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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 959 - (show annotations) (download) (as text)
Sat Apr 14 16:16:58 2012 UTC (2 years, 5 months ago) by ph10
File MIME type: text/html
File size: 121082 byte(s)
Update documentation to clarify that UTF-8/16 checking is done on complete 
strings before any other processing.

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>
17 <li><a name="TOC2" href="#SEC2">PCRE NATIVE API STRING EXTRACTION FUNCTIONS</a>
18 <li><a name="TOC3" href="#SEC3">PCRE NATIVE API AUXILIARY FUNCTIONS</a>
19 <li><a name="TOC4" href="#SEC4">PCRE NATIVE API INDIRECTED 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>
24 <li><a name="TOC9" href="#SEC9">SAVING PRECOMPILED PATTERNS FOR LATER USE</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>
32 <li><a name="TOC17" href="#SEC17">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
33 <li><a name="TOC18" href="#SEC18">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
34 <li><a name="TOC19" href="#SEC19">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
35 <li><a name="TOC20" href="#SEC20">DUPLICATE SUBPATTERN NAMES</a>
36 <li><a name="TOC21" href="#SEC21">FINDING ALL POSSIBLE MATCHES</a>
37 <li><a name="TOC22" href="#SEC22">OBTAINING AN ESTIMATE OF STACK USAGE</a>
38 <li><a name="TOC23" href="#SEC23">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
39 <li><a name="TOC24" href="#SEC24">SEE ALSO</a>
40 <li><a name="TOC25" href="#SEC25">AUTHOR</a>
41 <li><a name="TOC26" href="#SEC26">REVISION</a>
42 </ul>
43 <P>
44 <b>#include &#60;pcre.h&#62;</b>
45 </P>
46 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API BASIC FUNCTIONS</a><br>
47 <P>
48 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
49 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
50 <b>const unsigned char *<i>tableptr</i>);</b>
51 </P>
52 <P>
53 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
54 <b>int *<i>errorcodeptr</i>,</b>
55 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
56 <b>const unsigned char *<i>tableptr</i>);</b>
57 </P>
58 <P>
59 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
60 <b>const char **<i>errptr</i>);</b>
61 </P>
62 <P>
63 <b>void pcre_free_study(pcre_extra *<i>extra</i>);</b>
64 </P>
65 <P>
66 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
67 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
68 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
69 </P>
70 <P>
71 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
72 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
73 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
74 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
75 </P>
76 <br><a name="SEC2" href="#TOC1">PCRE NATIVE API STRING EXTRACTION FUNCTIONS</a><br>
77 <P>
78 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
79 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
80 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
81 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
82 </P>
83 <P>
84 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
85 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
86 <b>int <i>buffersize</i>);</b>
87 </P>
88 <P>
89 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
90 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
91 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
92 <b>const char **<i>stringptr</i>);</b>
93 </P>
94 <P>
95 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
96 <b>const char *<i>name</i>);</b>
97 </P>
98 <P>
99 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
100 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
101 </P>
102 <P>
103 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
104 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
105 <b>const char **<i>stringptr</i>);</b>
106 </P>
107 <P>
108 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
109 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
110 </P>
111 <P>
112 <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
113 </P>
114 <P>
115 <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
116 </P>
117 <br><a name="SEC3" href="#TOC1">PCRE NATIVE API AUXILIARY FUNCTIONS</a><br>
118 <P>
119 <b>pcre_jit_stack *pcre_jit_stack_alloc(int <i>startsize</i>, int <i>maxsize</i>);</b>
120 </P>
121 <P>
122 <b>void pcre_jit_stack_free(pcre_jit_stack *<i>stack</i>);</b>
123 </P>
124 <P>
125 <b>void pcre_assign_jit_stack(pcre_extra *<i>extra</i>,</b>
126 <b>pcre_jit_callback <i>callback</i>, void *<i>data</i>);</b>
127 </P>
128 <P>
129 <b>const unsigned char *pcre_maketables(void);</b>
130 </P>
131 <P>
132 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
133 <b>int <i>what</i>, void *<i>where</i>);</b>
134 </P>
135 <P>
136 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
137 </P>
138 <P>
139 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
140 </P>
141 <P>
142 <b>const char *pcre_version(void);</b>
143 </P>
144 <P>
145 <b>int pcre_pattern_to_host_byte_order(pcre *<i>code</i>,</b>
146 <b>pcre_extra *<i>extra</i>, const unsigned char *<i>tables</i>);</b>
147 </P>
148 <br><a name="SEC4" href="#TOC1">PCRE NATIVE API INDIRECTED FUNCTIONS</a><br>
149 <P>
150 <b>void *(*pcre_malloc)(size_t);</b>
151 </P>
152 <P>
153 <b>void (*pcre_free)(void *);</b>
154 </P>
155 <P>
156 <b>void *(*pcre_stack_malloc)(size_t);</b>
157 </P>
158 <P>
159 <b>void (*pcre_stack_free)(void *);</b>
160 </P>
161 <P>
162 <b>int (*pcre_callout)(pcre_callout_block *);</b>
163 </P>
164 <br><a name="SEC5" href="#TOC1">PCRE 8-BIT AND 16-BIT LIBRARIES</a><br>
165 <P>
166 From release 8.30, PCRE can be compiled as a library for handling 16-bit
167 character strings as well as, or instead of, the original library that handles
168 8-bit character strings. To avoid too much complication, this document
169 describes the 8-bit versions of the functions, with only occasional references
170 to the 16-bit library.
171 </P>
172 <P>
173 The 16-bit functions operate in the same way as their 8-bit counterparts; they
174 just use different data types for their arguments and results, and their names
175 start with <b>pcre16_</b> instead of <b>pcre_</b>. For every option that has UTF8
176 in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
177 UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
178 option names define the same bit values.
179 </P>
180 <P>
181 References to bytes and UTF-8 in this document should be read as references to
182 16-bit data quantities and UTF-16 when using the 16-bit library, unless
183 specified otherwise. More details of the specific differences for the 16-bit
184 library are given in the
185 <a href="pcre16.html"><b>pcre16</b></a>
186 page.
187 </P>
188 <br><a name="SEC6" href="#TOC1">PCRE API OVERVIEW</a><br>
189 <P>
190 PCRE has its own native API, which is described in this document. There are
191 also some wrapper functions (for the 8-bit library only) that correspond to the
192 POSIX regular expression API, but they do not give access to all the
193 functionality. They are described in the
194 <a href="pcreposix.html"><b>pcreposix</b></a>
195 documentation. Both of these APIs define a set of C function calls. A C++
196 wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
197 documented in the
198 <a href="pcrecpp.html"><b>pcrecpp</b></a>
199 page.
200 </P>
201 <P>
202 The native API C function prototypes are defined in the header file
203 <b>pcre.h</b>, and on Unix-like systems the (8-bit) library itself is called
204 <b>libpcre</b>. It can normally be accessed by adding <b>-lpcre</b> to the
205 command for linking an application that uses PCRE. The header file defines the
206 macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
207 for the library. Applications can use these to include support for different
208 releases of PCRE.
209 </P>
210 <P>
211 In a Windows environment, if you want to statically link an application program
212 against a non-dll <b>pcre.a</b> file, you must define PCRE_STATIC before
213 including <b>pcre.h</b> or <b>pcrecpp.h</b>, because otherwise the
214 <b>pcre_malloc()</b> and <b>pcre_free()</b> exported functions will be declared
215 <b>__declspec(dllimport)</b>, with unwanted results.
