ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 869 - (show annotations)
Sat Jan 14 11:16:23 2012 UTC (7 years, 8 months ago) by ph10
File MIME type: text/html
File size: 118920 byte(s)
Bring HTML docs up to date.
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 BASIC FUNCTIONS</a>
18 <li><a name="TOC3" href="#SEC3">PCRE NATIVE API AUXILIARY FUNCTIONS</a>
20 <li><a name="TOC5" href="#SEC5">PCRE 8-BIT AND 16-BIT LIBRARIES</a>
21 <li><a name="TOC6" href="#SEC6">PCRE API OVERVIEW</a>
22 <li><a name="TOC7" href="#SEC7">NEWLINES</a>
23 <li><a name="TOC8" href="#SEC8">MULTITHREADING</a>
25 <li><a name="TOC10" href="#SEC10">CHECKING BUILD-TIME OPTIONS</a>
26 <li><a name="TOC11" href="#SEC11">COMPILING A PATTERN</a>
27 <li><a name="TOC12" href="#SEC12">COMPILATION ERROR CODES</a>
28 <li><a name="TOC13" href="#SEC13">STUDYING A PATTERN</a>
29 <li><a name="TOC14" href="#SEC14">LOCALE SUPPORT</a>
30 <li><a name="TOC15" href="#SEC15">INFORMATION ABOUT A PATTERN</a>
31 <li><a name="TOC16" href="#SEC16">REFERENCE COUNTS</a>
35 <li><a name="TOC20" href="#SEC20">DUPLICATE SUBPATTERN NAMES</a>
36 <li><a name="TOC21" href="#SEC21">FINDING ALL POSSIBLE MATCHES</a>
38 <li><a name="TOC23" href="#SEC23">SEE ALSO</a>
39 <li><a name="TOC24" href="#SEC24">AUTHOR</a>
40 <li><a name="TOC25" href="#SEC25">REVISION</a>
41 </ul>
42 <P>
43 <b>#include &#60;pcre.h&#62;</b>
44 </P>
45 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API BASIC FUNCTIONS</a><br>
46 <P>
47 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</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 *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
53 <b>int *<i>errorcodeptr</i>,</b>
54 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
55 <b>const unsigned char *<i>tableptr</i>);</b>
56 </P>
57 <P>
58 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
59 <b>const char **<i>errptr</i>);</b>
60 </P>
61 <P>
62 <b>void pcre_free_study(pcre_extra *<i>extra</i>);</b>
63 </P>
64 <P>
65 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
66 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
67 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
68 </P>
69 <P>
70 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
71 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
72 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
73 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
74 </P>
76 <P>
77 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
78 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
79 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
80 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
81 </P>
82 <P>
83 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
84 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
85 <b>int <i>buffersize</i>);</b>
86 </P>
87 <P>
88 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
89 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
90 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
91 <b>const char **<i>stringptr</i>);</b>
92 </P>
93 <P>
94 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
95 <b>const char *<i>name</i>);</b>
96 </P>
97 <P>
98 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
99 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
100 </P>
101 <P>
102 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
103 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
104 <b>const char **<i>stringptr</i>);</b>
105 </P>
106 <P>
107 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
108 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
109 </P>
110 <P>
111 <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
112 </P>
113 <P>
114 <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
115 </P>
116 <br><a name="SEC3" href="#TOC1">PCRE NATIVE API AUXILIARY FUNCTIONS</a><br>
117 <P>
118 <b>pcre_jit_stack *pcre_jit_stack_alloc(int <i>startsize</i>, int <i>maxsize</i>);</b>
119 </P>
120 <P>
121 <b>void pcre_jit_stack_free(pcre_jit_stack *<i>stack</i>);</b>
122 </P>
123 <P>
124 <b>void pcre_assign_jit_stack(pcre_extra *<i>extra</i>,</b>
125 <b>pcre_jit_callback <i>callback</i>, void *<i>data</i>);</b>
126 </P>
127 <P>
128 <b>const unsigned char *pcre_maketables(void);</b>
129 </P>
130 <P>
131 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
132 <b>int <i>what</i>, void *<i>where</i>);</b>
133 </P>
134 <P>
135 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
136 </P>
137 <P>
138 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
139 </P>
140 <P>
141 <b>const char *pcre_version(void);</b>
142 </P>
143 <P>
144 <b>int pcre_pattern_to_host_byte_order(pcre *<i>code</i>,</b>
145 <b>pcre_extra *<i>extra</i>, const unsigned char *<i>tables</i>);</b>
146 </P>
147 <br><a name="SEC4" href="#TOC1">PCRE NATIVE API INDIRECTED FUNCTIONS</a><br>
148 <P>
149 <b>void *(*pcre_malloc)(size_t);</b>
150 </P>
151 <P>
152 <b>void (*pcre_free)(void *);</b>
153 </P>
154 <P>
155 <b>void *(*pcre_stack_malloc)(size_t);</b>
156 </P>
157 <P>
158 <b>void (*pcre_stack_free)(void *);</b>
159 </P>
160 <P>
161 <b>int (*pcre_callout)(pcre_callout_block *);</b>
162 </P>
163 <br><a name="SEC5" href="#TOC1">PCRE 8-BIT AND 16-BIT LIBRARIES</a><br>
164 <P>
165 From release 8.30, PCRE can be compiled as a library for handling 16-bit
166 character strings as well as, or instead of, the original library that handles
167 8-bit character strings. To avoid too much complication, this document
168 describes the 8-bit versions of the functions, with only occasional references
169 to the 16-bit library.
170 </P>
171 <P>
172 The 16-bit functions operate in the same way as their 8-bit counterparts; they
173 just use different data types for their arguments and results, and their names
174 start with <b>pcre16_</b> instead of <b>pcre_</b>. For every option that has UTF8
175 in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
176 UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
177 option names define the same bit values.
178 </P>
179 <P>
180 References to bytes and UTF-8 in this document should be read as references to
181 16-bit data quantities and UTF-16 when using the 16-bit library, unless
182 specified otherwise. More details of the specific differences for the 16-bit
183 library are given in the
184 <a href="pcre16.html"><b>pcre16</b></a>
185 page.
186 </P>
187 <br><a name="SEC6" href="#TOC1">PCRE API OVERVIEW</a><br>
188 <P>
189 PCRE has its own native API, which is described in this document. There are
190 also some wrapper functions (for the 8-bit library only) that correspond to the
191 POSIX regular expression API, but they do not give access to all the
192 functionality. They are described in the
193 <a href="pcreposix.html"><b>pcreposix</b></a>
194 documentation. Both of these APIs define a set of C function calls. A C++
195 wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
196 documented in the
197 <a href="pcrecpp.html"><b>pcrecpp</b></a>
198 page.
199 </P>
200 <P>
201 The native API C function prototypes are defined in the header file
202 <b>pcre.h</b>, and on Unix-like systems the (8-bit) library itself is called
203 <b>libpcre</b>. It can normally be accessed by adding <b>-lpcre</b> to the
204 command for linking an application that uses PCRE. The header file defines the
205 macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
206 for the library. Applications can use these to include support for different
207 releases of PCRE.
208 </P>
209 <P>
210 In a Windows environment, if you want to statically link an application program
211 against a non-dll <b>pcre.a</b> file, you must define PCRE_STATIC before
212 including <b>pcre.h</b> or <b>pcrecpp.h</b>, because otherwise the
213 <b>pcre_malloc()</b> and <b>pcre_free()</b> exported functions will be declared
214 <b>__declspec(dllimport)</b>, with unwanted results.
215 </P>
216 <P>
217 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
218 and <b>pcre_exec()</b> are used for compiling and matching regular expressions
219 in a Perl-compatible manner. A sample program that demonstrates the simplest
220 way of using them is provided in the file called <i>pcredemo.c</i> in the PCRE
221 source distribution. A listing of this program is given in the
222 <a href="pcredemo.html"><b>pcredemo</b></a>
223 documentation, and the
224 <a href="pcresample.html"><b>pcresample</b></a>
225 documentation describes how to compile and run it.
226 </P>
227 <P>
228 Just-in-time compiler support is an optional feature of PCRE that can be built
229 in appropriate hardware environments. It greatly speeds up the matching
230 performance of many patterns. Simple programs can easily request that it be
231 used if available, by setting an option that is ignored when it is not
232 relevant. More complicated programs might need to make use of the functions
233 <b>pcre_jit_stack_alloc()</b>, <b>pcre_jit_stack_free()</b>, and
234 <b>pcre_assign_jit_stack()</b> in order to control the JIT code's memory usage.
235 These functions are discussed in the
236 <a href="pcrejit.html"><b>pcrejit</b></a>
237 documentation.
238 </P>
239 <P>
240 A second matching function, <b>pcre_dfa_exec()</b>, which is not
241 Perl-compatible, is also provided. This uses a different algorithm for the
242 matching. The alternative algorithm finds all possible matches (at a given
243 point in the subject), and scans the subject just once (unless there are
244 lookbehind assertions). However, this algorithm does not return captured
245 substrings. A description of the two matching algorithms and their advantages
246 and disadvantages is given in the
247 <a href="pcrematching.html"><b>pcrematching</b></a>
248 documentation.
249 </P>
250 <P>
251 In addition to the main compiling and matching functions, there are convenience
252 functions for extracting captured substrings from a subject string that is
253 matched by <b>pcre_exec()</b>. They are:
254 <pre>
255 <b>pcre_copy_substring()</b>
256 <b>pcre_copy_named_substring()</b>
257 <b>pcre_get_substring()</b>
258 <b>pcre_get_named_substring()</b>
259 <b>pcre_get_substring_list()</b>
260 <b>pcre_get_stringnumber()</b>
261 <b>pcre_get_stringtable_entries()</b>
262 </pre>
263 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
264 provided, to free the memory used for extracted strings.
265 </P>
266 <P>
267 The function <b>pcre_maketables()</b> is used to build a set of character tables
268 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
269 or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
270 specialist use. Most commonly, no special tables are passed, in which case
271 internal tables that are generated when PCRE is built are used.
272 </P>
273 <P>
274 The function <b>pcre_fullinfo()</b> is used to find out information about a
275 compiled pattern. The function <b>pcre_version()</b> returns a pointer to a
276 string containing the version of PCRE and its date of release.
277 </P>
278 <P>
279 The function <b>pcre_refcount()</b> maintains a reference count in a data block
280 containing a compiled pattern. This is provided for the benefit of
281 object-oriented applications.
282 </P>
283 <P>
284 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
285 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
286 respectively. PCRE calls the memory management functions via these variables,
287 so a calling program can replace them if it wishes to intercept the calls. This
288 should be done before calling any PCRE functions.
289 </P>
290 <P>
291 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
292 indirections to memory management functions. These special functions are used
293 only when PCRE is compiled to use the heap for remembering data, instead of
294 recursive function calls, when running the <b>pcre_exec()</b> function. See the
295 <a href="pcrebuild.html"><b>pcrebuild</b></a>
296 documentation for details of how to do this. It is a non-standard way of
297 building PCRE, for use in environments that have limited stacks. Because of the
298 greater use of memory management, it runs more slowly. Separate functions are
299 provided so that special-purpose external code can be used for this case. When
300 used, these functions are always called in a stack-like manner (last obtained,
301 first freed), and always for memory blocks of the same size. There is a
302 discussion about PCRE's stack usage in the
303 <a href="pcrestack.html"><b>pcrestack</b></a>
304 documentation.
305 </P>
306 <P>
307 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
308 by the caller to a "callout" function, which PCRE will then call at specified
309 points during a matching operation. Details are given in the
310 <a href="pcrecallout.html"><b>pcrecallout</b></a>
311 documentation.
312 <a name="newlines"></a></P>
313 <br><a name="SEC7" href="#TOC1">NEWLINES</a><br>
314 <P>
315 PCRE supports five different conventions for indicating line breaks in
316 strings: a single CR (carriage return) character, a single LF (linefeed)
317 character, the two-character sequence CRLF, any of the three preceding, or any
318 Unicode newline sequence. The Unicode newline sequences are the three just
319 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
320 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
321 (paragraph separator, U+2029).
