1206 part of the subject when subpattern <i>n</i> has not been used at all. For
1207 example, if the string "abc" is matched against the pattern (a|(z))(bc)
1208 subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset
1209 values corresponding to the unused subpattern are set to -1.
1210 </P>
1211 <P>
1212 If a capturing subpattern is matched repeatedly, it is the last portion of the
1213 string that it matched that is returned.
1214 </P>
1215 <P>
1216 If the vector is too small to hold all the captured substring offsets, it is
1217 used as far as possible (up to two-thirds of its length), and the function
1218 returns a value of zero. In particular, if the substring offsets are not of
1219 interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1220 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1221 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1222 has to get additional memory for use during matching. Thus it is usually
1223 advisable to supply an <i>ovector</i>.
1224 </P>
1225 <P>
1226 Note that <b>pcre_info()</b> can be used to find out how many capturing
1227 subpatterns there are in a compiled pattern. The smallest size for
1228 <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1229 the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1230 <a name="errorlist"></a></P>
1231 <br><b>
1232 Return values from <b>pcre_exec()</b>
1233 </b><br>
1234 <P>
1235 If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1236 defined in the header file:
1237 <pre>
1238 PCRE_ERROR_NOMATCH (-1)
1239 </pre>
1240 The subject string did not match the pattern.
1241 <pre>
1242 PCRE_ERROR_NULL (-2)
1243 </pre>
1244 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1245 NULL and <i>ovecsize</i> was not zero.
1246 <pre>
1247 PCRE_ERROR_BADOPTION (-3)
1248 </pre>
1249 An unrecognized bit was set in the <i>options</i> argument.
1250 <pre>
1251 PCRE_ERROR_BADMAGIC (-4)
1252 </pre>
1253 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1254 the case when it is passed a junk pointer and to detect when a pattern that was
1255 compiled in an environment of one endianness is run in an environment with the
1256 other endianness. This is the error that PCRE gives when the magic number is
1257 not present.
1258 <pre>
1259 PCRE_ERROR_UNKNOWN_NODE (-5)
1260 </pre>
1261 While running the pattern match, an unknown item was encountered in the
1262 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1263 of the compiled pattern.
1264 <pre>
1265 PCRE_ERROR_NOMEMORY (-6)
1266 </pre>
1267 If a pattern contains back references, but the <i>ovector</i> that is passed to
1268 <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1269 gets a block of memory at the start of matching to use for this purpose. If the
1270 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1271 automatically freed at the end of matching.
1272 <pre>
1273 PCRE_ERROR_NOSUBSTRING (-7)
1274 </pre>
1275 This error is used by the <b>pcre_copy_substring()</b>,
1276 <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1277 below). It is never returned by <b>pcre_exec()</b>.
1278 <pre>
1279 PCRE_ERROR_MATCHLIMIT (-8)
1280 </pre>
1281 The backtracking limit, as specified by the <i>match_limit</i> field in a
1282 <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1283 above.
1284 <pre>
1285 PCRE_ERROR_RECURSIONLIMIT (-21)
1286 </pre>
1287 The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1288 field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1289 description above.
1290 <pre>
1291 PCRE_ERROR_CALLOUT (-9)
1292 </pre>
1293 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1294 use by callout functions that want to yield a distinctive error code. See the
1295 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1296 documentation for details.
1297 <pre>
1298 PCRE_ERROR_BADUTF8 (-10)
1299 </pre>
1300 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1301 <pre>
1302 PCRE_ERROR_BADUTF8_OFFSET (-11)
1303 </pre>
1304 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1305 of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1306 <pre>
1307 PCRE_ERROR_PARTIAL (-12)
1308 </pre>
1309 The subject string did not match, but it did match partially. See the
1310 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1311 documentation for details of partial matching.
1312 <pre>
1313 PCRE_ERROR_BADPARTIAL (-13)
1314 </pre>
1315 The PCRE_PARTIAL option was used with a compiled pattern containing items that
1316 are not supported for partial matching. See the
1317 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1318 documentation for details of partial matching.