216 </P>
217 <P>
218 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
219 and <b>pcre_exec()</b> are used for compiling and matching regular expressions
220 in a Perl-compatible manner. A sample program that demonstrates the simplest
221 way of using them is provided in the file called <i>pcredemo.c</i> in the PCRE
222 source distribution. A listing of this program is given in the
223 <a href="pcredemo.html"><b>pcredemo</b></a>
224 documentation, and the
225 <a href="pcresample.html"><b>pcresample</b></a>
226 documentation describes how to compile and run it.
227 </P>
228 <P>
229 Just-in-time compiler support is an optional feature of PCRE that can be built
230 in appropriate hardware environments. It greatly speeds up the matching
231 performance of many patterns. Simple programs can easily request that it be
232 used if available, by setting an option that is ignored when it is not
233 relevant. More complicated programs might need to make use of the functions
234 <b>pcre_jit_stack_alloc()</b>, <b>pcre_jit_stack_free()</b>, and
235 <b>pcre_assign_jit_stack()</b> in order to control the JIT code's memory usage.
236 These functions are discussed in the
237 <a href="pcrejit.html"><b>pcrejit</b></a>
238 documentation.
239 </P>
240 <P>
241 A second matching function, <b>pcre_dfa_exec()</b>, which is not
242 Perl-compatible, is also provided. This uses a different algorithm for the
243 matching. The alternative algorithm finds all possible matches (at a given
244 point in the subject), and scans the subject just once (unless there are
245 lookbehind assertions). However, this algorithm does not return captured
246 substrings. A description of the two matching algorithms and their advantages
247 and disadvantages is given in the
248 <a href="pcrematching.html"><b>pcrematching</b></a>
249 documentation.
250 </P>
251 <P>
252 In addition to the main compiling and matching functions, there are convenience
253 functions for extracting captured substrings from a subject string that is
254 matched by <b>pcre_exec()</b>. They are:
255 <pre>
256 <b>pcre_copy_substring()</b>
257 <b>pcre_copy_named_substring()</b>
258 <b>pcre_get_substring()</b>
259 <b>pcre_get_named_substring()</b>
260 <b>pcre_get_substring_list()</b>
261 <b>pcre_get_stringnumber()</b>
262 <b>pcre_get_stringtable_entries()</b>
263 </pre>
264 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
265 provided, to free the memory used for extracted strings.
266 </P>
267 <P>
268 The function <b>pcre_maketables()</b> is used to build a set of character tables
269 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
270 or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
271 specialist use. Most commonly, no special tables are passed, in which case
272 internal tables that are generated when PCRE is built are used.
273 </P>
274 <P>
275 The function <b>pcre_fullinfo()</b> is used to find out information about a
276 compiled pattern. The function <b>pcre_version()</b> returns a pointer to a
277 string containing the version of PCRE and its date of release.
278 </P>
279 <P>
280 The function <b>pcre_refcount()</b> maintains a reference count in a data block
281 containing a compiled pattern. This is provided for the benefit of
282 object-oriented applications.
283 </P>
284 <P>
285 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
286 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
287 respectively. PCRE calls the memory management functions via these variables,
288 so a calling program can replace them if it wishes to intercept the calls. This
289 should be done before calling any PCRE functions.
290 </P>
291 <P>
292 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
293 indirections to memory management functions. These special functions are used
294 only when PCRE is compiled to use the heap for remembering data, instead of
295 recursive function calls, when running the <b>pcre_exec()</b> function. See the
296 <a href="pcrebuild.html"><b>pcrebuild</b></a>
297 documentation for details of how to do this. It is a non-standard way of
298 building PCRE, for use in environments that have limited stacks. Because of the
299 greater use of memory management, it runs more slowly. Separate functions are
300 provided so that special-purpose external code can be used for this case. When
301 used, these functions are always called in a stack-like manner (last obtained,
302 first freed), and always for memory blocks of the same size. There is a
303 discussion about PCRE's stack usage in the
304 <a href="pcrestack.html"><b>pcrestack</b></a>
305 documentation.
306 </P>
307 <P>
308 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
309 by the caller to a "callout" function, which PCRE will then call at specified
310 points during a matching operation. Details are given in the
311 <a href="pcrecallout.html"><b>pcrecallout</b></a>
312 documentation.
313 <a name="newlines"></a></P>
314 <br><a name="SEC7" href="#TOC1">NEWLINES</a><br>
315 <P>
316 PCRE supports five different conventions for indicating line breaks in
317 strings: a single CR (carriage return) character, a single LF (linefeed)
318 character, the two-character sequence CRLF, any of the three preceding, or any
319 Unicode newline sequence. The Unicode newline sequences are the three just
320 mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
321 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
322 (paragraph separator, U+2029).
323 </P>
324 <P>
325 Each of the first three conventions is used by at least one operating system as
326 its standard newline sequence. When PCRE is built, a default can be specified.
327 The default default is LF, which is the Unix standard. When PCRE is run, the
328 default can be overridden, either when a pattern is compiled, or when it is
329 matched.
330 </P>
331 <P>
332 At compile time, the newline convention can be specified by the <i>options</i>
333 argument of <b>pcre_compile()</b>, or it can be specified by special text at the
334 start of the pattern itself; this overrides any other settings. See the
335 <a href="pcrepattern.html"><b>pcrepattern</b></a>
336 page for details of the special character sequences.
337 </P>
338 <P>
339 In the PCRE documentation the word "newline" is used to mean "the character or
340 pair of characters that indicate a line break". The choice of newline
341 convention affects the handling of the dot, circumflex, and dollar
342 metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
343 recognized line ending sequence, the match position advancement for a
344 non-anchored pattern. There is more detail about this in the
345 <a href="#execoptions">section on <b>pcre_exec()</b> options</a>
346 below.
347 </P>
348 <P>
349 The choice of newline convention does not affect the interpretation of
350 the \n or \r escape sequences, nor does it affect what \R matches, which is
351 controlled in a similar way, but by separate options.
352 </P>
353 <br><a name="SEC8" href="#TOC1">MULTITHREADING</a><br>
354 <P>
355 The PCRE functions can be used in multi-threading applications, with the
356 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
357 <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
358 callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
359 </P>
360 <P>
361 The compiled form of a regular expression is not altered during matching, so
362 the same compiled pattern can safely be used by several threads at once.
363 </P>
364 <P>
365 If the just-in-time optimization feature is being used, it needs separate
366 memory stack areas for each thread. See the
367 <a href="pcrejit.html"><b>pcrejit</b></a>
368 documentation for more details.
369 </P>
370 <br><a name="SEC9" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
371 <P>
372 The compiled form of a regular expression can be saved and re-used at a later
373 time, possibly by a different program, and even on a host other than the one on
374 which it was compiled. Details are given in the
375 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
376 documentation, which includes a description of the
377 <b>pcre_pattern_to_host_byte_order()</b> function. However, compiling a regular
378 expression with one version of PCRE for use with a different version is not
379 guaranteed to work and may cause crashes.
380 </P>
381 <br><a name="SEC10" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
382 <P>
383 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
384 </P>
385 <P>
386 The function <b>pcre_config()</b> makes it possible for a PCRE client to
387 discover which optional features have been compiled into the PCRE library. The
388 <a href="pcrebuild.html"><b>pcrebuild</b></a>
389 documentation has more details about these optional features.
390 </P>
391 <P>
392 The first argument for <b>pcre_config()</b> is an integer, specifying which
393 information is required; the second argument is a pointer to a variable into
394 which the information is placed. The returned value is zero on success, or the
395 negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
396 not recognized. The following information is available:
397 <pre>
398 PCRE_CONFIG_UTF8
399 </pre>
400 The output is an integer that is set to one if UTF-8 support is available;
401 otherwise it is set to zero. If this option is given to the 16-bit version of
402 this function, <b>pcre16_config()</b>, the result is PCRE_ERROR_BADOPTION.