322 </P>
323 <P>
324 Each of the first three conventions is used by at least one operating system as
325 its standard newline sequence. When PCRE is built, a default can be specified.
326 The default default is LF, which is the Unix standard. When PCRE is run, the
327 default can be overridden, either when a pattern is compiled, or when it is
328 matched.
329 </P>
330 <P>
331 At compile time, the newline convention can be specified by the <i>options</i>
332 argument of <b>pcre_compile()</b>, or it can be specified by special text at the
333 start of the pattern itself; this overrides any other settings. See the
334 <a href="pcrepattern.html"><b>pcrepattern</b></a>
335 page for details of the special character sequences.
336 </P>
337 <P>
338 In the PCRE documentation the word "newline" is used to mean "the character or
339 pair of characters that indicate a line break". The choice of newline
340 convention affects the handling of the dot, circumflex, and dollar
341 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
342 recognized line ending sequence, the match position advancement for a
343 non-anchored pattern. There is more detail about this in the
344 <a href="#execoptions">section on <b>pcre_exec()</b> options</a>
345 below.
346 </P>
347 <P>
348 The choice of newline convention does not affect the interpretation of
349 the \n or \r escape sequences, nor does it affect what \R matches, which is
350 controlled in a similar way, but by separate options.
351 </P>
352 <br><a name="SEC8" href="#TOC1">MULTITHREADING</a><br>
353 <P>
354 The PCRE functions can be used in multi-threading applications, with the
355 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
356 <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
357 callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
358 </P>
359 <P>
360 The compiled form of a regular expression is not altered during matching, so
361 the same compiled pattern can safely be used by several threads at once.
362 </P>
363 <P>
364 If the just-in-time optimization feature is being used, it needs separate
365 memory stack areas for each thread. See the
366 <a href="pcrejit.html"><b>pcrejit</b></a>
367 documentation for more details.
368 </P>
369 <br><a name="SEC9" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
370 <P>
371 The compiled form of a regular expression can be saved and re-used at a later
372 time, possibly by a different program, and even on a host other than the one on
373 which it was compiled. Details are given in the
374 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
375 documentation, which includes a description of the
376 <b>pcre_pattern_to_host_byte_order()</b> function. However, compiling a regular
377 expression with one version of PCRE for use with a different version is not
378 guaranteed to work and may cause crashes.
379 </P>
380 <br><a name="SEC10" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
381 <P>
382 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
383 </P>
384 <P>
385 The function <b>pcre_config()</b> makes it possible for a PCRE client to
386 discover which optional features have been compiled into the PCRE library. The
387 <a href="pcrebuild.html"><b>pcrebuild</b></a>
388 documentation has more details about these optional features.
389 </P>
390 <P>
391 The first argument for <b>pcre_config()</b> is an integer, specifying which
392 information is required; the second argument is a pointer to a variable into
393 which the information is placed. The returned value is zero on success, or the
394 negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
395 not recognized. The following information is available:
396 <pre>
398 </pre>
399 The output is an integer that is set to one if UTF-8 support is available;
400 otherwise it is set to zero. If this option is given to the 16-bit version of
401 this function, <b>pcre16_config()</b>, the result is PCRE_ERROR_BADOPTION.
402 <pre>
404 </pre>
405 The output is an integer that is set to one if UTF-16 support is available;
406 otherwise it is set to zero. This value should normally be given to the 16-bit
407 version of this function, <b>pcre16_config()</b>. If it is given to the 8-bit
408 version of this function, the result is PCRE_ERROR_BADOPTION.
409 <pre>
411 </pre>
412 The output is an integer that is set to one if support for Unicode character
413 properties is available; otherwise it is set to zero.
414 <pre>
416 </pre>
417 The output is an integer that is set to one if support for just-in-time
418 compiling is available; otherwise it is set to zero.
419 <pre>
421 </pre>
422 The output is an integer whose value specifies the default character sequence
423 that is recognized as meaning "newline". The four values that are supported
424 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
425 Though they are derived from ASCII, the same values are returned in EBCDIC
426 environments. The default should normally correspond to the standard sequence
427 for your operating system.
428 <pre>
430 </pre>
431 The output is an integer whose value indicates what character sequences the \R
432 escape sequence matches by default. A value of 0 means that \R matches any
433 Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
434 or CRLF. The default can be overridden when a pattern is compiled or matched.
435 <pre>
437 </pre>
438 The output is an integer that contains the number of bytes used for internal
439 linkage in compiled regular expressions. For the 8-bit library, the value can
440 be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
441 a number of bytes. The default value of 2 is sufficient for all but the most
442 massive patterns, since it allows the compiled pattern to be up to 64K in size.
443 Larger values allow larger regular expressions to be compiled, at the expense
444 of slower matching.
445 <pre>
447 </pre>
448 The output is an integer that contains the threshold above which the POSIX
449 interface uses <b>malloc()</b> for output vectors. Further details are given in
450 the
451 <a href="pcreposix.html"><b>pcreposix</b></a>
452 documentation.
453 <pre>
455 </pre>
456 The output is a long integer that gives the default limit for the number of
457 internal matching function calls in a <b>pcre_exec()</b> execution. Further
458 details are given with <b>pcre_exec()</b> below.
459 <pre>
461 </pre>
462 The output is a long integer that gives the default limit for the depth of
463 recursion when calling the internal matching function in a <b>pcre_exec()</b>
464 execution. Further details are given with <b>pcre_exec()</b> below.
465 <pre>
467 </pre>
468 The output is an integer that is set to one if internal recursion when running
469 <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
470 to remember their state. This is the usual way that PCRE is compiled. The
471 output is zero if PCRE was compiled to use blocks of data on the heap instead
472 of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
473 <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
474 avoiding the use of the stack.
475 </P>
476 <br><a name="SEC11" href="#TOC1">COMPILING A PATTERN</a><br>
477 <P>
478 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
479 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
480 <b>const unsigned char *<i>tableptr</i>);</b>
481 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
482 <b>int *<i>errorcodeptr</i>,</b>
483 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
484 <b>const unsigned char *<i>tableptr</i>);</b>
485 </P>
486 <P>
487 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
488 called to compile a pattern into an internal form. The only difference between
489 the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
490 <i>errorcodeptr</i>, via which a numerical error code can be returned. To avoid
491 too much repetition, we refer just to <b>pcre_compile()</b> below, but the
492 information applies equally to <b>pcre_compile2()</b>.
493 </P>
494 <P>
495 The pattern is a C string terminated by a binary zero, and is passed in the
496 <i>pattern</i> argument. A pointer to a single block of memory that is obtained
497 via <b>pcre_malloc</b> is returned. This contains the compiled code and related
498 data. The <b>pcre</b> type is defined for the returned block; this is a typedef
499 for a structure whose contents are not externally defined. It is up to the
500 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
501 </P>
502 <P>
503 Although the compiled code of a PCRE regex is relocatable, that is, it does not
504 depend on memory location, the complete <b>pcre</b> data block is not
505 fully relocatable, because it may contain a copy of the <i>tableptr</i>
506 argument, which is an address (see below).
507 </P>
508 <P>
509 The <i>options</i> argument contains various bit settings that affect the
510 compilation. It should be zero if no options are required. The available
511 options are described below. Some of them (in particular, those that are
512 compatible with Perl, but some others as well) can also be set and unset from
513 within the pattern (see the detailed description in the
514 <a href="pcrepattern.html"><b>pcrepattern</b></a>
515 documentation). For those options that can be different in different parts of
516 the pattern, the contents of the <i>options</i> argument specifies their
517 settings at the start of compilation and execution. The PCRE_ANCHORED,
518 PCRE_BSR_<i>xxx</i>, PCRE_NEWLINE_<i>xxx</i>, PCRE_NO_UTF8_CHECK, and
519 PCRE_NO_START_OPT options can be set at the time of matching as well as at
520 compile time.
521 </P>
522 <P>
523 If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
524 Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
525 NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
526 error message. This is a static string that is part of the library. You must
527 not try to free it. Normally, the offset from the start of the pattern to the
528 byte that was being processed when the error was discovered is placed in the
529 variable pointed to by <i>erroffset</i>, which must not be NULL (if it is, an
530 immediate error is given). However, for an invalid UTF-8 string, the offset is
531 that of the first byte of the failing character.
532 </P>
533 <P>
534 Some errors are not detected until the whole pattern has been scanned; in these
535 cases, the offset passed back is the length of the pattern. Note that the
536 offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
537 into the middle of a UTF-8 character.
538 </P>
539 <P>
540 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
541 <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
542 returned via this argument in the event of an error. This is in addition to the
543 textual error message. Error codes and messages are listed below.
544 </P>
545 <P>
546 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
547 character tables that are built when PCRE is compiled, using the default C
548 locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
549 call to <b>pcre_maketables()</b>. This value is stored with the compiled
550 pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
551 passed to it. For more discussion, see the section on locale support below.
552 </P>
553 <P>
554 This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
555 <pre>
556 pcre *re;
557 const char *error;
558 int erroffset;
559 re = pcre_compile(
560 "^A.*Z", /* the pattern */
561 0, /* default options */
562 &error, /* for error message */
563 &erroffset, /* for error offset */
564 NULL); /* use default character tables */
565 </pre>
566 The following names for option bits are defined in the <b>pcre.h</b> header
567 file:
568 <pre>
570 </pre>
571 If this bit is set, the pattern is forced to be "anchored", that is, it is
572 constrained to match only at the first matching point in the string that is
573 being searched (the "subject string"). This effect can also be achieved by
574 appropriate constructs in the pattern itself, which is the only way to do it in
575 Perl.
576 <pre>
578 </pre>
579 If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
580 all with number 255, before each pattern item. For discussion of the callout
581 facility, see the
582 <a href="pcrecallout.html"><b>pcrecallout</b></a>
583 documentation.
584 <pre>
587 </pre>
588 These options (which are mutually exclusive) control what the \R escape
589 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
590 match any Unicode newline sequence. The default is specified when PCRE is
591 built. It can be overridden from within the pattern, or by setting an option
592 when a compiled pattern is matched.
593 <pre>
595 </pre>
596 If this bit is set, letters in the pattern match both upper and lower case
597 letters. It is equivalent to Perl's /i option, and it can be changed within a
598 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
599 concept of case for characters whose values are less than 128, so caseless
600 matching is always possible. For characters with higher values, the concept of
601 case is supported if PCRE is compiled with Unicode property support, but not
602 otherwise. If you want to use caseless matching for characters 128 and above,
603 you must ensure that PCRE is compiled with Unicode property support as well as
604 with UTF-8 support.
605 <pre>
607 </pre>
608 If this bit is set, a dollar metacharacter in the pattern matches only at the
609 end of the subject string. Without this option, a dollar also matches
610 immediately before a newline at the end of the string (but not before any other
611 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
612 There is no equivalent to this option in Perl, and no way to set it within a
613 pattern.
614 <pre>
616 </pre>
617 If this bit is set, a dot metacharacter in the pattern matches a character of
618 any value, including one that indicates a newline. However, it only ever
619 matches one character, even if newlines are coded as CRLF. Without this option,
620 a dot does not match when the current position is at a newline. This option is
621 equivalent to Perl's /s option, and it can be changed within a pattern by a
622 (?s) option setting. A negative class such as [^a] always matches newline
623 characters, independent of the setting of this option.
624 <pre>
626 </pre>
627 If this bit is set, names used to identify capturing subpatterns need not be
628 unique. This can be helpful for certain types of pattern when it is known that
629 only one instance of the named subpattern can ever be matched. There are more
630 details of named subpatterns below; see also the
631 <a href="pcrepattern.html"><b>pcrepattern</b></a>
632 documentation.
633 <pre>
635 </pre>
636 If this bit is set, whitespace data characters in the pattern are totally
637 ignored except when escaped or inside a character class. Whitespace does not
638 include the VT character (code 11). In addition, characters between an
639 unescaped # outside a character class and the next newline, inclusive, are also
640 ignored. This is equivalent to Perl's /x option, and it can be changed within a
641 pattern by a (?x) option setting.