1319 <pre>
1320 PCRE_ERROR_INTERNAL (-14)
1321 </pre>
1322 An unexpected internal error has occurred. This error could be caused by a bug
1323 in PCRE or by overwriting of the compiled pattern.
1324 <pre>
1325 PCRE_ERROR_BADCOUNT (-15)
1326 </pre>
1327 This error is given if the value of the <i>ovecsize</i> argument is negative.
1328 </P>
1329 <br><a name="SEC14" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1330 <P>
1331 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1332 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1333 <b>int <i>buffersize</i>);</b>
1334 </P>
1335 <P>
1336 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1337 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1338 <b>const char **<i>stringptr</i>);</b>
1339 </P>
1340 <P>
1341 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1342 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1343 </P>
1344 <P>
1345 Captured substrings can be accessed directly by using the offsets returned by
1346 <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1347 <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1348 <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1349 as new, separate, zero-terminated strings. These functions identify substrings
1350 by number. The next section describes functions for extracting named
1351 substrings. A substring that contains a binary zero is correctly extracted and
1352 has a further zero added on the end, but the result is not, of course,
1353 a C string.
1354 </P>
1355 <P>
1356 The first three arguments are the same for all three of these functions:
1357 <i>subject</i> is the subject string that has just been successfully matched,
1358 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1359 <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1360 captured by the match, including the substring that matched the entire regular
1361 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1362 than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1363 space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1364 number of elements in the vector divided by three.
1365 </P>
1366 <P>
1367 The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1368 extract a single substring, whose number is given as <i>stringnumber</i>. A
1369 value of zero extracts the substring that matched the entire pattern, whereas
1370 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1371 the string is placed in <i>buffer</i>, whose length is given by
1372 <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1373 obtained via <b>pcre_malloc</b>, and its address is returned via
1374 <i>stringptr</i>. The yield of the function is the length of the string, not
1375 including the terminating zero, or one of
1376 <pre>
1377 PCRE_ERROR_NOMEMORY (-6)
1378 </pre>
1379 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1380 memory failed for <b>pcre_get_substring()</b>.
1381 <pre>
1382 PCRE_ERROR_NOSUBSTRING (-7)
1383 </pre>
1384 There is no substring whose number is <i>stringnumber</i>.
1385 </P>
1386 <P>
1387 The <b>pcre_get_substring_list()</b> function extracts all available substrings
1388 and builds a list of pointers to them. All this is done in a single block of
1389 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1390 is returned via <i>listptr</i>, which is also the start of the list of string
1391 pointers. The end of the list is marked by a NULL pointer. The yield of the
1392 function is zero if all went well, or
1393 <pre>
1394 PCRE_ERROR_NOMEMORY (-6)
1395 </pre>
1396 if the attempt to get the memory block failed.
1397 </P>
1398 <P>
1399 When any of these functions encounter a substring that is unset, which can
1400 happen when capturing subpattern number <i>n+1</i> matches some part of the
1401 subject, but subpattern <i>n</i> has not been used at all, they return an empty
1402 string. This can be distinguished from a genuine zero-length substring by
1403 inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1404 substrings.
1405 </P>
1406 <P>
1407 The two convenience functions <b>pcre_free_substring()</b> and
1408 <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1409 a previous call of <b>pcre_get_substring()</b> or
1410 <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1411 the function pointed to by <b>pcre_free</b>, which of course could be called
1412 directly from a C program. However, PCRE is used in some situations where it is
1413 linked via a special interface to another programming language which cannot use
1414 <b>pcre_free</b> directly; it is for these cases that the functions are
1415 provided.
1416 </P>
1417 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1418 <P>
1419 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1420 <b>const char *<i>name</i>);</b>
1421 </P>
1422 <P>
1423 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1424 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1425 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1426 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1427 </P>
1428 <P>
1429 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1430 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1431 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1432 <b>const char **<i>stringptr</i>);</b>
1433 </P>
1434 <P>
1435 To extract a substring by name, you first have to find associated number.
1436 For example, for this pattern
1437 <pre>
1438 (a+)b(?P&#60;xxx&#62;\d+)...