403 <pre>
404 PCRE_CONFIG_UTF16
405 </pre>
406 The output is an integer that is set to one if UTF-16 support is available;
407 otherwise it is set to zero. This value should normally be given to the 16-bit
408 version of this function, <b>pcre16_config()</b>. If it is given to the 8-bit
409 version of this function, the result is PCRE_ERROR_BADOPTION.
410 <pre>
411 PCRE_CONFIG_UNICODE_PROPERTIES
412 </pre>
413 The output is an integer that is set to one if support for Unicode character
414 properties is available; otherwise it is set to zero.
415 <pre>
416 PCRE_CONFIG_JIT
417 </pre>
418 The output is an integer that is set to one if support for just-in-time
419 compiling is available; otherwise it is set to zero.
420 <pre>
421 PCRE_CONFIG_JITTARGET
422 </pre>
423 The output is a pointer to a zero-terminated "const char *" string. If JIT
424 support is available, the string contains the name of the architecture for
425 which the JIT compiler is configured, for example "x86 32bit (little endian +
426 unaligned)". If JIT support is not available, the result is NULL.
427 <pre>
428 PCRE_CONFIG_NEWLINE
429 </pre>
430 The output is an integer whose value specifies the default character sequence
431 that is recognized as meaning "newline". The four values that are supported
432 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
433 Though they are derived from ASCII, the same values are returned in EBCDIC
434 environments. The default should normally correspond to the standard sequence
435 for your operating system.
436 <pre>
437 PCRE_CONFIG_BSR
438 </pre>
439 The output is an integer whose value indicates what character sequences the \R
440 escape sequence matches by default. A value of 0 means that \R matches any
441 Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
442 or CRLF. The default can be overridden when a pattern is compiled or matched.
443 <pre>
444 PCRE_CONFIG_LINK_SIZE
445 </pre>
446 The output is an integer that contains the number of bytes used for internal
447 linkage in compiled regular expressions. For the 8-bit library, the value can
448 be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
449 a number of bytes. The default value of 2 is sufficient for all but the most
450 massive patterns, since it allows the compiled pattern to be up to 64K in size.
451 Larger values allow larger regular expressions to be compiled, at the expense
452 of slower matching.
453 <pre>
454 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
455 </pre>
456 The output is an integer that contains the threshold above which the POSIX
457 interface uses <b>malloc()</b> for output vectors. Further details are given in
458 the
459 <a href="pcreposix.html"><b>pcreposix</b></a>
460 documentation.
461 <pre>
462 PCRE_CONFIG_MATCH_LIMIT
463 </pre>
464 The output is a long integer that gives the default limit for the number of
465 internal matching function calls in a <b>pcre_exec()</b> execution. Further
466 details are given with <b>pcre_exec()</b> below.
467 <pre>
468 PCRE_CONFIG_MATCH_LIMIT_RECURSION
469 </pre>
470 The output is a long integer that gives the default limit for the depth of
471 recursion when calling the internal matching function in a <b>pcre_exec()</b>
472 execution. Further details are given with <b>pcre_exec()</b> below.
473 <pre>
474 PCRE_CONFIG_STACKRECURSE
475 </pre>
476 The output is an integer that is set to one if internal recursion when running
477 <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
478 to remember their state. This is the usual way that PCRE is compiled. The
479 output is zero if PCRE was compiled to use blocks of data on the heap instead
480 of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
481 <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
482 avoiding the use of the stack.
483 </P>
484 <br><a name="SEC11" href="#TOC1">COMPILING A PATTERN</a><br>
485 <P>
486 <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
487 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
488 <b>const unsigned char *<i>tableptr</i>);</b>
489 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
490 <b>int *<i>errorcodeptr</i>,</b>
491 <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
492 <b>const unsigned char *<i>tableptr</i>);</b>
493 </P>
494 <P>
495 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
496 called to compile a pattern into an internal form. The only difference between
497 the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
498 <i>errorcodeptr</i>, via which a numerical error code can be returned. To avoid
499 too much repetition, we refer just to <b>pcre_compile()</b> below, but the
500 information applies equally to <b>pcre_compile2()</b>.
501 </P>
502 <P>
503 The pattern is a C string terminated by a binary zero, and is passed in the
504 <i>pattern</i> argument. A pointer to a single block of memory that is obtained
505 via <b>pcre_malloc</b> is returned. This contains the compiled code and related
506 data. The <b>pcre</b> type is defined for the returned block; this is a typedef
507 for a structure whose contents are not externally defined. It is up to the
508 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
509 </P>
510 <P>
511 Although the compiled code of a PCRE regex is relocatable, that is, it does not
512 depend on memory location, the complete <b>pcre</b> data block is not
513 fully relocatable, because it may contain a copy of the <i>tableptr</i>
514 argument, which is an address (see below).
515 </P>
516 <P>
517 The <i>options</i> argument contains various bit settings that affect the
518 compilation. It should be zero if no options are required. The available
519 options are described below. Some of them (in particular, those that are
520 compatible with Perl, but some others as well) can also be set and unset from
521 within the pattern (see the detailed description in the
522 <a href="pcrepattern.html"><b>pcrepattern</b></a>
523 documentation). For those options that can be different in different parts of
524 the pattern, the contents of the <i>options</i> argument specifies their
525 settings at the start of compilation and execution. The PCRE_ANCHORED,
526 PCRE_BSR_<i>xxx</i>, PCRE_NEWLINE_<i>xxx</i>, PCRE_NO_UTF8_CHECK, and
527 PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
528 compile time.
529 </P>
530 <P>
531 If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
532 Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
533 NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
534 error message. This is a static string that is part of the library. You must
535 not try to free it. Normally, the offset from the start of the pattern to the
536 byte that was being processed when the error was discovered is placed in the
537 variable pointed to by <i>erroffset</i>, which must not be NULL (if it is, an
538 immediate error is given). However, for an invalid UTF-8 string, the offset is
539 that of the first byte of the failing character.
540 </P>
541 <P>
542 Some errors are not detected until the whole pattern has been scanned; in these
543 cases, the offset passed back is the length of the pattern. Note that the
544 offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
545 into the middle of a UTF-8 character.
546 </P>
547 <P>
548 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
549 <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
550 returned via this argument in the event of an error. This is in addition to the
551 textual error message. Error codes and messages are listed below.
552 </P>
553 <P>
554 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
555 character tables that are built when PCRE is compiled, using the default C
556 locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
557 call to <b>pcre_maketables()</b>. This value is stored with the compiled
558 pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
559 passed to it. For more discussion, see the section on locale support below.
560 </P>
561 <P>
562 This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
563 <pre>
564 pcre *re;
565 const char *error;
566 int erroffset;
567 re = pcre_compile(
568 "^A.*Z", /* the pattern */
569 0, /* default options */
570 &error, /* for error message */
571 &erroffset, /* for error offset */
572 NULL); /* use default character tables */
573 </pre>
574 The following names for option bits are defined in the <b>pcre.h</b> header
575 file:
576 <pre>
577 PCRE_ANCHORED
578 </pre>
579 If this bit is set, the pattern is forced to be "anchored", that is, it is
580 constrained to match only at the first matching point in the string that is
581 being searched (the "subject string"). This effect can also be achieved by
582 appropriate constructs in the pattern itself, which is the only way to do it in
583 Perl.
584 <pre>
585 PCRE_AUTO_CALLOUT
586 </pre>
587 If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
588 all with number 255, before each pattern item. For discussion of the callout
589 facility, see the
590 <a href="pcrecallout.html"><b>pcrecallout</b></a>
591 documentation.