642 </P>
643 <P>
644 Which characters are interpreted as newlines is controlled by the options
645 passed to <b>pcre_compile()</b> or by a special sequence at the start of the
646 pattern, as described in the section entitled
647 <a href="pcrepattern.html#newlines">"Newline conventions"</a>
648 in the <b>pcrepattern</b> documentation. Note that the end of this type of
649 comment is a literal newline sequence in the pattern; escape sequences that
650 happen to represent a newline do not count.
651 </P>
652 <P>
653 This option makes it possible to include comments inside complicated patterns.
654 Note, however, that this applies only to data characters. Whitespace characters
655 may never appear within special character sequences in a pattern, for example
656 within the sequence (?( that introduces a conditional subpattern.
657 <pre>
659 </pre>
660 This option was invented in order to turn on additional functionality of PCRE
661 that is incompatible with Perl, but it is currently of very little use. When
662 set, any backslash in a pattern that is followed by a letter that has no
663 special meaning causes an error, thus reserving these combinations for future
664 expansion. By default, as in Perl, a backslash followed by a letter with no
665 special meaning is treated as a literal. (Perl can, however, be persuaded to
666 give an error for this, by running it with the -w option.) There are at present
667 no other features controlled by this option. It can also be set by a (?X)
668 option setting within a pattern.
669 <pre>
671 </pre>
672 If this option is set, an unanchored pattern is required to match before or at
673 the first newline in the subject string, though the matched text may continue
674 over the newline.
675 <pre>
677 </pre>
678 If this option is set, PCRE's behaviour is changed in some ways so that it is
679 compatible with JavaScript rather than Perl. The changes are as follows:
680 </P>
681 <P>
682 (1) A lone closing square bracket in a pattern causes a compile-time error,
683 because this is illegal in JavaScript (by default it is treated as a data
684 character). Thus, the pattern AB]CD becomes illegal when this option is set.
685 </P>
686 <P>
687 (2) At run time, a back reference to an unset subpattern group matches an empty
688 string (by default this causes the current matching alternative to fail). A
689 pattern such as (\1)(a) succeeds when this option is set (assuming it can find
690 an "a" in the subject), whereas it fails by default, for Perl compatibility.
691 </P>
692 <P>
693 (3) \U matches an upper case "U" character; by default \U causes a compile
694 time error (Perl uses \U to upper case subsequent characters).
695 </P>
696 <P>
697 (4) \u matches a lower case "u" character unless it is followed by four
698 hexadecimal digits, in which case the hexadecimal number defines the code point
699 to match. By default, \u causes a compile time error (Perl uses it to upper
700 case the following character).
701 </P>
702 <P>
703 (5) \x matches a lower case "x" character unless it is followed by two
704 hexadecimal digits, in which case the hexadecimal number defines the code point
705 to match. By default, as in Perl, a hexadecimal number is always expected after
706 \x, but it may have zero, one, or two digits (so, for example, \xz matches a
707 binary zero character followed by z).
708 <pre>
710 </pre>
711 By default, PCRE treats the subject string as consisting of a single line of
712 characters (even if it actually contains newlines). The "start of line"
713 metacharacter (^) matches only at the start of the string, while the "end of
714 line" metacharacter ($) matches only at the end of the string, or before a
715 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
716 Perl.
717 </P>
718 <P>
719 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
720 match immediately following or immediately before internal newlines in the
721 subject string, respectively, as well as at the very start and end. This is
722 equivalent to Perl's /m option, and it can be changed within a pattern by a
723 (?m) option setting. If there are no newlines in a subject string, or no
724 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
725 <pre>
731 </pre>
732 These options override the default newline definition that was chosen when PCRE
733 was built. Setting the first or the second specifies that a newline is
734 indicated by a single character (CR or LF, respectively). Setting
735 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
736 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
737 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
738 that any Unicode newline sequence should be recognized. The Unicode newline
739 sequences are the three just mentioned, plus the single characters VT (vertical
740 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
741 separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
742 library, the last two are recognized only in UTF-8 mode.
743 </P>
744 <P>
745 The newline setting in the options word uses three bits that are treated
746 as a number, giving eight possibilities. Currently only six are used (default
747 plus the five values above). This means that if you set more than one newline
748 option, the combination may or may not be sensible. For example,
750 other combinations may yield unused numbers and cause an error.
751 </P>
752 <P>
753 The only time that a line break in a pattern is specially recognized when
754 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
755 and so are ignored in this mode. Also, an unescaped # outside a character class
756 indicates a comment that lasts until after the next line break sequence. In
757 other circumstances, line break sequences in patterns are treated as literal
758 data.
759 </P>
760 <P>
761 The newline option that is set at compile time becomes the default that is used
762 for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
763 <pre>
765 </pre>
766 If this option is set, it disables the use of numbered capturing parentheses in
767 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
768 were followed by ?: but named parentheses can still be used for capturing (and
769 they acquire numbers in the usual way). There is no equivalent of this option
770 in Perl.
771 <pre>
773 </pre>
774 This is an option that acts at matching time; that is, it is really an option
775 for <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. If it is set at compile time,
776 it is remembered with the compiled pattern and assumed at matching time. For
777 details see the discussion of PCRE_NO_START_OPTIMIZE
778 <a href="#execoptions">below.</a>
779 <pre>
781 </pre>
782 This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
783 \w, and some of the POSIX character classes. By default, only ASCII characters
784 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
785 classify characters. More details are given in the section on
786 <a href="pcre.html#genericchartypes">generic character types</a>
787 in the
788 <a href="pcrepattern.html"><b>pcrepattern</b></a>
789 page. If you set PCRE_UCP, matching one of the items it affects takes much
790 longer. The option is available only if PCRE has been compiled with Unicode
791 property support.
792 <pre>
794 </pre>
795 This option inverts the "greediness" of the quantifiers so that they are not
796 greedy by default, but become greedy if followed by "?". It is not compatible
797 with Perl. It can also be set by a (?U) option setting within the pattern.
798 <pre>
800 </pre>
801 This option causes PCRE to regard both the pattern and the subject as strings
802 of UTF-8 characters instead of single-byte strings. However, it is available
803 only when PCRE is built to include UTF support. If not, the use of this option
804 provokes an error. Details of how this option changes the behaviour of PCRE are
805 given in the
806 <a href="pcreunicode.html"><b>pcreunicode</b></a>
807 page.
808 <pre>
810 </pre>
811 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8
812 string is automatically checked. There is a discussion about the
813 <a href="pcreunicode.html#utf8strings">validity of UTF-8 strings</a>
814 in the
815 <a href="pcreunicode.html"><b>pcreunicode</b></a>
816 page. If an invalid UTF-8 sequence is found, <b>pcre_compile()</b> returns an
817 error. If you already know that your pattern is valid, and you want to skip
818 this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
819 When it is set, the effect of passing an invalid UTF-8 string as a pattern is
820 undefined. It may cause your program to crash. Note that this option can also
821 be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress the
822 validity checking of subject strings.
823 </P>
824 <br><a name="SEC12" href="#TOC1">COMPILATION ERROR CODES</a><br>
825 <P>
826 The following table lists the error codes than may be returned by
827 <b>pcre_compile2()</b>, along with the error messages that may be returned by
828 both compiling functions. Note that error messages are always 8-bit ASCII
829 strings, even in 16-bit mode. As PCRE has developed, some error codes have
830 fallen out of use. To avoid confusion, they have not been re-used.
831 <pre>
832 0 no error
833 1 \ at end of pattern
834 2 \c at end of pattern
835 3 unrecognized character follows \
836 4 numbers out of order in {} quantifier
837 5 number too big in {} quantifier
838 6 missing terminating ] for character class
839 7 invalid escape sequence in character class
840 8 range out of order in character class
841 9 nothing to repeat
842 10 [this code is not in use]
843 11 internal error: unexpected repeat
844 12 unrecognized character after (? or (?-
845 13 POSIX named classes are supported only within a class
846 14 missing )
847 15 reference to non-existent subpattern
848 16 erroffset passed as NULL
849 17 unknown option bit(s) set
850 18 missing ) after comment
851 19 [this code is not in use]
852 20 regular expression is too large
853 21 failed to get memory
854 22 unmatched parentheses
855 23 internal error: code overflow
856 24 unrecognized character after (?&#60;
857 25 lookbehind assertion is not fixed length
858 26 malformed number or name after (?(
859 27 conditional group contains more than two branches
860 28 assertion expected after (?(
861 29 (?R or (?[+-]digits must be followed by )
862 30 unknown POSIX class name
863 31 POSIX collating elements are not supported
864 32 this version of PCRE is compiled without UTF support
865 33 [this code is not in use]
866 34 character value in \x{...} sequence is too large
867 35 invalid condition (?(0)
868 36 \C not allowed in lookbehind assertion
869 37 PCRE does not support \L, \l, \N{name}, \U, or \u
870 38 number after (?C is &#62; 255
871 39 closing ) for (?C expected
872 40 recursive call could loop indefinitely
873 41 unrecognized character after (?P
874 42 syntax error in subpattern name (missing terminator)
875 43 two named subpatterns have the same name
876 44 invalid UTF-8 string (specifically UTF-8)
877 45 support for \P, \p, and \X has not been compiled
878 46 malformed \P or \p sequence
879 47 unknown property name after \P or \p
880 48 subpattern name is too long (maximum 32 characters)
881 49 too many named subpatterns (maximum 10000)
882 50 [this code is not in use]
883 51 octal value is greater than \377 in 8-bit non-UTF-8 mode
884 52 internal error: overran compiling workspace
885 53 internal error: previously-checked referenced subpattern
886 not found
887 54 DEFINE group contains more than one branch
888 55 repeating a DEFINE group is not allowed
889 56 inconsistent NEWLINE options
890 57 \g is not followed by a braced, angle-bracketed, or quoted
891 name/number or by a plain number
892 58 a numbered reference must not be zero
893 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
894 60 (*VERB) not recognized
895 61 number is too big
896 62 subpattern name expected
897 63 digit expected after (?+
898 64 ] is an invalid data character in JavaScript compatibility mode
899 65 different names for subpatterns of the same number are
900 not allowed
901 66 (*MARK) must have an argument
902 67 this version of PCRE is not compiled with Unicode property
903 support
904 68 \c must be followed by an ASCII character
905 69 \k is not followed by a braced, angle-bracketed, or quoted name
906 70 internal error: unknown opcode in find_fixedlength()
907 71 \N is not supported in a class
908 72 too many forward references
909 73 disallowed Unicode code point (&#62;= 0xd800 && &#60;= 0xdfff)
910 74 invalid UTF-16 string (specifically UTF-16)
911 </pre>
912 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
913 be used if the limits were changed when PCRE was built.
914 <a name="studyingapattern"></a></P>
915 <br><a name="SEC13" href="#TOC1">STUDYING A PATTERN</a><br>
916 <P>
917 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
918 <b>const char **<i>errptr</i>);</b>
919 </P>
920 <P>
921 If a compiled pattern is going to be used several times, it is worth spending
922 more time analyzing it in order to speed up the time taken for matching. The
923 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
924 argument. If studying the pattern produces additional information that will
925 help speed up matching, <b>pcre_study()</b> returns a pointer to a
926 <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
927 results of the study.
928 </P>
929 <P>
930 The returned value from <b>pcre_study()</b> can be passed directly to
931 <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. However, a <b>pcre_extra</b> block
932 also contains other fields that can be set by the caller before the block is
933 passed; these are described
934 <a href="#extradata">below</a>
935 in the section on matching a pattern.
936 </P>
937 <P>
938 If studying the pattern does not produce any useful information,
939 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
940 wants to pass any of the other fields to <b>pcre_exec()</b> or
941 <b>pcre_dfa_exec()</b>, it must set up its own <b>pcre_extra</b> block.