1439 </pre>
1440 the number of the subpattern called "xxx" is 2. You can find the number from
1441 the name by calling <b>pcre_get_stringnumber()</b>. The first argument is the
1442 compiled pattern, and the second is the name. The yield of the function is the
1443 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1444 that name.
1445 </P>
1446 <P>
1447 Given the number, you can extract the substring directly, or use one of the
1448 functions described in the previous section. For convenience, there are also
1449 two functions that do the whole job.
1450 </P>
1451 <P>
1452 Most of the arguments of <i>pcre_copy_named_substring()</i> and
1453 <i>pcre_get_named_substring()</i> are the same as those for the similarly named
1454 functions that extract by number. As these are described in the previous
1455 section, they are not re-described here. There are just two differences:
1456 </P>
1457 <P>
1458 First, instead of a substring number, a substring name is given. Second, there
1459 is an extra argument, given at the start, which is a pointer to the compiled
1460 pattern. This is needed in order to gain access to the name-to-number
1461 translation table.
1462 </P>
1463 <P>
1464 These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1465 then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as
1466 appropriate.
1467 </P>
1468 <br><a name="SEC16" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1469 <P>
1470 The traditional matching function uses a similar algorithm to Perl, which stops
1471 when it finds the first match, starting at a given point in the subject. If you
1472 want to find all possible matches, or the longest possible match, consider
1473 using the alternative matching function (see below) instead. If you cannot use
1474 the alternative function, but still need to find all possible matches, you
1475 can kludge it up by making use of the callout facility, which is described in
1476 the
1477 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1478 documentation.
1479 </P>
1480 <P>
1481 What you have to do is to insert a callout right at the end of the pattern.
1482 When your callout function is called, extract and save the current matched
1483 substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1484 other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1485 will yield PCRE_ERROR_NOMATCH.
1486 <a name="dfamatch"></a></P>
1487 <br><a name="SEC17" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1488 <P>
1489 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1490 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1491 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1492 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1493 </P>
1494 <P>
1495 The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1496 a compiled pattern, using a "DFA" matching algorithm. This has different
1497 characteristics to the normal algorithm, and is not compatible with Perl. Some
1498 of the features of PCRE patterns are not supported. Nevertheless, there are
1499 times when this kind of matching can be useful. For a discussion of the two
1500 matching algorithms, see the
1501 <a href="pcrematching.html"><b>pcrematching</b></a>
1502 documentation.
1503 </P>
1504 <P>
1505 The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1506 <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1507 different way, and this is described below. The other common arguments are used
1508 in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1509 here.
1510 </P>
1511 <P>
1512 The two additional arguments provide workspace for the function. The workspace
1513 vector should contain at least 20 elements. It is used for keeping track of
1514 multiple paths through the pattern tree. More workspace will be needed for
1515 patterns and subjects where there are a lot of possible matches.
1516 </P>
1517 <P>
1518 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1519 <pre>
1520 int rc;
1521 int ovector[10];
1522 int wspace[20];
1523 rc = pcre_dfa_exec(
1524 re, /* result of pcre_compile() */
1525 NULL, /* we didn't study the pattern */
1526 "some string", /* the subject string */
1527 11, /* the length of the subject string */
1528 0, /* start at offset 0 in the subject */
1529 0, /* default options */
1530 ovector, /* vector of integers for substring information */
1531 10, /* number of elements (NOT size in bytes) */
1532 wspace, /* working space vector */
1533 20); /* number of elements (NOT size in bytes) */
1534 </PRE>
1535 </P>
1536 <br><b>
1537 Option bits for <b>pcre_dfa_exec()</b>
1538 </b><br>
1539 <P>
1540 The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1541 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,
1542 PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1543 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1544 the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1545 <pre>
1546 PCRE_PARTIAL
1547 </pre>
1548 This has the same general effect as it does for <b>pcre_exec()</b>, but the
1549 details are slightly different. When PCRE_PARTIAL is set for
1550 <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1551 PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1552 complete matches, but there is still at least one matching possibility. The
1553 portion of the string that provided the partial match is set as the first
1554 matching string.