592 <pre>
593 PCRE_BSR_ANYCRLF
594 PCRE_BSR_UNICODE
595 </pre>
596 These options (which are mutually exclusive) control what the \R escape
597 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
598 match any Unicode newline sequence. The default is specified when PCRE is
599 built. It can be overridden from within the pattern, or by setting an option
600 when a compiled pattern is matched.
601 <pre>
602 PCRE_CASELESS
603 </pre>
604 If this bit is set, letters in the pattern match both upper and lower case
605 letters. It is equivalent to Perl's /i option, and it can be changed within a
606 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
607 concept of case for characters whose values are less than 128, so caseless
608 matching is always possible. For characters with higher values, the concept of
609 case is supported if PCRE is compiled with Unicode property support, but not
610 otherwise. If you want to use caseless matching for characters 128 and above,
611 you must ensure that PCRE is compiled with Unicode property support as well as
612 with UTF-8 support.
613 <pre>
614 PCRE_DOLLAR_ENDONLY
615 </pre>
616 If this bit is set, a dollar metacharacter in the pattern matches only at the
617 end of the subject string. Without this option, a dollar also matches
618 immediately before a newline at the end of the string (but not before any other
619 newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
620 There is no equivalent to this option in Perl, and no way to set it within a
621 pattern.
622 <pre>
623 PCRE_DOTALL
624 </pre>
625 If this bit is set, a dot metacharacter in the pattern matches a character of
626 any value, including one that indicates a newline. However, it only ever
627 matches one character, even if newlines are coded as CRLF. Without this option,
628 a dot does not match when the current position is at a newline. This option is
629 equivalent to Perl's /s option, and it can be changed within a pattern by a
630 (?s) option setting. A negative class such as [^a] always matches newline
631 characters, independent of the setting of this option.
632 <pre>
633 PCRE_DUPNAMES
634 </pre>
635 If this bit is set, names used to identify capturing subpatterns need not be
636 unique. This can be helpful for certain types of pattern when it is known that
637 only one instance of the named subpattern can ever be matched. There are more
638 details of named subpatterns below; see also the
639 <a href="pcrepattern.html"><b>pcrepattern</b></a>
640 documentation.
641 <pre>
642 PCRE_EXTENDED
643 </pre>
644 If this bit is set, whitespace data characters in the pattern are totally
645 ignored except when escaped or inside a character class. Whitespace does not
646 include the VT character (code 11). In addition, characters between an
647 unescaped # outside a character class and the next newline, inclusive, are also
648 ignored. This is equivalent to Perl's /x option, and it can be changed within a
649 pattern by a (?x) option setting.
650 </P>
651 <P>
652 Which characters are interpreted as newlines is controlled by the options
653 passed to <b>pcre_compile()</b> or by a special sequence at the start of the
654 pattern, as described in the section entitled
655 <a href="pcrepattern.html#newlines">"Newline conventions"</a>
656 in the <b>pcrepattern</b> documentation. Note that the end of this type of
657 comment is a literal newline sequence in the pattern; escape sequences that
658 happen to represent a newline do not count.
659 </P>
660 <P>
661 This option makes it possible to include comments inside complicated patterns.
662 Note, however, that this applies only to data characters. Whitespace characters
663 may never appear within special character sequences in a pattern, for example
664 within the sequence (?( that introduces a conditional subpattern.
665 <pre>
666 PCRE_EXTRA
667 </pre>
668 This option was invented in order to turn on additional functionality of PCRE
669 that is incompatible with Perl, but it is currently of very little use. When
670 set, any backslash in a pattern that is followed by a letter that has no
671 special meaning causes an error, thus reserving these combinations for future
672 expansion. By default, as in Perl, a backslash followed by a letter with no
673 special meaning is treated as a literal. (Perl can, however, be persuaded to
674 give an error for this, by running it with the -w option.) There are at present
675 no other features controlled by this option. It can also be set by a (?X)
676 option setting within a pattern.
677 <pre>
678 PCRE_FIRSTLINE
679 </pre>
680 If this option is set, an unanchored pattern is required to match before or at
681 the first newline in the subject string, though the matched text may continue
682 over the newline.
683 <pre>
684 PCRE_JAVASCRIPT_COMPAT
685 </pre>
686 If this option is set, PCRE's behaviour is changed in some ways so that it is
687 compatible with JavaScript rather than Perl. The changes are as follows:
688 </P>
689 <P>
690 (1) A lone closing square bracket in a pattern causes a compile-time error,
691 because this is illegal in JavaScript (by default it is treated as a data
692 character). Thus, the pattern AB]CD becomes illegal when this option is set.
693 </P>
694 <P>
695 (2) At run time, a back reference to an unset subpattern group matches an empty
696 string (by default this causes the current matching alternative to fail). A
697 pattern such as (\1)(a) succeeds when this option is set (assuming it can find
698 an "a" in the subject), whereas it fails by default, for Perl compatibility.
699 </P>
700 <P>
701 (3) \U matches an upper case "U" character; by default \U causes a compile
702 time error (Perl uses \U to upper case subsequent characters).
703 </P>
704 <P>
705 (4) \u matches a lower case "u" character unless it is followed by four
706 hexadecimal digits, in which case the hexadecimal number defines the code point
707 to match. By default, \u causes a compile time error (Perl uses it to upper
708 case the following character).
709 </P>
710 <P>
711 (5) \x matches a lower case "x" character unless it is followed by two
712 hexadecimal digits, in which case the hexadecimal number defines the code point
713 to match. By default, as in Perl, a hexadecimal number is always expected after
714 \x, but it may have zero, one, or two digits (so, for example, \xz matches a
715 binary zero character followed by z).
716 <pre>
717 PCRE_MULTILINE
718 </pre>
719 By default, PCRE treats the subject string as consisting of a single line of
720 characters (even if it actually contains newlines). The "start of line"
721 metacharacter (^) matches only at the start of the string, while the "end of
722 line" metacharacter ($) matches only at the end of the string, or before a
723 terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
724 Perl.
725 </P>
726 <P>
727 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
728 match immediately following or immediately before internal newlines in the
729 subject string, respectively, as well as at the very start and end. This is
730 equivalent to Perl's /m option, and it can be changed within a pattern by a
731 (?m) option setting. If there are no newlines in a subject string, or no
732 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
733 <pre>
734 PCRE_NEWLINE_CR
735 PCRE_NEWLINE_LF
736 PCRE_NEWLINE_CRLF
737 PCRE_NEWLINE_ANYCRLF
738 PCRE_NEWLINE_ANY
739 </pre>
740 These options override the default newline definition that was chosen when PCRE
741 was built. Setting the first or the second specifies that a newline is
742 indicated by a single character (CR or LF, respectively). Setting
743 PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
744 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
745 preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
746 that any Unicode newline sequence should be recognized. The Unicode newline
747 sequences are the three just mentioned, plus the single characters VT (vertical
748 tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
749 separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
750 library, the last two are recognized only in UTF-8 mode.
751 </P>
752 <P>
753 The newline setting in the options word uses three bits that are treated
754 as a number, giving eight possibilities. Currently only six are used (default
755 plus the five values above). This means that if you set more than one newline
756 option, the combination may or may not be sensible. For example,
757 PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
758 other combinations may yield unused numbers and cause an error.