942 </P>
943 <P>
944 The second argument of <b>pcre_study()</b> contains option bits. There is only
945 one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
946 compiler is available, the pattern is further compiled into machine code that
947 executes much faster than the <b>pcre_exec()</b> matching function. If
948 the just-in-time compiler is not available, this option is ignored. All other
949 bits in the <i>options</i> argument must be zero.
950 </P>
951 <P>
952 JIT compilation is a heavyweight optimization. It can take some time for
953 patterns to be analyzed, and for one-off matches and simple patterns the
954 benefit of faster execution might be offset by a much slower study time.
955 Not all patterns can be optimized by the JIT compiler. For those that cannot be
956 handled, matching automatically falls back to the <b>pcre_exec()</b>
957 interpreter. For more details, see the
958 <a href="pcrejit.html"><b>pcrejit</b></a>
959 documentation.
960 </P>
961 <P>
962 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
963 studying succeeds (even if no data is returned), the variable it points to is
964 set to NULL. Otherwise it is set to point to a textual error message. This is a
965 static string that is part of the library. You must not try to free it. You
966 should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
967 sure that it has run successfully.
968 </P>
969 <P>
970 When you are finished with a pattern, you can free the memory used for the
971 study data by calling <b>pcre_free_study()</b>. This function was added to the
972 API for release 8.20. For earlier versions, the memory could be freed with
973 <b>pcre_free()</b>, just like the pattern itself. This will still work in cases
974 where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
975 new function when convenient.
976 </P>
977 <P>
978 This is a typical way in which <b>pcre_study</b>() is used (except that in a
979 real application there should be tests for errors):
980 <pre>
981 int rc;
982 pcre *re;
983 pcre_extra *sd;
984 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
985 sd = pcre_study(
986 re, /* result of pcre_compile() */
987 0, /* no options */
988 &error); /* set to NULL or points to a message */
989 rc = pcre_exec( /* see below for details of pcre_exec() options */
990 re, sd, "subject", 7, 0, 0, ovector, 30);
991 ...
992 pcre_free_study(sd);
993 pcre_free(re);
994 </pre>
995 Studying a pattern does two things: first, a lower bound for the length of
996 subject string that is needed to match the pattern is computed. This does not
997 mean that there are any strings of that length that match, but it does
998 guarantee that no shorter strings match. The value is used by
999 <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b> to avoid wasting time by trying to
1000 match strings that are shorter than the lower bound. You can find out the value
1001 in a calling program via the <b>pcre_fullinfo()</b> function.
1002 </P>
1003 <P>
1004 Studying a pattern is also useful for non-anchored patterns that do not have a
1005 single fixed starting character. A bitmap of possible starting bytes is
1006 created. This speeds up finding a position in the subject at which to start
1007 matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1008 </P>
1009 <P>
1010 These two optimizations apply to both <b>pcre_exec()</b> and
1011 <b>pcre_dfa_exec()</b>. However, they are not used by <b>pcre_exec()</b> if
1012 <b>pcre_study()</b> is called with the PCRE_STUDY_JIT_COMPILE option, and
1013 just-in-time compiling is successful. The optimizations can be disabled by
1014 setting the PCRE_NO_START_OPTIMIZE option when calling <b>pcre_exec()</b> or
1015 <b>pcre_dfa_exec()</b>. You might want to do this if your pattern contains
1016 callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
1017 to make use of these facilities in cases where matching fails. See the
1018 discussion of PCRE_NO_START_OPTIMIZE
1019 <a href="#execoptions">below.</a>
1020 <a name="localesupport"></a></P>
1021 <br><a name="SEC14" href="#TOC1">LOCALE SUPPORT</a><br>
1022 <P>
1023 PCRE handles caseless matching, and determines whether characters are letters,
1024 digits, or whatever, by reference to a set of tables, indexed by character
1025 value. When running in UTF-8 mode, this applies only to characters
1026 with codes less than 128. By default, higher-valued codes never match escapes
1027 such as \w or \d, but they can be tested with \p if PCRE is built with
1028 Unicode character property support. Alternatively, the PCRE_UCP option can be
1029 set at compile time; this causes \w and friends to use Unicode property
1030 support instead of built-in tables. The use of locales with Unicode is
1031 discouraged. If you are handling characters with codes greater than 128, you
1032 should either use UTF-8 and Unicode, or use locales, but not try to mix the
1033 two.
1034 </P>
1035 <P>
1036 PCRE contains an internal set of tables that are used when the final argument
1037 of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
1038 Normally, the internal tables recognize only ASCII characters. However, when
1039 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1040 default "C" locale of the local system, which may cause them to be different.
1041 </P>
1042 <P>
1043 The internal tables can always be overridden by tables supplied by the
1044 application that calls PCRE. These may be created in a different locale from
1045 the default. As more and more applications change to using Unicode, the need
1046 for this locale support is expected to die away.
1047 </P>
1048 <P>
1049 External tables are built by calling the <b>pcre_maketables()</b> function,
1050 which has no arguments, in the relevant locale. The result can then be passed
1051 to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
1052 example, to build and use tables that are appropriate for the French locale
1053 (where accented characters with values greater than 128 are treated as letters),
1054 the following code could be used:
1055 <pre>
1056 setlocale(LC_CTYPE, "fr_FR");
1057 tables = pcre_maketables();
1058 re = pcre_compile(..., tables);
1059 </pre>
1060 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1061 are using Windows, the name for the French locale is "french".
1062 </P>
1063 <P>
1064 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
1065 obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
1066 that the memory containing the tables remains available for as long as it is
1067 needed.
1068 </P>
1069 <P>
1070 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
1071 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
1072 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
1073 pattern, compilation, studying and matching all happen in the same locale, but
1074 different patterns can be compiled in different locales.
1075 </P>
1076 <P>
1077 It is possible to pass a table pointer or NULL (indicating the use of the
1078 internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
1079 this facility could be used to match a pattern in a different locale from the
1080 one in which it was compiled. Passing table pointers at run time is discussed
1081 below in the section on matching a pattern.
1082 <a name="infoaboutpattern"></a></P>
1083 <br><a name="SEC15" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
1084 <P>
1085 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1086 <b>int <i>what</i>, void *<i>where</i>);</b>
1087 </P>
1088 <P>
1089 The <b>pcre_fullinfo()</b> function returns information about a compiled
1090 pattern. It replaces the <b>pcre_info()</b> function, which was removed from the
1091 library at version 8.30, after more than 10 years of obsolescence.
1092 </P>
1093 <P>
1094 The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
1095 pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
1096 the pattern was not studied. The third argument specifies which piece of
1097 information is required, and the fourth argument is a pointer to a variable
1098 to receive the data. The yield of the function is zero for success, or one of
1099 the following negative numbers:
1100 <pre>
1101 PCRE_ERROR_NULL the argument <i>code</i> was NULL
1102 the argument <i>where</i> was NULL
1103 PCRE_ERROR_BADMAGIC the "magic number" was not found
1104 PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
1105 endianness
1106 PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
1107 </pre>
1108 The "magic number" is placed at the start of each compiled pattern as an simple
1109 check against passing an arbitrary memory pointer. The endianness error can
1110 occur if a compiled pattern is saved and reloaded on a different host. Here is
1111 a typical call of <b>pcre_fullinfo()</b>, to obtain the length of the compiled
1112 pattern:
1113 <pre>
1114 int rc;
1115 size_t length;
1116 rc = pcre_fullinfo(
1117 re, /* result of pcre_compile() */
1118 sd, /* result of pcre_study(), or NULL */
1119 PCRE_INFO_SIZE, /* what is required */
1120 &length); /* where to put the data */
1121 </pre>
1122 The possible values for the third argument are defined in <b>pcre.h</b>, and are
1123 as follows:
1124 <pre>
1126 </pre>
1127 Return the number of the highest back reference in the pattern. The fourth
1128 argument should point to an <b>int</b> variable. Zero is returned if there are
1129 no back references.
1130 <pre>
1132 </pre>
1133 Return the number of capturing subpatterns in the pattern. The fourth argument
1134 should point to an <b>int</b> variable.
1135 <pre>
1137 </pre>
1138 Return a pointer to the internal default character tables within PCRE. The
1139 fourth argument should point to an <b>unsigned char *</b> variable. This
1140 information call is provided for internal use by the <b>pcre_study()</b>
1141 function. External callers can cause PCRE to use its internal tables by passing
1142 a NULL table pointer.
1143 <pre>
1145 </pre>
1146 Return information about the first data unit of any matched string, for a
1147 non-anchored pattern. (The name of this option refers to the 8-bit library,
1148 where data units are bytes.) The fourth argument should point to an <b>int</b>
1149 variable.
1150 </P>
1151 <P>
1152 If there is a fixed first value, for example, the letter "c" from a pattern
1153 such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1154 value is always less than 256; in the 16-bit library the value can be up to
1155 0xffff.
1156 </P>
1157 <P>
1158 If there is no fixed first value, and if either
1159 <br>
1160 <br>
1161 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1162 starts with "^", or
1163 <br>
1164 <br>
1165 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1166 (if it were set, the pattern would be anchored),
1167 <br>
1168 <br>
1169 -1 is returned, indicating that the pattern matches only at the start of a
1170 subject string or after any newline within the string. Otherwise -2 is
1171 returned. For anchored patterns, -2 is returned.
1172 <pre>
1174 </pre>
1175 If the pattern was studied, and this resulted in the construction of a 256-bit
1176 table indicating a fixed set of values for the first data unit in any matching
1177 string, a pointer to the table is returned. Otherwise NULL is returned. The
1178 fourth argument should point to an <b>unsigned char *</b> variable.
1179 <pre>
1181 </pre>
1182 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1183 otherwise 0. The fourth argument should point to an <b>int</b> variable. An
1184 explicit match is either a literal CR or LF character, or \r or \n.
1185 <pre>
1187 </pre>
1188 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1189 0. The fourth argument should point to an <b>int</b> variable. (?J) and
1190 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1191 <pre>
1193 </pre>
1194 Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1195 just-in-time compiling was successful. The fourth argument should point to an
1196 <b>int</b> variable. A return value of 0 means that JIT support is not available
1197 in this version of PCRE, or that the pattern was not studied with the
1198 PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1199 particular pattern. See the
1200 <a href="pcrejit.html"><b>pcrejit</b></a>
1201 documentation for details of what can and cannot be handled.
1202 <pre>
1204 </pre>
1205 If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1206 return the size of the JIT compiled code, otherwise return zero. The fourth
1207 argument should point to a <b>size_t</b> variable.
1208 <pre>
1210 </pre>
1211 Return the value of the rightmost literal data unit that must exist in any
1212 matched string, other than at its start, if such a value has been recorded. The
1213 fourth argument should point to an <b>int</b> variable. If there is no such
1214 value, -1 is returned. For anchored patterns, a last literal value is recorded
1215 only if it follows something of variable length. For example, for the pattern
1216 /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
1217 is -1.
1218 <pre>
1220 </pre>
1221 If the pattern was studied and a minimum length for matching subject strings
1222 was computed, its value is returned. Otherwise the returned value is -1. The
1223 value is a number of characters, which in UTF-8 mode may be different from the
1224 number of bytes. The fourth argument should point to an <b>int</b> variable. A
1225 non-negative value is a lower bound to the length of any matching string. There
1226 may not be any strings of that length that do actually match, but every string
1227 that does match is at least that long.
1228 <pre>
1232 </pre>
1233 PCRE supports the use of named as well as numbered capturing parentheses. The
1234 names are just an additional way of identifying the parentheses, which still
1235 acquire numbers. Several convenience functions such as
1236 <b>pcre_get_named_substring()</b> are provided for extracting captured
1237 substrings by name. It is also possible to extract the data directly, by first
1238 converting the name to a number in order to access the correct pointers in the
1239 output vector (described with <b>pcre_exec()</b> below). To do the conversion,
1240 you need to use the name-to-number map, which is described by these three
1241 values.
1242 </P>
1243 <P>
1244 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1245 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1246 entry; both of these return an <b>int</b> value. The entry size depends on the
1247 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1248 entry of the table. This is a pointer to <b>char</b> in the 8-bit library, where
1249 the first two bytes of each entry are the number of the capturing parenthesis,
1250 most significant byte first. In the 16-bit library, the pointer points to
1251 16-bit data units, the first of which contains the parenthesis number. The rest
1252 of the entry is the corresponding name, zero terminated.