1555 <pre>
1556 PCRE_DFA_SHORTEST
1557 </pre>
1558 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1559 soon as it has found one match. Because of the way the DFA algorithm works,
1560 this is necessarily the shortest possible match at the first possible matching
1561 point in the subject string.
1562 <pre>
1563 PCRE_DFA_RESTART
1564 </pre>
1565 When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1566 a partial match, it is possible to call it again, with additional subject
1567 characters, and have it continue with the same match. The PCRE_DFA_RESTART
1568 option requests this action; when it is set, the <i>workspace</i> and
1569 <i>wscount</i> options must reference the same vector as before because data
1570 about the match so far is left in them after a partial match. There is more
1571 discussion of this facility in the
1572 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1573 documentation.
1574 </P>
1575 <br><b>
1576 Successful returns from <b>pcre_dfa_exec()</b>
1577 </b><br>
1578 <P>
1579 When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1580 substring in the subject. Note, however, that all the matches from one run of
1581 the function start at the same point in the subject. The shorter matches are
1582 all initial substrings of the longer matches. For example, if the pattern
1583 <pre>
1584 &#60;.*&#62;
1585 </pre>
1586 is matched against the string
1587 <pre>
1588 This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1589 </pre>
1590 the three matched strings are
1591 <pre>
1592 &#60;something&#62;
1593 &#60;something&#62; &#60;something else&#62;
1594 &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1595 </pre>
1596 On success, the yield of the function is a number greater than zero, which is
1597 the number of matched substrings. The substrings themselves are returned in
1598 <i>ovector</i>. Each string uses two elements; the first is the offset to the
1599 start, and the second is the offset to the end. All the strings have the same
1600 start offset. (Space could have been saved by giving this only once, but it was
1601 decided to retain some compatibility with the way <b>pcre_exec()</b> returns
1602 data, even though the meaning of the strings is different.)
1603 </P>
1604 <P>
1605 The strings are returned in reverse order of length; that is, the longest
1606 matching string is given first. If there were too many matches to fit into
1607 <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1608 the longest matches.
1609 </P>
1610 <br><b>
1611 Error returns from <b>pcre_dfa_exec()</b>
1612 </b><br>
1613 <P>
1614 The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1615 Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1616 described
1617 <a href="#errorlist">above.</a>
1618 There are in addition the following errors that are specific to
1619 <b>pcre_dfa_exec()</b>:
1620 <pre>
1621 PCRE_ERROR_DFA_UITEM (-16)
1622 </pre>
1623 This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1624 that it does not support, for instance, the use of \C or a back reference.
1625 <pre>
1626 PCRE_ERROR_DFA_UCOND (-17)
1627 </pre>
1628 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item in a
1629 pattern that uses a back reference for the condition. This is not supported.
1630 <pre>
1631 PCRE_ERROR_DFA_UMLIMIT (-18)
1632 </pre>
1633 This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1634 block that contains a setting of the <i>match_limit</i> field. This is not
1635 supported (it is meaningless).
1636 <pre>
1637 PCRE_ERROR_DFA_WSSIZE (-19)
1638 </pre>
1639 This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1640 <i>workspace</i> vector.
1641 <pre>
1642 PCRE_ERROR_DFA_RECURSE (-20)
1643 </pre>
1644 When a recursive subpattern is processed, the matching function calls itself
1645 recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1646 error is given if the output vector is not large enough. This should be
1647 extremely rare, as a vector of size 1000 is used.
1648 </P>
1649 <P>
1650 Last updated: 18 January 2006
1651 <br>
1652 Copyright &copy; 1997-2006 University of Cambridge.
1653 <p>
1654 Return to the <a href="index.html">PCRE index page</a>.
1655 </p>

  ViewVC Help
Powered by ViewVC 1.1.5