759 </P>
760 <P>
761 The only time that a line break in a pattern is specially recognized when
762 compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
763 and so are ignored in this mode. Also, an unescaped # outside a character class
764 indicates a comment that lasts until after the next line break sequence. In
765 other circumstances, line break sequences in patterns are treated as literal
766 data.
767 </P>
768 <P>
769 The newline option that is set at compile time becomes the default that is used
770 for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
771 <pre>
772 PCRE_NO_AUTO_CAPTURE
773 </pre>
774 If this option is set, it disables the use of numbered capturing parentheses in
775 the pattern. Any opening parenthesis that is not followed by ? behaves as if it
776 were followed by ?: but named parentheses can still be used for capturing (and
777 they acquire numbers in the usual way). There is no equivalent of this option
778 in Perl.
779 <pre>
780 NO_START_OPTIMIZE
781 </pre>
782 This is an option that acts at matching time; that is, it is really an option
783 for <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. If it is set at compile time,
784 it is remembered with the compiled pattern and assumed at matching time. For
785 details see the discussion of PCRE_NO_START_OPTIMIZE
786 <a href="#execoptions">below.</a>
787 <pre>
788 PCRE_UCP
789 </pre>
790 This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
791 \w, and some of the POSIX character classes. By default, only ASCII characters
792 are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
793 classify characters. More details are given in the section on
794 <a href="pcre.html#genericchartypes">generic character types</a>
795 in the
796 <a href="pcrepattern.html"><b>pcrepattern</b></a>
797 page. If you set PCRE_UCP, matching one of the items it affects takes much
798 longer. The option is available only if PCRE has been compiled with Unicode
799 property support.
800 <pre>
801 PCRE_UNGREEDY
802 </pre>
803 This option inverts the "greediness" of the quantifiers so that they are not
804 greedy by default, but become greedy if followed by "?". It is not compatible
805 with Perl. It can also be set by a (?U) option setting within the pattern.
806 <pre>
807 PCRE_UTF8
808 </pre>
809 This option causes PCRE to regard both the pattern and the subject as strings
810 of UTF-8 characters instead of single-byte strings. However, it is available
811 only when PCRE is built to include UTF support. If not, the use of this option
812 provokes an error. Details of how this option changes the behaviour of PCRE are
813 given in the
814 <a href="pcreunicode.html"><b>pcreunicode</b></a>
815 page.
816 <pre>
817 PCRE_NO_UTF8_CHECK
818 </pre>
819 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8
820 string is automatically checked. There is a discussion about the
821 <a href="pcreunicode.html#utf8strings">validity of UTF-8 strings</a>
822 in the
823 <a href="pcreunicode.html"><b>pcreunicode</b></a>
824 page. If an invalid UTF-8 sequence is found, <b>pcre_compile()</b> returns an
825 error. If you already know that your pattern is valid, and you want to skip
826 this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
827 When it is set, the effect of passing an invalid UTF-8 string as a pattern is
828 undefined. It may cause your program to crash. Note that this option can also
829 be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress the
830 validity checking of subject strings.
831 </P>
832 <br><a name="SEC12" href="#TOC1">COMPILATION ERROR CODES</a><br>
833 <P>
834 The following table lists the error codes than may be returned by
835 <b>pcre_compile2()</b>, along with the error messages that may be returned by
836 both compiling functions. Note that error messages are always 8-bit ASCII
837 strings, even in 16-bit mode. As PCRE has developed, some error codes have
838 fallen out of use. To avoid confusion, they have not been re-used.
839 <pre>
840 0 no error
841 1 \ at end of pattern
842 2 \c at end of pattern
843 3 unrecognized character follows \
844 4 numbers out of order in {} quantifier
845 5 number too big in {} quantifier
846 6 missing terminating ] for character class
847 7 invalid escape sequence in character class
848 8 range out of order in character class
849 9 nothing to repeat
850 10 [this code is not in use]
851 11 internal error: unexpected repeat
852 12 unrecognized character after (? or (?-
853 13 POSIX named classes are supported only within a class
854 14 missing )
855 15 reference to non-existent subpattern
856 16 erroffset passed as NULL
857 17 unknown option bit(s) set
858 18 missing ) after comment
859 19 [this code is not in use]
860 20 regular expression is too large
861 21 failed to get memory
862 22 unmatched parentheses
863 23 internal error: code overflow
864 24 unrecognized character after (?&#60;
865 25 lookbehind assertion is not fixed length
866 26 malformed number or name after (?(
867 27 conditional group contains more than two branches
868 28 assertion expected after (?(
869 29 (?R or (?[+-]digits must be followed by )
870 30 unknown POSIX class name
871 31 POSIX collating elements are not supported
872 32 this version of PCRE is compiled without UTF support
873 33 [this code is not in use]
874 34 character value in \x{...} sequence is too large
875 35 invalid condition (?(0)
876 36 \C not allowed in lookbehind assertion
877 37 PCRE does not support \L, \l, \N{name}, \U, or \u
878 38 number after (?C is &#62; 255
879 39 closing ) for (?C expected
880 40 recursive call could loop indefinitely
881 41 unrecognized character after (?P
882 42 syntax error in subpattern name (missing terminator)
883 43 two named subpatterns have the same name
884 44 invalid UTF-8 string (specifically UTF-8)
885 45 support for \P, \p, and \X has not been compiled
886 46 malformed \P or \p sequence
887 47 unknown property name after \P or \p
888 48 subpattern name is too long (maximum 32 characters)
889 49 too many named subpatterns (maximum 10000)
890 50 [this code is not in use]
891 51 octal value is greater than \377 in 8-bit non-UTF-8 mode
892 52 internal error: overran compiling workspace
893 53 internal error: previously-checked referenced subpattern
894 not found
895 54 DEFINE group contains more than one branch
896 55 repeating a DEFINE group is not allowed
897 56 inconsistent NEWLINE options
898 57 \g is not followed by a braced, angle-bracketed, or quoted
899 name/number or by a plain number
900 58 a numbered reference must not be zero
901 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
902 60 (*VERB) not recognized
903 61 number is too big
904 62 subpattern name expected
905 63 digit expected after (?+
906 64 ] is an invalid data character in JavaScript compatibility mode
907 65 different names for subpatterns of the same number are
908 not allowed
909 66 (*MARK) must have an argument
910 67 this version of PCRE is not compiled with Unicode property
911 support
912 68 \c must be followed by an ASCII character
913 69 \k is not followed by a braced, angle-bracketed, or quoted name
914 70 internal error: unknown opcode in find_fixedlength()
915 71 \N is not supported in a class
916 72 too many forward references
917 73 disallowed Unicode code point (&#62;= 0xd800 && &#60;= 0xdfff)
918 74 invalid UTF-16 string (specifically UTF-16)
919 </pre>
920 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
921 be used if the limits were changed when PCRE was built.
922 <a name="studyingapattern"></a></P>
923 <br><a name="SEC13" href="#TOC1">STUDYING A PATTERN</a><br>
924 <P>
925 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
926 <b>const char **<i>errptr</i>);</b>
927 </P>
928 <P>
929 If a compiled pattern is going to be used several times, it is worth spending
930 more time analyzing it in order to speed up the time taken for matching. The
931 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
932 argument. If studying the pattern produces additional information that will
933 help speed up matching, <b>pcre_study()</b> returns a pointer to a
934 <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
935 results of the study.
936 </P>
937 <P>
938 The returned value from <b>pcre_study()</b> can be passed directly to
939 <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. However, a <b>pcre_extra</b> block
940 also contains other fields that can be set by the caller before the block is
941 passed; these are described
942 <a href="#extradata">below</a>
943 in the section on matching a pattern.