1253 </P>
1254 <P>
1255 The names are in alphabetical order. Duplicate names may appear if (?| is used
1256 to create multiple groups with the same number, as described in the
1257 <a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
1258 in the
1259 <a href="pcrepattern.html"><b>pcrepattern</b></a>
1260 page. Duplicate names for subpatterns with different numbers are permitted only
1261 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1262 table in the order in which they were found in the pattern. In the absence of
1263 (?| this is the order of increasing number; when (?| is used this is not
1264 necessarily the case because later subpatterns may have lower numbers.
1265 </P>
1266 <P>
1267 As a simple example of the name/number table, consider the following pattern
1268 after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1269 space - including newlines - is ignored):
1270 <pre>
1271 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
1272 </pre>
1273 There are four named subpatterns, so the table has four entries, and each entry
1274 in the table is eight bytes long. The table is as follows, with non-printing
1275 bytes shows in hexadecimal, and undefined bytes shown as ??:
1276 <pre>
1277 00 01 d a t e 00 ??
1278 00 05 d a y 00 ?? ??
1279 00 04 m o n t h 00
1280 00 02 y e a r 00 ??
1281 </pre>
1282 When writing code to extract data from named subpatterns using the
1283 name-to-number map, remember that the length of the entries is likely to be
1284 different for each compiled pattern.
1285 <pre>
1287 </pre>
1288 Return 1 if the pattern can be used for partial matching with
1289 <b>pcre_exec()</b>, otherwise 0. The fourth argument should point to an
1290 <b>int</b> variable. From release 8.00, this always returns 1, because the
1291 restrictions that previously applied to partial matching have been lifted. The
1292 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1293 documentation gives details of partial matching.
1294 <pre>
1296 </pre>
1297 Return a copy of the options with which the pattern was compiled. The fourth
1298 argument should point to an <b>unsigned long int</b> variable. These option bits
1299 are those specified in the call to <b>pcre_compile()</b>, modified by any
1300 top-level option settings at the start of the pattern itself. In other words,
1301 they are the options that will be in force when matching starts. For example,
1302 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1304 </P>
1305 <P>
1306 A pattern is automatically anchored by PCRE if all of its top-level
1307 alternatives begin with one of the following:
1308 <pre>
1309 ^ unless PCRE_MULTILINE is set
1310 \A always
1311 \G always
1312 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
1313 </pre>
1314 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1315 <b>pcre_fullinfo()</b>.
1316 <pre>
1318 </pre>
1319 Return the size of the compiled pattern in bytes (for both libraries). The
1320 fourth argument should point to a <b>size_t</b> variable. This value does not
1321 include the size of the <b>pcre</b> structure that is returned by
1322 <b>pcre_compile()</b>. The value that is passed as the argument to
1323 <b>pcre_malloc()</b> when <b>pcre_compile()</b> is getting memory in which to
1324 place the compiled data is the value returned by this option plus the size of
1325 the <b>pcre</b> structure. Studying a compiled pattern, with or without JIT,
1326 does not alter the value returned by this option.
1327 <pre>
1329 </pre>
1330 Return the size in bytes of the data block pointed to by the <i>study_data</i>
1331 field in a <b>pcre_extra</b> block. If <b>pcre_extra</b> is NULL, or there is no
1332 study data, zero is returned. The fourth argument should point to a
1333 <b>size_t</b> variable. The <i>study_data</i> field is set by <b>pcre_study()</b>
1334 to record information that will speed up matching (see the section entitled
1335 <a href="#studyingapattern">"Studying a pattern"</a>
1336 above). The format of the <i>study_data</i> block is private, but its length
1337 is made available via this option so that it can be saved and restored (see the
1338 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1339 documentation for details).
1340 </P>
1341 <br><a name="SEC16" href="#TOC1">REFERENCE COUNTS</a><br>
1342 <P>
1343 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1344 </P>
1345 <P>
1346 The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1347 data block that contains a compiled pattern. It is provided for the benefit of
1348 applications that operate in an object-oriented manner, where different parts
1349 of the application may be using the same compiled pattern, but you want to free
1350 the block when they are all done.
1351 </P>
1352 <P>
1353 When a pattern is compiled, the reference count field is initialized to zero.
1354 It is changed only by calling this function, whose action is to add the
1355 <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1356 function is the new value. However, the value of the count is constrained to
1357 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1358 it is forced to the appropriate limit value.
1359 </P>
1360 <P>
1361 Except when it is zero, the reference count is not correctly preserved if a
1362 pattern is compiled on one host and then transferred to a host whose byte-order
1363 is different. (This seems a highly unlikely scenario.)
1364 </P>
1365 <br><a name="SEC17" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1366 <P>
1367 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1368 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1369 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1370 </P>
1371 <P>
1372 The function <b>pcre_exec()</b> is called to match a subject string against a
1373 compiled pattern, which is passed in the <i>code</i> argument. If the
1374 pattern was studied, the result of the study should be passed in the
1375 <i>extra</i> argument. You can call <b>pcre_exec()</b> with the same <i>code</i>
1376 and <i>extra</i> arguments as many times as you like, in order to match
1377 different subject strings with the same pattern.
1378 </P>
1379 <P>
1380 This function is the main matching facility of the library, and it operates in
1381 a Perl-like manner. For specialist use there is also an alternative matching
1382 function, which is described
1383 <a href="#dfamatch">below</a>
1384 in the section about the <b>pcre_dfa_exec()</b> function.
1385 </P>
1386 <P>
1387 In most applications, the pattern will have been compiled (and optionally
1388 studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1389 possible to save compiled patterns and study data, and then use them later
1390 in different processes, possibly even on different hosts. For a discussion
1391 about this, see the
1392 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1393 documentation.
1394 </P>
1395 <P>
1396 Here is an example of a simple call to <b>pcre_exec()</b>:
1397 <pre>
1398 int rc;
1399 int ovector[30];
1400 rc = pcre_exec(
1401 re, /* result of pcre_compile() */
1402 NULL, /* we didn't study the pattern */
1403 "some string", /* the subject string */
1404 11, /* the length of the subject string */
1405 0, /* start at offset 0 in the subject */
1406 0, /* default options */
1407 ovector, /* vector of integers for substring information */
1408 30); /* number of elements (NOT size in bytes) */
1409 <a name="extradata"></a></PRE>
1410 </P>
1411 <br><b>
1412 Extra data for <b>pcre_exec()</b>
1413 </b><br>
1414 <P>
1415 If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1416 data block. The <b>pcre_study()</b> function returns such a block (when it
1417 doesn't return NULL), but you can also create one for yourself, and pass
1418 additional information in it. The <b>pcre_extra</b> block contains the following
1419 fields (not necessarily in this order):
1420 <pre>
1421 unsigned long int <i>flags</i>;
1422 void *<i>study_data</i>;
1423 void *<i>executable_jit</i>;
1424 unsigned long int <i>match_limit</i>;
1425 unsigned long int <i>match_limit_recursion</i>;
1426 void *<i>callout_data</i>;
1427 const unsigned char *<i>tables</i>;
1428 unsigned char **<i>mark</i>;
1429 </pre>
1430 In the 16-bit version of this structure, the <i>mark</i> field has type
1431 "PCRE_UCHAR16 **".
1432 </P>
1433 <P>
1434 The <i>flags</i> field is a bitmap that specifies which of the other fields
1435 are set. The flag bits are:
1436 <pre>
1444 </pre>
1445 Other flag bits should be set to zero. The <i>study_data</i> field and sometimes
1446 the <i>executable_jit</i> field are set in the <b>pcre_extra</b> block that is
1447 returned by <b>pcre_study()</b>, together with the appropriate flag bits. You
1448 should not set these yourself, but you may add to the block by setting the
1449 other fields and their corresponding flag bits.
1450 </P>
1451 <P>
1452 The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1453 vast amount of resources when running patterns that are not going to match,
1454 but which have a very large number of possibilities in their search trees. The
1455 classic example is a pattern that uses nested unlimited repeats.
1456 </P>
1457 <P>
1458 Internally, <b>pcre_exec()</b> uses a function called <b>match()</b>, which it
1459 calls repeatedly (sometimes recursively). The limit set by <i>match_limit</i> is
1460 imposed on the number of times this function is called during a match, which
1461 has the effect of limiting the amount of backtracking that can take place. For
1462 patterns that are not anchored, the count restarts from zero for each position
1463 in the subject string.
1464 </P>
1465 <P>
1466 When <b>pcre_exec()</b> is called with a pattern that was successfully studied
1467 with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1468 is entirely different. However, there is still the possibility of runaway
1469 matching that goes on for a very long time, and so the <i>match_limit</i> value
1470 is also used in this case (but in a different way) to limit how long the
1471 matching can continue.
1472 </P>
1473 <P>
1474 The default value for the limit can be set when PCRE is built; the default
1475 default is 10 million, which handles all but the most extreme cases. You can
1476 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1477 block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1478 the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1480 </P>
1481 <P>
1482 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1483 instead of limiting the total number of times that <b>match()</b> is called, it
1484 limits the depth of recursion. The recursion depth is a smaller number than the
1485 total number of calls, because not all calls to <b>match()</b> are recursive.
1486 This limit is of use only if it is set smaller than <i>match_limit</i>.
1487 </P>
1488 <P>
1489 Limiting the recursion depth limits the amount of machine stack that can be
1490 used, or, when PCRE has been compiled to use memory on the heap instead of the
1491 stack, the amount of heap memory that can be used. This limit is not relevant,
1492 and is ignored, if the pattern was successfully studied with
1494 </P>
1495 <P>
1496 The default value for <i>match_limit_recursion</i> can be set when PCRE is
1497 built; the default default is the same value as the default for
1498 <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1499 with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1500 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1501 is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1502 </P>
1503 <P>
1504 The <i>callout_data</i> field is used in conjunction with the "callout" feature,
1505 and is described in the
1506 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1507 documentation.
1508 </P>
1509 <P>
1510 The <i>tables</i> field is used to pass a character tables pointer to
1511 <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1512 pattern. A non-NULL value is stored with the compiled pattern only if custom
1513 tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1514 If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1515 internal tables to be used. This facility is helpful when re-using patterns
1516 that have been saved after compiling with an external set of tables, because
1517 the external tables might be at a different address when <b>pcre_exec()</b> is
1518 called. See the
1519 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1520 documentation for a discussion of saving compiled patterns for later use.
1521 </P>
1522 <P>
1523 If PCRE_EXTRA_MARK is set in the <i>flags</i> field, the <i>mark</i> field must
1524 be set to point to a suitable variable. If the pattern contains any
1525 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1526 a name to pass back, a pointer to the name string (zero terminated) is placed
1527 in the variable pointed to by the <i>mark</i> field. The names are within the
1528 compiled pattern; if you wish to retain such a name you must copy it before
1529 freeing the memory of a compiled pattern. If there is no name to pass back, the
1530 variable pointed to by the <i>mark</i> field is set to NULL. For details of the
1531 backtracking control verbs, see the section entitled
1532 <a href="pcrepattern#backtrackcontrol">"Backtracking control"</a>
1533 in the
1534 <a href="pcrepattern.html"><b>pcrepattern</b></a>
1535 documentation.
1536 <a name="execoptions"></a></P>
1537 <br><b>
1538 Option bits for <b>pcre_exec()</b>
1539 </b><br>
1540 <P>
1541 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1542 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1546 </P>
1547 <P>
1548 If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1549 the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1551 particular that partial matching is not supported. If an unsupported option is
1552 used, JIT execution is disabled and the normal interpretive code in
1553 <b>pcre_exec()</b> is run.
1554 <pre>
1556 </pre>
1557 The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1558 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1559 to be anchored by virtue of its contents, it cannot be made unachored at
1560 matching time.