944 </P>
945 <P>
946 If studying the pattern does not produce any useful information,
947 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
948 wants to pass any of the other fields to <b>pcre_exec()</b> or
949 <b>pcre_dfa_exec()</b>, it must set up its own <b>pcre_extra</b> block.
950 </P>
951 <P>
952 The second argument of <b>pcre_study()</b> contains option bits. There are three
953 options:
954 <pre>
955 PCRE_STUDY_JIT_COMPILE
956 PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
957 PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
958 </pre>
959 If any of these are set, and the just-in-time compiler is available, the
960 pattern is further compiled into machine code that executes much faster than
961 the <b>pcre_exec()</b> interpretive matching function. If the just-in-time
962 compiler is not available, these options are ignored. All other bits in the
963 <i>options</i> argument must be zero.
964 </P>
965 <P>
966 JIT compilation is a heavyweight optimization. It can take some time for
967 patterns to be analyzed, and for one-off matches and simple patterns the
968 benefit of faster execution might be offset by a much slower study time.
969 Not all patterns can be optimized by the JIT compiler. For those that cannot be
970 handled, matching automatically falls back to the <b>pcre_exec()</b>
971 interpreter. For more details, see the
972 <a href="pcrejit.html"><b>pcrejit</b></a>
973 documentation.
974 </P>
975 <P>
976 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
977 studying succeeds (even if no data is returned), the variable it points to is
978 set to NULL. Otherwise it is set to point to a textual error message. This is a
979 static string that is part of the library. You must not try to free it. You
980 should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
981 sure that it has run successfully.
982 </P>
983 <P>
984 When you are finished with a pattern, you can free the memory used for the
985 study data by calling <b>pcre_free_study()</b>. This function was added to the
986 API for release 8.20. For earlier versions, the memory could be freed with
987 <b>pcre_free()</b>, just like the pattern itself. This will still work in cases
988 where JIT optimization is not used, but it is advisable to change to the new
989 function when convenient.
990 </P>
991 <P>
992 This is a typical way in which <b>pcre_study</b>() is used (except that in a
993 real application there should be tests for errors):
994 <pre>
995 int rc;
996 pcre *re;
997 pcre_extra *sd;
998 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
999 sd = pcre_study(
1000 re, /* result of pcre_compile() */
1001 0, /* no options */
1002 &error); /* set to NULL or points to a message */
1003 rc = pcre_exec( /* see below for details of pcre_exec() options */
1004 re, sd, "subject", 7, 0, 0, ovector, 30);
1005 ...
1006 pcre_free_study(sd);
1007 pcre_free(re);
1008 </pre>
1009 Studying a pattern does two things: first, a lower bound for the length of
1010 subject string that is needed to match the pattern is computed. This does not
1011 mean that there are any strings of that length that match, but it does
1012 guarantee that no shorter strings match. The value is used by
1013 <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b> to avoid wasting time by trying to
1014 match strings that are shorter than the lower bound. You can find out the value
1015 in a calling program via the <b>pcre_fullinfo()</b> function.
1016 </P>
1017 <P>
1018 Studying a pattern is also useful for non-anchored patterns that do not have a
1019 single fixed starting character. A bitmap of possible starting bytes is
1020 created. This speeds up finding a position in the subject at which to start
1021 matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1022 </P>
1023 <P>
1024 These two optimizations apply to both <b>pcre_exec()</b> and
1025 <b>pcre_dfa_exec()</b>, and the information is also used by the JIT compiler.
1026 The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option
1027 when calling <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>, but if this is done,
1028 JIT execution is also disabled. You might want to do this if your pattern
1029 contains callouts or (*MARK) and you want to make use of these facilities in
1030 cases where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
1031 <a href="#execoptions">below.</a>
1032 <a name="localesupport"></a></P>
1033 <br><a name="SEC14" href="#TOC1">LOCALE SUPPORT</a><br>
1034 <P>
1035 PCRE handles caseless matching, and determines whether characters are letters,
1036 digits, or whatever, by reference to a set of tables, indexed by character
1037 value. When running in UTF-8 mode, this applies only to characters
1038 with codes less than 128. By default, higher-valued codes never match escapes
1039 such as \w or \d, but they can be tested with \p if PCRE is built with
1040 Unicode character property support. Alternatively, the PCRE_UCP option can be
1041 set at compile time; this causes \w and friends to use Unicode property
1042 support instead of built-in tables. The use of locales with Unicode is
1043 discouraged. If you are handling characters with codes greater than 128, you
1044 should either use UTF-8 and Unicode, or use locales, but not try to mix the
1045 two.
1046 </P>
1047 <P>
1048 PCRE contains an internal set of tables that are used when the final argument
1049 of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
1050 Normally, the internal tables recognize only ASCII characters. However, when
1051 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1052 default "C" locale of the local system, which may cause them to be different.
1053 </P>
1054 <P>
1055 The internal tables can always be overridden by tables supplied by the
1056 application that calls PCRE. These may be created in a different locale from
1057 the default. As more and more applications change to using Unicode, the need
1058 for this locale support is expected to die away.
1059 </P>
1060 <P>
1061 External tables are built by calling the <b>pcre_maketables()</b> function,
1062 which has no arguments, in the relevant locale. The result can then be passed
1063 to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
1064 example, to build and use tables that are appropriate for the French locale
1065 (where accented characters with values greater than 128 are treated as letters),
1066 the following code could be used:
1067 <pre>
1068 setlocale(LC_CTYPE, "fr_FR");
1069 tables = pcre_maketables();
1070 re = pcre_compile(..., tables);
1071 </pre>
1072 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1073 are using Windows, the name for the French locale is "french".
1074 </P>
1075 <P>
1076 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
1077 obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
1078 that the memory containing the tables remains available for as long as it is
1079 needed.
1080 </P>
1081 <P>
1082 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
1083 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
1084 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
1085 pattern, compilation, studying and matching all happen in the same locale, but
1086 different patterns can be compiled in different locales.
1087 </P>
1088 <P>
1089 It is possible to pass a table pointer or NULL (indicating the use of the
1090 internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
1091 this facility could be used to match a pattern in a different locale from the
1092 one in which it was compiled. Passing table pointers at run time is discussed
1093 below in the section on matching a pattern.
1094 <a name="infoaboutpattern"></a></P>
1095 <br><a name="SEC15" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
1096 <P>
1097 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1098 <b>int <i>what</i>, void *<i>where</i>);</b>
1099 </P>
1100 <P>
1101 The <b>pcre_fullinfo()</b> function returns information about a compiled
1102 pattern. It replaces the <b>pcre_info()</b> function, which was removed from the
1103 library at version 8.30, after more than 10 years of obsolescence.
1104 </P>
1105 <P>
1106 The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
1107 pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
1108 the pattern was not studied. The third argument specifies which piece of
1109 information is required, and the fourth argument is a pointer to a variable
1110 to receive the data. The yield of the function is zero for success, or one of
1111 the following negative numbers:
1112 <pre>
1113 PCRE_ERROR_NULL the argument <i>code</i> was NULL
1114 the argument <i>where</i> was NULL
1115 PCRE_ERROR_BADMAGIC the "magic number" was not found
1116 PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
1117 endianness
1118 PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
1119 </pre>
1120 The "magic number" is placed at the start of each compiled pattern as an simple
1121 check against passing an arbitrary memory pointer. The endianness error can
1122 occur if a compiled pattern is saved and reloaded on a different host. Here is
1123 a typical call of <b>pcre_fullinfo()</b>, to obtain the length of the compiled
1124 pattern:
1125 <pre>
1126 int rc;
1127 size_t length;
1128 rc = pcre_fullinfo(
1129 re, /* result of pcre_compile() */
1130 sd, /* result of pcre_study(), or NULL */
1131 PCRE_INFO_SIZE, /* what is required */
1132 &length); /* where to put the data */
1133 </pre>
1134 The possible values for the third argument are defined in <b>pcre.h</b>, and are
1135 as follows:
1136 <pre>
1137 PCRE_INFO_BACKREFMAX
1138 </pre>
1139 Return the number of the highest back reference in the pattern. The fourth
1140 argument should point to an <b>int</b> variable. Zero is returned if there are
1141 no back references.
1142 <pre>
1143 PCRE_INFO_CAPTURECOUNT
1144 </pre>
1145 Return the number of capturing subpatterns in the pattern. The fourth argument
1146 should point to an <b>int</b> variable.
1147 <pre>
1148 PCRE_INFO_DEFAULT_TABLES
1149 </pre>
1150 Return a pointer to the internal default character tables within PCRE. The
1151 fourth argument should point to an <b>unsigned char *</b> variable. This
1152 information call is provided for internal use by the <b>pcre_study()</b>
1153 function. External callers can cause PCRE to use its internal tables by passing
1154 a NULL table pointer.
1155 <pre>
1156 PCRE_INFO_FIRSTBYTE
1157 </pre>
1158 Return information about the first data unit of any matched string, for a
1159 non-anchored pattern. (The name of this option refers to the 8-bit library,
1160 where data units are bytes.) The fourth argument should point to an <b>int</b>
1161 variable.
1162 </P>
1163 <P>
1164 If there is a fixed first value, for example, the letter "c" from a pattern
1165 such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1166 value is always less than 256; in the 16-bit library the value can be up to
1167 0xffff.
1168 </P>
1169 <P>
1170 If there is no fixed first value, and if either
1171 <br>
1172 <br>
1173 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1174 starts with "^", or
1175 <br>
1176 <br>
1177 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1178 (if it were set, the pattern would be anchored),
1179 <br>
1180 <br>
1181 -1 is returned, indicating that the pattern matches only at the start of a
1182 subject string or after any newline within the string. Otherwise -2 is
1183 returned. For anchored patterns, -2 is returned.
1184 <pre>
1185 PCRE_INFO_FIRSTTABLE
1186 </pre>
1187 If the pattern was studied, and this resulted in the construction of a 256-bit
1188 table indicating a fixed set of values for the first data unit in any matching
1189 string, a pointer to the table is returned. Otherwise NULL is returned. The
1190 fourth argument should point to an <b>unsigned char *</b> variable.
1191 <pre>
1192 PCRE_INFO_HASCRORLF
1193 </pre>
1194 Return 1 if the pattern contains any explicit matches for CR or LF characters,
1195 otherwise 0. The fourth argument should point to an <b>int</b> variable. An
1196 explicit match is either a literal CR or LF character, or \r or \n.
1197 <pre>
1198 PCRE_INFO_JCHANGED
1199 </pre>
1200 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1201 0. The fourth argument should point to an <b>int</b> variable. (?J) and
1202 (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1203 <pre>
1204 PCRE_INFO_JIT
1205 </pre>
1206 Return 1 if the pattern was studied with one of the JIT options, and
1207 just-in-time compiling was successful. The fourth argument should point to an
1208 <b>int</b> variable. A return value of 0 means that JIT support is not available
1209 in this version of PCRE, or that the pattern was not studied with a JIT option,
1210 or that the JIT compiler could not handle this particular pattern. See the
1211 <a href="pcrejit.html"><b>pcrejit</b></a>
1212 documentation for details of what can and cannot be handled.
1213 <pre>
1214 PCRE_INFO_JITSIZE
1215 </pre>
1216 If the pattern was successfully studied with a JIT option, return the size of
1217 the JIT compiled code, otherwise return zero. The fourth argument should point
1218 to a <b>size_t</b> variable.
1219 <pre>
1220 PCRE_INFO_LASTLITERAL
1221 </pre>
1222 Return the value of the rightmost literal data unit that must exist in any
1223 matched string, other than at its start, if such a value has been recorded. The
1224 fourth argument should point to an <b>int</b> variable. If there is no such
1225 value, -1 is returned. For anchored patterns, a last literal value is recorded
1226 only if it follows something of variable length. For example, for the pattern
1227 /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
1228 is -1.
1229 <pre>
1230 PCRE_INFO_MAXLOOKBEHIND
1231 </pre>
1232 Return the number of characters (NB not bytes) in the longest lookbehind
1233 assertion in the pattern. Note that the simple assertions \b and \B require a
1234 one-character lookbehind. This information is useful when doing multi-segment
1235 matching using the partial matching facilities.
1236 <pre>
1237 PCRE_INFO_MINLENGTH
1238 </pre>
1239 If the pattern was studied and a minimum length for matching subject strings
1240 was computed, its value is returned. Otherwise the returned value is -1. The
1241 value is a number of characters, which in UTF-8 mode may be different from the
1242 number of bytes. The fourth argument should point to an <b>int</b> variable. A
1243 non-negative value is a lower bound to the length of any matching string. There
1244 may not be any strings of that length that do actually match, but every string
1245 that does match is at least that long.
1246 <pre>
1247 PCRE_INFO_NAMECOUNT
1248 PCRE_INFO_NAMEENTRYSIZE
1249 PCRE_INFO_NAMETABLE
1250 </pre>
1251 PCRE supports the use of named as well as numbered capturing parentheses. The
1252 names are just an additional way of identifying the parentheses, which still
1253 acquire numbers. Several convenience functions such as
1254 <b>pcre_get_named_substring()</b> are provided for extracting captured
1255 substrings by name. It is also possible to extract the data directly, by first
1256 converting the name to a number in order to access the correct pointers in the
1257 output vector (described with <b>pcre_exec()</b> below). To do the conversion,
1258 you need to use the name-to-number map, which is described by these three
1259 values.
1260 </P>
1261 <P>
1262 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1263 the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1264 entry; both of these return an <b>int</b> value. The entry size depends on the
1265 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1266 entry of the table. This is a pointer to <b>char</b> in the 8-bit library, where
1267 the first two bytes of each entry are the number of the capturing parenthesis,
1268 most significant byte first. In the 16-bit library, the pointer points to
1269 16-bit data units, the first of which contains the parenthesis number. The rest
1270 of the entry is the corresponding name, zero terminated.
1271 </P>
1272 <P>
1273 The names are in alphabetical order. Duplicate names may appear if (?| is used
1274 to create multiple groups with the same number, as described in the
1275 <a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
1276 in the
1277 <a href="pcrepattern.html"><b>pcrepattern</b></a>
1278 page. Duplicate names for subpatterns with different numbers are permitted only
1279 if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1280 table in the order in which they were found in the pattern. In the absence of
1281 (?| this is the order of increasing number; when (?| is used this is not
1282 necessarily the case because later subpatterns may have lower numbers.
1283 </P>
1284 <P>
1285 As a simple example of the name/number table, consider the following pattern
1286 after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1287 space - including newlines - is ignored):
1288 <pre>
1289 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
1290 </pre>
1291 There are four named subpatterns, so the table has four entries, and each entry
1292 in the table is eight bytes long. The table is as follows, with non-printing
1293 bytes shows in hexadecimal, and undefined bytes shown as ??:
1294 <pre>
1295 00 01 d a t e 00 ??
1296 00 05 d a y 00 ?? ??