1561 <pre>
1564 </pre>
1565 These options (which are mutually exclusive) control what the \R escape
1566 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1567 match any Unicode newline sequence. These options override the choice that was
1568 made or defaulted when the pattern was compiled.
1569 <pre>
1575 </pre>
1576 These options override the newline definition that was chosen or defaulted when
1577 the pattern was compiled. For details, see the description of
1578 <b>pcre_compile()</b> above. During matching, the newline choice affects the
1579 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1580 the way the match position is advanced after a match failure for an unanchored
1581 pattern.
1582 </P>
1583 <P>
1585 match attempt for an unanchored pattern fails when the current position is at a
1586 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1587 characters, the match position is advanced by two characters instead of one, in
1588 other words, to after the CRLF.
1589 </P>
1590 <P>
1591 The above rule is a compromise that makes the most common cases work as
1592 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1593 set), it does not match the string "\r\nA" because, after failing at the
1594 start, it skips both the CR and the LF before retrying. However, the pattern
1595 [\r\n]A does match that string, because it contains an explicit CR or LF
1596 reference, and so advances only by one character after the first failure.
1597 </P>
1598 <P>
1599 An explicit match for CR of LF is either a literal appearance of one of those
1600 characters, or one of the \r or \n escape sequences. Implicit matches such as
1601 [^X] do not count, nor does \s (which includes CR and LF in the characters
1602 that it matches).
1603 </P>
1604 <P>
1605 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1606 valid newline sequence and explicit \r or \n escapes appear in the pattern.
1607 <pre>
1609 </pre>
1610 This option specifies that first character of the subject string is not the
1611 beginning of a line, so the circumflex metacharacter should not match before
1612 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1613 never to match. This option affects only the behaviour of the circumflex
1614 metacharacter. It does not affect \A.
1615 <pre>
1617 </pre>
1618 This option specifies that the end of the subject string is not the end of a
1619 line, so the dollar metacharacter should not match it nor (except in multiline
1620 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1621 compile time) causes dollar never to match. This option affects only the
1622 behaviour of the dollar metacharacter. It does not affect \Z or \z.
1623 <pre>
1625 </pre>
1626 An empty string is not considered to be a valid match if this option is set. If
1627 there are alternatives in the pattern, they are tried. If all the alternatives
1628 match the empty string, the entire match fails. For example, if the pattern
1629 <pre>
1630 a?b?
1631 </pre>
1632 is applied to a string not beginning with "a" or "b", it matches an empty
1633 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1634 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1635 <pre>
1637 </pre>
1638 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1639 the start of the subject is permitted. If the pattern is anchored, such a match
1640 can occur only if the pattern contains \K.
1641 </P>
1642 <P>
1643 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1644 does make a special case of a pattern match of the empty string within its
1645 <b>split()</b> function, and when using the /g modifier. It is possible to
1646 emulate Perl's behaviour after matching a null string by first trying the match
1647 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1648 if that fails, by advancing the starting offset (see below) and trying an
1649 ordinary match again. There is some code that demonstrates how to do this in
1650 the
1651 <a href="pcredemo.html"><b>pcredemo</b></a>
1652 sample program. In the most general case, you have to check to see if the
1653 newline convention recognizes CRLF as a newline, and if so, and the current
1654 character is CR followed by LF, advance the starting offset by two characters
1655 instead of one.
1656 <pre>
1658 </pre>
1659 There are a number of optimizations that <b>pcre_exec()</b> uses at the start of
1660 a match, in order to speed up the process. For example, if it is known that an
1661 unanchored match must start with a specific character, it searches the subject
1662 for that character, and fails immediately if it cannot find it, without
1663 actually running the main matching function. This means that a special item
1664 such as (*COMMIT) at the start of a pattern is not considered until after a
1665 suitable starting point for the match has been found. When callouts or (*MARK)
1666 items are in use, these "start-up" optimizations can cause them to be skipped
1667 if the pattern is never actually used. The start-up optimizations are in effect
1668 a pre-scan of the subject that takes place before the pattern is run.
1669 </P>
1670 <P>
1671 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1672 causing performance to suffer, but ensuring that in cases where the result is
1673 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1674 are considered at every possible starting position in the subject string. If
1675 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1676 time.
1677 </P>
1678 <P>
1679 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1680 Consider the pattern
1681 <pre>
1683 </pre>
1684 When this is compiled, PCRE records the fact that a match must start with the
1685 character "A". Suppose the subject string is "DEFABC". The start-up
1686 optimization scans along the subject, finds "A" and runs the first match
1687 attempt from there. The (*COMMIT) item means that the pattern must match the
1688 current starting position, which in this case, it does. However, if the same
1689 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1690 subject string does not happen. The first match attempt is run starting from
1691 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1692 the overall result is "no match". If the pattern is studied, more start-up
1693 optimizations may be used. For example, a minimum length for the subject may be
1694 recorded. Consider the pattern
1695 <pre>
1696 (*MARK:A)(X|Y)
1697 </pre>
1698 The minimum length for a match is one character. If the subject is "ABC", there
1699 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1700 If the pattern is studied, the final attempt does not take place, because PCRE
1701 knows that the subject is too short, and so the (*MARK) is never encountered.
1702 In this case, studying the pattern does not affect the overall match result,
1703 which is still "no match", but it does affect the auxiliary information that is
1704 returned.
1705 <pre>
1707 </pre>
1708 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1709 string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1710 The value of <i>startoffset</i> is also checked to ensure that it points to the
1711 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1712 strings in the
1713 <a href="pcreunicode.html"><b>pcreunicode</b></a>
1714 page. If an invalid sequence of bytes is found, <b>pcre_exec()</b> returns the
1715 error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1716 truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1717 cases, information about the precise nature of the error may also be returned
1718 (see the descriptions of these errors in the section entitled \fIError return
1719 values from\fP <b>pcre_exec()</b>
1720 <a href="#errorlist">below).</a>
1721 If <i>startoffset</i> contains a value that does not point to the start of a
1722 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1723 returned.
1724 </P>
1725 <P>
1726 If you already know that your subject is valid, and you want to skip these
1727 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1728 calling <b>pcre_exec()</b>. You might want to do this for the second and
1729 subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1730 all the matches in a single subject string. However, you should be sure that
1731 the value of <i>startoffset</i> points to the start of a character (or the end
1732 of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1733 invalid string as a subject or an invalid value of <i>startoffset</i> is
1734 undefined. Your program may crash.
1735 <pre>
1738 </pre>
1739 These options turn on the partial matching feature. For backwards
1740 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1741 occurs if the end of the subject string is reached successfully, but there are
1742 not enough subject characters to complete the match. If this happens when
1743 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1744 testing any remaining alternatives. Only if no complete match can be found is
1745 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1746 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1747 but only if no complete match can be found.
1748 </P>
1749 <P>
1750 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1751 partial match is found, <b>pcre_exec()</b> immediately returns
1752 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1753 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1754 important that an alternative complete match.
1755 </P>
1756 <P>
1757 In both cases, the portion of the string that was inspected when the partial
1758 match was found is set as the first matching string. There is a more detailed
1759 discussion of partial and multi-segment matching, with examples, in the
1760 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1761 documentation.
1762 </P>
1763 <br><b>
1764 The string to be matched by <b>pcre_exec()</b>
1765 </b><br>
1766 <P>
1767 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1768 <i>subject</i>, a length in bytes in <i>length</i>, and a starting byte offset
1769 in <i>startoffset</i>. If this is negative or greater than the length of the
1770 subject, <b>pcre_exec()</b> returns PCRE_ERROR_BADOFFSET. When the starting
1771 offset is zero, the search for a match starts at the beginning of the subject,
1772 and this is by far the most common case. In UTF-8 mode, the byte offset must
1773 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1774 pattern string, the subject may contain binary zero bytes.
1775 </P>
1776 <P>
1777 A non-zero starting offset is useful when searching for another match in the
1778 same subject by calling <b>pcre_exec()</b> again after a previous success.
1779 Setting <i>startoffset</i> differs from just passing over a shortened string and
1780 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1781 lookbehind. For example, consider the pattern
1782 <pre>
1783 \Biss\B
1784 </pre>
1785 which finds occurrences of "iss" in the middle of words. (\B matches only if
1786 the current position in the subject is not a word boundary.) When applied to
1787 the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1788 occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1789 subject, namely "issipi", it does not match, because \B is always false at the
1790 start of the subject, which is deemed to be a word boundary. However, if
1791 <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1792 set to 4, it finds the second occurrence of "iss" because it is able to look
1793 behind the starting point to discover that it is preceded by a letter.
1794 </P>
1795 <P>
1796 Finding all the matches in a subject is tricky when the pattern can match an
1797 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1798 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1799 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1800 and trying an ordinary match again. There is some code that demonstrates how to
1801 do this in the
1802 <a href="pcredemo.html"><b>pcredemo</b></a>
1803 sample program. In the most general case, you have to check to see if the
1804 newline convention recognizes CRLF as a newline, and if so, and the current
1805 character is CR followed by LF, advance the starting offset by two characters
1806 instead of one.
1807 </P>
1808 <P>
1809 If a non-zero starting offset is passed when the pattern is anchored, one
1810 attempt to match at the given offset is made. This can only succeed if the
1811 pattern does not require the match to be at the start of the subject.
1812 </P>
1813 <br><b>
1814 How <b>pcre_exec()</b> returns captured substrings
1815 </b><br>
1816 <P>
1817 In general, a pattern matches a certain portion of the subject, and in
1818 addition, further substrings from the subject may be picked out by parts of the
1819 pattern. Following the usage in Jeffrey Friedl's book, this is called
1820 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1821 a fragment of a pattern that picks out a substring. PCRE supports several other
1822 kinds of parenthesized subpattern that do not cause substrings to be captured.
1823 </P>
1824 <P>
1825 Captured substrings are returned to the caller via a vector of integers whose
1826 address is passed in <i>ovector</i>. The number of elements in the vector is
1827 passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>: this
1828 argument is NOT the size of <i>ovector</i> in bytes.
1829 </P>
1830 <P>
1831 The first two-thirds of the vector is used to pass back captured substrings,
1832 each substring using a pair of integers. The remaining third of the vector is
1833 used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1834 and is not available for passing back information. The number passed in
1835 <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1836 rounded down.
1837 </P>
1838 <P>
1839 When a match is successful, information about captured substrings is returned
1840 in pairs of integers, starting at the beginning of <i>ovector</i>, and
1841 continuing up to two-thirds of its length at the most. The first element of
1842 each pair is set to the byte offset of the first character in a substring, and
1843 the second is set to the byte offset of the first character after the end of a
1844 substring. <b>Note</b>: these values are always byte offsets, even in UTF-8
1845 mode. They are not character counts.
1846 </P>
1847 <P>
1848 The first pair of integers, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the
1849 portion of the subject string matched by the entire pattern. The next pair is
1850 used for the first capturing subpattern, and so on. The value returned by
1851 <b>pcre_exec()</b> is one more than the highest numbered pair that has been set.
1852 For example, if two substrings have been captured, the returned value is 3. If
1853 there are no capturing subpatterns, the return value from a successful match is
1854 1, indicating that just the first pair of offsets has been set.
1855 </P>
1856 <P>
1857 If a capturing subpattern is matched repeatedly, it is the last portion of the
1858 string that it matched that is returned.
1859 </P>
1860 <P>
1861 If the vector is too small to hold all the captured substring offsets, it is
1862 used as far as possible (up to two-thirds of its length), and the function
1863 returns a value of zero. If neither the actual string matched not any captured
1864 substrings are of interest, <b>pcre_exec()</b> may be called with <i>ovector</i>
1865 passed as NULL and <i>ovecsize</i> as zero. However, if the pattern contains
1866 back references and the <i>ovector</i> is not big enough to remember the related
1867 substrings, PCRE has to get additional memory for use during matching. Thus it
1868 is usually advisable to supply an <i>ovector</i> of reasonable size.
1869 </P>
1870 <P>
1871 There are some cases where zero is returned (indicating vector overflow) when
1872 in fact the vector is exactly the right size for the final match. For example,
1873 consider the pattern
1874 <pre>
1875 (a)(?:(b)c|bd)
1876 </pre>
1877 If a vector of 6 elements (allowing for only 1 captured substring) is given
1878 with subject string "abd", <b>pcre_exec()</b> will try to set the second
1879 captured string, thereby recording a vector overflow, before failing to match
1880 "c" and backing up to try the second alternative. The zero return, however,
1881 does correctly indicate that the maximum number of slots (namely 2) have been
1882 filled. In similar cases where there is temporary overflow, but the final
1883 number of used slots is actually less than the maximum, a non-zero value is
1884 returned.
1885 </P>
1886 <P>
1887 The <b>pcre_fullinfo()</b> function can be used to find out how many capturing
1888 subpatterns there are in a compiled pattern. The smallest size for
1889 <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1890 the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1891 </P>
1892 <P>
1893 It is possible for capturing subpattern number <i>n+1</i> to match some part of
1894 the subject when subpattern <i>n</i> has not been used at all. For example, if
1895 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1896 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1897 happens, both values in the offset pairs corresponding to unused subpatterns
1898 are set to -1.
1899 </P>
1900 <P>
1901 Offset values that correspond to unused subpatterns at the end of the
1902 expression are also set to -1. For example, if the string "abc" is matched
1903 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1904 return from the function is 2, because the highest used capturing subpattern
1905 number is 1, and the offsets for for the second and third capturing subpatterns
1906 (assuming the vector is large enough, of course) are set to -1.
1907 </P>
1908 <P>
1909 <b>Note</b>: Elements in the first two-thirds of <i>ovector</i> that do not
1910 correspond to capturing parentheses in the pattern are never changed. That is,
1911 if a pattern contains <i>n</i> capturing parentheses, no more than
1912 <i>ovector[0]</i> to <i>ovector[2n+1]</i> are set by <b>pcre_exec()</b>. The other
1913 elements (in the first two-thirds) retain whatever values they previously had.
1914 </P>
1915 <P>
1916 Some convenience functions are provided for extracting the captured substrings
1917 as separate strings. These are described below.
1918 <a name="errorlist"></a></P>
1919 <br><b>
1920 Error return values from <b>pcre_exec()</b>
1921 </b><br>
1922 <P>
1923 If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1924 defined in the header file:
1925 <pre>
1927 </pre>
1928 The subject string did not match the pattern.
1929 <pre>
1931 </pre>
1932 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1933 NULL and <i>ovecsize</i> was not zero.
1934 <pre>
1936 </pre>
1937 An unrecognized bit was set in the <i>options</i> argument.
1938 <pre>
1940 </pre>
1941 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1942 the case when it is passed a junk pointer and to detect when a pattern that was
1943 compiled in an environment of one endianness is run in an environment with the
1944 other endianness. This is the error that PCRE gives when the magic number is
1945 not present.
1946 <pre>
1948 </pre>
1949 While running the pattern match, an unknown item was encountered in the
1950 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1951 of the compiled pattern.
1952 <pre>
1954 </pre>
1955 If a pattern contains back references, but the <i>ovector</i> that is passed to
1956 <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1957 gets a block of memory at the start of matching to use for this purpose. If the
1958 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1959 automatically freed at the end of matching.
1960 </P>
1961 <P>
1962 This error is also given if <b>pcre_stack_malloc()</b> fails in
1963 <b>pcre_exec()</b>. This can happen only when PCRE has been compiled with
1964 <b>--disable-stack-for-recursion</b>.
1965 <pre>
1967 </pre>
1968 This error is used by the <b>pcre_copy_substring()</b>,
1969 <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1970 below). It is never returned by <b>pcre_exec()</b>.
1971 <pre>
1973 </pre>
1974 The backtracking limit, as specified by the <i>match_limit</i> field in a
1975 <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1976 above.
1977 <pre>
1979 </pre>
1980 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1981 use by callout functions that want to yield a distinctive error code. See the
1982 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1983 documentation for details.
1984 <pre>
1986 </pre>
1987 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1988 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1989 (<i>ovecsize</i>) is at least 2, the byte offset to the start of the the invalid
1990 UTF-8 character is placed in the first element, and a reason code is placed in
1991 the second element. The reason codes are listed in the
1992 <a href="#badutf8reasons">following section.</a>
1993 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1994 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1995 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1996 <pre>
1998 </pre>
1999 The UTF-8 byte sequence that was passed as a subject was checked and found to
2000 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
2001 <i>startoffset</i> did not point to the beginning of a UTF-8 character or the
2002 end of the subject.
2003 <pre>
2005 </pre>
2006 The subject string did not match, but it did match partially. See the
2007 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2008 documentation for details of partial matching.
2009 <pre>
2011 </pre>
2012 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2013 option was used with a compiled pattern containing items that were not
2014 supported for partial matching. From release 8.00 onwards, there are no
2015 restrictions on partial matching.
2016 <pre>
2018 </pre>
2019 An unexpected internal error has occurred. This error could be caused by a bug
2020 in PCRE or by overwriting of the compiled pattern.
2021 <pre>
2023 </pre>
2024 This error is given if the value of the <i>ovecsize</i> argument is negative.
2025 <pre>
2027 </pre>
2028 The internal recursion limit, as specified by the <i>match_limit_recursion</i>
2029 field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
2030 description above.
2031 <pre>
2033 </pre>
2034 An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
2035 <pre>
2037 </pre>
2038 The value of <i>startoffset</i> was negative or greater than the length of the
2039 subject, that is, the value in <i>length</i>.
2040 <pre>
2042 </pre>
2043 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2044 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2045 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2046 fact sufficient to detect this case, but this special error code for
2047 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2048 retained for backwards compatibility.
2049 <pre>
2051 </pre>
2052 This error is returned when <b>pcre_exec()</b> detects a recursion loop within
2053 the pattern. Specifically, it means that either the whole pattern or a
2054 subpattern has been called recursively for the second time at the same position
2055 in the subject string. Some simple patterns that might do this are detected and
2056 faulted at compile time, but more complicated cases, in particular mutual
2057 recursions between two different subpatterns, cannot be detected until run
2058 time.
2059 <pre>
2061 </pre>
2062 This error is returned when a pattern that was successfully studied using the
2063 PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2064 the just-in-time processing stack is not large enough. See the
2065 <a href="pcrejit.html"><b>pcrejit</b></a>
2066 documentation for more details.
2067 <pre>
2069 </pre>
2070 This error is given if a pattern that was compiled by the 8-bit library is
2071 passed to a 16-bit library function, or vice versa.
2072 <pre>
2074 </pre>
2075 This error is given if a pattern that was compiled and saved is reloaded on a
2076 host with different endianness. The utility function
2077 <b>pcre_pattern_to_host_byte_order()</b> can be used to convert such a pattern
2078 so that it runs on the new host.
2079 </P>
2080 <P>
2081 Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
2082 <a name="badutf8reasons"></a></P>
2083 <br><b>
2084 Reason codes for invalid UTF-8 strings
2085 </b><br>
2086 <P>
2087 This section applies only to the 8-bit library. The corresponding information
2088 for the 16-bit library is given in the
2089 <a href="pcre16.html"><b>pcre16</b></a>
2090 page.
2091 </P>
2092 <P>
2093 When <b>pcre_exec()</b> returns either PCRE_ERROR_BADUTF8 or
2094 PCRE_ERROR_SHORTUTF8, and the size of the output vector (<i>ovecsize</i>) is at
2095 least 2, the offset of the start of the invalid UTF-8 character is placed in
2096 the first output vector element (<i>ovector[0]</i>) and a reason code is placed
2097 in the second element (<i>ovector[1]</i>). The reason codes are given names in
2098 the <b>pcre.h</b> header file:
2099 <pre>
2105 </pre>
2106 The string ends with a truncated UTF-8 character; the code specifies how many
2107 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2108 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2109 allows for up to 6 bytes, and this is checked first; hence the possibility of
2110 4 or 5 missing bytes.
2111 <pre>
2116 PCRE_UTF8_ERR10
2117 </pre>
2118 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2119 character do not have the binary value 0b10 (that is, either the most
2120 significant bit is 0, or the next bit is 1).
2121 <pre>
2122 PCRE_UTF8_ERR11
2123 PCRE_UTF8_ERR12
2124 </pre>
2125 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2126 these code points are excluded by RFC 3629.
2127 <pre>
2128 PCRE_UTF8_ERR13
2129 </pre>
2130 A 4-byte character has a value greater than 0x10fff; these code points are
2131 excluded by RFC 3629.
2132 <pre>
2133 PCRE_UTF8_ERR14
2134 </pre>
2135 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2136 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2137 from UTF-8.
2138 <pre>
2139 PCRE_UTF8_ERR15
2140 PCRE_UTF8_ERR16
2141 PCRE_UTF8_ERR17
2142 PCRE_UTF8_ERR18
2143 PCRE_UTF8_ERR19
2144 </pre>
2145 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2146 value that can be represented by fewer bytes, which is invalid. For example,
2147 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2148 one byte.
2149 <pre>
2150 PCRE_UTF8_ERR20
2151 </pre>
2152 The two most significant bits of the first byte of a character have the binary
2153 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2154 byte can only validly occur as the second or subsequent byte of a multi-byte
2155 character.
2156 <pre>
2157 PCRE_UTF8_ERR21
2158 </pre>
2159 The first byte of a character has the value 0xfe or 0xff. These values can
2160 never occur in a valid UTF-8 string.
2161 </P>
2162 <br><a name="SEC18" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
2163 <P>
2164 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
2165 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
2166 <b>int <i>buffersize</i>);</b>
2167 </P>
2168 <P>
2169 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
2170 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
2171 <b>const char **<i>stringptr</i>);</b>
2172 </P>
2173 <P>
2174 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
2175 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
2176 </P>
2177 <P>
2178 Captured substrings can be accessed directly by using the offsets returned by
2179 <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
2180 <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
2181 <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
2182 as new, separate, zero-terminated strings. These functions identify substrings
2183 by number. The next section describes functions for extracting named
2184 substrings.
2185 </P>
2186 <P>
2187 A substring that contains a binary zero is correctly extracted and has a
2188 further zero added on the end, but the result is not, of course, a C string.
2189 However, you can process such a string by referring to the length that is
2190 returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
2191 Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
2192 for handling strings containing binary zeros, because the end of the final
2193 string is not independently indicated.
2194 </P>
2195 <P>
2196 The first three arguments are the same for all three of these functions:
2197 <i>subject</i> is the subject string that has just been successfully matched,
2198 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
2199 <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
2200 captured by the match, including the substring that matched the entire regular
2201 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
2202 than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
2203 space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
2204 number of elements in the vector divided by three.
2205 </P>
2206 <P>
2207 The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
2208 extract a single substring, whose number is given as <i>stringnumber</i>. A
2209 value of zero extracts the substring that matched the entire pattern, whereas
2210 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
2211 the string is placed in <i>buffer</i>, whose length is given by
2212 <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
2213 obtained via <b>pcre_malloc</b>, and its address is returned via
2214 <i>stringptr</i>. The yield of the function is the length of the string, not
2215 including the terminating zero, or one of these error codes:
2216 <pre>
2218 </pre>
2219 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
2220 memory failed for <b>pcre_get_substring()</b>.
2221 <pre>
2223 </pre>
2224 There is no substring whose number is <i>stringnumber</i>.
2225 </P>
2226 <P>
2227 The <b>pcre_get_substring_list()</b> function extracts all available substrings
2228 and builds a list of pointers to them. All this is done in a single block of
2229 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
2230 is returned via <i>listptr</i>, which is also the start of the list of string
2231 pointers. The end of the list is marked by a NULL pointer. The yield of the
2232 function is zero if all went well, or the error code
2233 <pre>
2235 </pre>
2236 if the attempt to get the memory block failed.
2237 </P>
2238 <P>
2239 When any of these functions encounter a substring that is unset, which can
2240 happen when capturing subpattern number <i>n+1</i> matches some part of the
2241 subject, but subpattern <i>n</i> has not been used at all, they return an empty
2242 string. This can be distinguished from a genuine zero-length substring by
2243 inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
2244 substrings.
2245 </P>
2246 <P>
2247 The two convenience functions <b>pcre_free_substring()</b> and
2248 <b>pcre_free_substring_list()</b> can be used to free the memory returned by
2249 a previous call of <b>pcre_get_substring()</b> or
2250 <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
2251 the function pointed to by <b>pcre_free</b>, which of course could be called
2252 directly from a C program. However, PCRE is used in some situations where it is
2253 linked via a special interface to another programming language that cannot use
2254 <b>pcre_free</b> directly; it is for these cases that the functions are
2255 provided.
2256 </P>
2257 <br><a name="SEC19" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
2258 <P>
2259 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
2260 <b>const char *<i>name</i>);</b>
2261 </P>
2262 <P>
2263 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
2264 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
2265 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
2266 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
2267 </P>
2268 <P>
2269 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
2270 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
2271 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
2272 <b>const char **<i>stringptr</i>);</b>
2273 </P>
2274 <P>
2275 To extract a substring by name, you first have to find associated number.
2276 For example, for this pattern
2277 <pre>
2278 (a+)b(?&#60;xxx&#62;\d+)...
2279 </pre>
2280 the number of the subpattern called "xxx" is 2. If the name is known to be
2281 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2282 calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
2283 pattern, and the second is the name. The yield of the function is the
2284 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2285 that name.
2286 </P>
2287 <P>
2288 Given the number, you can extract the substring directly, or use one of the
2289 functions described in the previous section. For convenience, there are also
2290 two functions that do the whole job.
2291 </P>
2292 <P>
2293 Most of the arguments of <b>pcre_copy_named_substring()</b> and
2294 <b>pcre_get_named_substring()</b> are the same as those for the similarly named
2295 functions that extract by number. As these are described in the previous
2296 section, they are not re-described here. There are just two differences:
2297 </P>
2298 <P>
2299 First, instead of a substring number, a substring name is given. Second, there
2300 is an extra argument, given at the start, which is a pointer to the compiled
2301 pattern. This is needed in order to gain access to the name-to-number
2302 translation table.
2303 </P>
2304 <P>
2305 These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
2306 then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
2307 appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
2308 the behaviour may not be what you want (see the next section).
2309 </P>
2310 <P>
2311 <b>Warning:</b> If the pattern uses the (?| feature to set up multiple
2312 subpatterns with the same number, as described in the
2313 <a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
2314 in the
2315 <a href="pcrepattern.html"><b>pcrepattern</b></a>
2316 page, you cannot use names to distinguish the different subpatterns, because
2317 names are not included in the compiled code. The matching process uses only
2318 numbers. For this reason, the use of different names for subpatterns of the
2319 same number causes an error at compile time.
2320 </P>
2321 <br><a name="SEC20" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
2322 <P>
2323 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
2324 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
2325 </P>
2326 <P>
2327 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2328 are not required to be unique. (Duplicate names are always allowed for
2329 subpatterns with the same number, created by using the (?| feature. Indeed, if
2330 such subpatterns are named, they are required to use the same names.)
2331 </P>
2332 <P>
2333 Normally, patterns with duplicate names are such that in any one match, only
2334 one of the named subpatterns participates. An example is shown in the
2335 <a href="pcrepattern.html"><b>pcrepattern</b></a>
2336 documentation.
2337 </P>
2338 <P>
2339 When duplicates are present, <b>pcre_copy_named_substring()</b> and
2340 <b>pcre_get_named_substring()</b> return the first substring corresponding to
2341 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2342 returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
2343 returns one of the numbers that are associated with the name, but it is not
2344 defined which it is.
2345 </P>
2346 <P>
2347 If you want to get full details of all captured substrings for a given name,
2348 you must use the <b>pcre_get_stringtable_entries()</b> function. The first
2349 argument is the compiled pattern, and the second is the name. The third and
2350 fourth are pointers to variables which are updated by the function. After it
2351 has run, they point to the first and last entries in the name-to-number table
2352 for the given name. The function itself returns the length of each entry, or
2353 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2354 described above in the section entitled <i>Information about a pattern</i>
2355 <a href="#infoaboutpattern">above.</a>
2356 Given all the relevant entries for the name, you can extract each of their
2357 numbers, and hence the captured data, if any.
2358 </P>
2359 <br><a name="SEC21" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
2360 <P>
2361 The traditional matching function uses a similar algorithm to Perl, which stops
2362 when it finds the first match, starting at a given point in the subject. If you
2363 want to find all possible matches, or the longest possible match, consider
2364 using the alternative matching function (see below) instead. If you cannot use
2365 the alternative function, but still need to find all possible matches, you
2366 can kludge it up by making use of the callout facility, which is described in
2367 the
2368 <a href="pcrecallout.html"><b>pcrecallout</b></a>
2369 documentation.
2370 </P>
2371 <P>
2372 What you have to do is to insert a callout right at the end of the pattern.
2373 When your callout function is called, extract and save the current matched
2374 substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
2375 other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
2376 will yield PCRE_ERROR_NOMATCH.
2377 <a name="dfamatch"></a></P>
2378 <br><a name="SEC22" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
2379 <P>
2380 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
2381 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
2382 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
2383 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
2384 </P>
2385 <P>
2386 The function <b>pcre_dfa_exec()</b> is called to match a subject string against
2387 a compiled pattern, using a matching algorithm that scans the subject string
2388 just once, and does not backtrack. This has different characteristics to the
2389 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2390 patterns are not supported. Nevertheless, there are times when this kind of
2391 matching can be useful. For a discussion of the two matching algorithms, and a
2392 list of features that <b>pcre_dfa_exec()</b> does not support, see the
2393 <a href="pcrematching.html"><b>pcrematching</b></a>
2394 documentation.
2395 </P>
2396 <P>
2397 The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
2398 <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
2399 different way, and this is described below. The other common arguments are used
2400 in the same way as for <b>pcre_exec()</b>, so their description is not repeated
2401 here.
2402 </P>
2403 <P>
2404 The two additional arguments provide workspace for the function. The workspace
2405 vector should contain at least 20 elements. It is used for keeping track of
2406 multiple paths through the pattern tree. More workspace will be needed for
2407 patterns and subjects where there are a lot of potential matches.
2408 </P>
2409 <P>
2410 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
2411 <pre>
2412 int rc;
2413 int ovector[10];
2414 int wspace[20];
2415 rc = pcre_dfa_exec(
2416 re, /* result of pcre_compile() */
2417 NULL, /* we didn't study the pattern */
2418 "some string", /* the subject string */
2419 11, /* the length of the subject string */
2420 0, /* start at offset 0 in the subject */
2421 0, /* default options */
2422 ovector, /* vector of integers for substring information */
2423 10, /* number of elements (NOT size in bytes) */
2424 wspace, /* working space vector */
2425 20); /* number of elements (NOT size in bytes) */
2426 </PRE>
2427 </P>
2428 <br><b>
2429 Option bits for <b>pcre_dfa_exec()</b>
2430 </b><br>
2431 <P>
2432 The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
2433 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
2437 All but the last four of these are exactly the same as for <b>pcre_exec()</b>,
2438 so their description is not repeated here.
2439 <pre>
2442 </pre>
2443 These have the same general effect as they do for <b>pcre_exec()</b>, but the
2444 details are slightly different. When PCRE_PARTIAL_HARD is set for
2445 <b>pcre_dfa_exec()</b>, it returns PCRE_ERROR_PARTIAL if the end of the subject
2446 is reached and there is still at least one matching possibility that requires
2447 additional characters. This happens even if some complete matches have also
2448 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2449 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2450 there have been no complete matches, but there is still at least one matching
2451 possibility. The portion of the string that was inspected when the longest
2452 partial match was found is set as the first matching string in both cases.
2453 There is a more detailed discussion of partial and multi-segment matching, with
2454 examples, in the
2455 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2456 documentation.
2457 <pre>
2459 </pre>
2460 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2461 soon as it has found one match. Because of the way the alternative algorithm
2462 works, this is necessarily the shortest possible match at the first possible
2463 matching point in the subject string.
2464 <pre>
2466 </pre>
2467 When <b>pcre_dfa_exec()</b> returns a partial match, it is possible to call it
2468 again, with additional subject characters, and have it continue with the same
2469 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2470 <i>workspace</i> and <i>wscount</i> options must reference the same vector as
2471 before because data about the match so far is left in them after a partial
2472 match. There is more discussion of this facility in the
2473 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2474 documentation.
2475 </P>
2476 <br><b>
2477 Successful returns from <b>pcre_dfa_exec()</b>
2478 </b><br>
2479 <P>
2480 When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
2481 substring in the subject. Note, however, that all the matches from one run of
2482 the function start at the same point in the subject. The shorter matches are
2483 all initial substrings of the longer matches. For example, if the pattern
2484 <pre>
2485 &#60;.*&#62;
2486 </pre>
2487 is matched against the string
2488 <pre>
2489 This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
2490 </pre>
2491 the three matched strings are
2492 <pre>
2493 &#60;something&#62;
2494 &#60;something&#62; &#60;something else&#62;
2495 &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
2496 </pre>
2497 On success, the yield of the function is a number greater than zero, which is
2498 the number of matched substrings. The substrings themselves are returned in
2499 <i>ovector</i>. Each string uses two elements; the first is the offset to the
2500 start, and the second is the offset to the end. In fact, all the strings have
2501 the same start offset. (Space could have been saved by giving this only once,
2502 but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
2503 returns data, even though the meaning of the strings is different.)
2504 </P>
2505 <P>
2506 The strings are returned in reverse order of length; that is, the longest
2507 matching string is given first. If there were too many matches to fit into
2508 <i>ovector</i>, the yield of the function is zero, and the vector is filled with
2509 the longest matches. Unlike <b>pcre_exec()</b>, <b>pcre_dfa_exec()</b> can use
2510 the entire <i>ovector</i> for returning matched strings.
2511 </P>
2512 <br><b>
2513 Error returns from <b>pcre_dfa_exec()</b>
2514 </b><br>
2515 <P>
2516 The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
2517 Many of the errors are the same as for <b>pcre_exec()</b>, and these are
2518 described
2519 <a href="#errorlist">above.</a>
2520 There are in addition the following errors that are specific to
2521 <b>pcre_dfa_exec()</b>:
2522 <pre>
2524 </pre>
2525 This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
2526 that it does not support, for instance, the use of \C or a back reference.
2527 <pre>
2529 </pre>
2530 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
2531 uses a back reference for the condition, or a test for recursion in a specific
2532 group. These are not supported.
2533 <pre>
2535 </pre>
2536 This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
2537 block that contains a setting of the <i>match_limit</i> or
2538 <i>match_limit_recursion</i> fields. This is not supported (these fields are
2539 meaningless for DFA matching).
2540 <pre>
2542 </pre>
2543 This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
2544 <i>workspace</i> vector.
2545 <pre>
2547 </pre>
2548 When a recursive subpattern is processed, the matching function calls itself
2549 recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
2550 error is given if the output vector is not large enough. This should be
2551 extremely rare, as a vector of size 1000 is used.
2552 </P>
2553 <br><a name="SEC23" href="#TOC1">SEE ALSO</a><br>
2554 <P>
2555 <b>pcre16</b>(3), <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
2556 <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
2557 <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
2558 </P>
2559 <br><a name="SEC24" href="#TOC1">AUTHOR</a><br>
2560 <P>
2561 Philip Hazel
2562 <br>
2563 University Computing Service
2564 <br>
2565 Cambridge CB2 3QH, England.
2566 <br>
2567 </P>
2568 <br><a name="SEC25" href="#TOC1">REVISION</a><br>
2569 <P>
2570 Last updated: 07 January 2012
2571 <br>
2572 Copyright &copy; 1997-2012 University of Cambridge.
2573 <br>
2574 <p>
2575 Return to the <a href="index.html">PCRE index page</a>.
2576 </p>


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

  ViewVC Help
Powered by ViewVC 1.1.5