1297 00 04 m o n t h 00
1298 00 02 y e a r 00 ??
1299 </pre>
1300 When writing code to extract data from named subpatterns using the
1301 name-to-number map, remember that the length of the entries is likely to be
1302 different for each compiled pattern.
1303 <pre>
1304 PCRE_INFO_OKPARTIAL
1305 </pre>
1306 Return 1 if the pattern can be used for partial matching with
1307 <b>pcre_exec()</b>, otherwise 0. The fourth argument should point to an
1308 <b>int</b> variable. From release 8.00, this always returns 1, because the
1309 restrictions that previously applied to partial matching have been lifted. The
1310 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1311 documentation gives details of partial matching.
1312 <pre>
1313 PCRE_INFO_OPTIONS
1314 </pre>
1315 Return a copy of the options with which the pattern was compiled. The fourth
1316 argument should point to an <b>unsigned long int</b> variable. These option bits
1317 are those specified in the call to <b>pcre_compile()</b>, modified by any
1318 top-level option settings at the start of the pattern itself. In other words,
1319 they are the options that will be in force when matching starts. For example,
1320 if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1321 result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1322 </P>
1323 <P>
1324 A pattern is automatically anchored by PCRE if all of its top-level
1325 alternatives begin with one of the following:
1326 <pre>
1327 ^ unless PCRE_MULTILINE is set
1328 \A always
1329 \G always
1330 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
1331 </pre>
1332 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1333 <b>pcre_fullinfo()</b>.
1334 <pre>
1335 PCRE_INFO_SIZE
1336 </pre>
1337 Return the size of the compiled pattern in bytes (for both libraries). The
1338 fourth argument should point to a <b>size_t</b> variable. This value does not
1339 include the size of the <b>pcre</b> structure that is returned by
1340 <b>pcre_compile()</b>. The value that is passed as the argument to
1341 <b>pcre_malloc()</b> when <b>pcre_compile()</b> is getting memory in which to
1342 place the compiled data is the value returned by this option plus the size of
1343 the <b>pcre</b> structure. Studying a compiled pattern, with or without JIT,
1344 does not alter the value returned by this option.
1345 <pre>
1346 PCRE_INFO_STUDYSIZE
1347 </pre>
1348 Return the size in bytes of the data block pointed to by the <i>study_data</i>
1349 field in a <b>pcre_extra</b> block. If <b>pcre_extra</b> is NULL, or there is no
1350 study data, zero is returned. The fourth argument should point to a
1351 <b>size_t</b> variable. The <i>study_data</i> field is set by <b>pcre_study()</b>
1352 to record information that will speed up matching (see the section entitled
1353 <a href="#studyingapattern">"Studying a pattern"</a>
1354 above). The format of the <i>study_data</i> block is private, but its length
1355 is made available via this option so that it can be saved and restored (see the
1356 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1357 documentation for details).
1358 </P>
1359 <br><a name="SEC16" href="#TOC1">REFERENCE COUNTS</a><br>
1360 <P>
1361 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1362 </P>
1363 <P>
1364 The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1365 data block that contains a compiled pattern. It is provided for the benefit of
1366 applications that operate in an object-oriented manner, where different parts
1367 of the application may be using the same compiled pattern, but you want to free
1368 the block when they are all done.
1369 </P>
1370 <P>
1371 When a pattern is compiled, the reference count field is initialized to zero.
1372 It is changed only by calling this function, whose action is to add the
1373 <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1374 function is the new value. However, the value of the count is constrained to
1375 lie between 0 and 65535, inclusive. If the new value is outside these limits,
1376 it is forced to the appropriate limit value.
1377 </P>
1378 <P>
1379 Except when it is zero, the reference count is not correctly preserved if a
1380 pattern is compiled on one host and then transferred to a host whose byte-order
1381 is different. (This seems a highly unlikely scenario.)
1382 </P>
1383 <br><a name="SEC17" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1384 <P>
1385 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1386 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1387 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1388 </P>
1389 <P>
1390 The function <b>pcre_exec()</b> is called to match a subject string against a
1391 compiled pattern, which is passed in the <i>code</i> argument. If the
1392 pattern was studied, the result of the study should be passed in the
1393 <i>extra</i> argument. You can call <b>pcre_exec()</b> with the same <i>code</i>
1394 and <i>extra</i> arguments as many times as you like, in order to match
1395 different subject strings with the same pattern.
1396 </P>
1397 <P>
1398 This function is the main matching facility of the library, and it operates in
1399 a Perl-like manner. For specialist use there is also an alternative matching
1400 function, which is described
1401 <a href="#dfamatch">below</a>
1402 in the section about the <b>pcre_dfa_exec()</b> function.
1403 </P>
1404 <P>
1405 In most applications, the pattern will have been compiled (and optionally
1406 studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1407 possible to save compiled patterns and study data, and then use them later
1408 in different processes, possibly even on different hosts. For a discussion
1409 about this, see the
1410 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1411 documentation.
1412 </P>
1413 <P>
1414 Here is an example of a simple call to <b>pcre_exec()</b>:
1415 <pre>
1416 int rc;
1417 int ovector[30];
1418 rc = pcre_exec(
1419 re, /* result of pcre_compile() */
1420 NULL, /* we didn't study the pattern */
1421 "some string", /* the subject string */
1422 11, /* the length of the subject string */
1423 0, /* start at offset 0 in the subject */
1424 0, /* default options */
1425 ovector, /* vector of integers for substring information */
1426 30); /* number of elements (NOT size in bytes) */
1427 <a name="extradata"></a></PRE>
1428 </P>
1429 <br><b>
1430 Extra data for <b>pcre_exec()</b>
1431 </b><br>
1432 <P>
1433 If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1434 data block. The <b>pcre_study()</b> function returns such a block (when it
1435 doesn't return NULL), but you can also create one for yourself, and pass
1436 additional information in it. The <b>pcre_extra</b> block contains the following
1437 fields (not necessarily in this order):
1438 <pre>
1439 unsigned long int <i>flags</i>;
1440 void *<i>study_data</i>;
1441 void *<i>executable_jit</i>;
1442 unsigned long int <i>match_limit</i>;
1443 unsigned long int <i>match_limit_recursion</i>;
1444 void *<i>callout_data</i>;
1445 const unsigned char *<i>tables</i>;
1446 unsigned char **<i>mark</i>;
1447 </pre>
1448 In the 16-bit version of this structure, the <i>mark</i> field has type
1449 "PCRE_UCHAR16 **".
1450 </P>
1451 <P>
1452 The <i>flags</i> field is used to specify which of the other fields are set. The
1453 flag bits are:
1454 <pre>
1455 PCRE_EXTRA_CALLOUT_DATA
1456 PCRE_EXTRA_EXECUTABLE_JIT
1457 PCRE_EXTRA_MARK
1458 PCRE_EXTRA_MATCH_LIMIT
1459 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1460 PCRE_EXTRA_STUDY_DATA
1461 PCRE_EXTRA_TABLES
1462 </pre>
1463 Other flag bits should be set to zero. The <i>study_data</i> field and sometimes
1464 the <i>executable_jit</i> field are set in the <b>pcre_extra</b> block that is
1465 returned by <b>pcre_study()</b>, together with the appropriate flag bits. You
1466 should not set these yourself, but you may add to the block by setting other
1467 fields and their corresponding flag bits.
1468 </P>
1469 <P>
1470 The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1471 vast amount of resources when running patterns that are not going to match,
1472 but which have a very large number of possibilities in their search trees. The
1473 classic example is a pattern that uses nested unlimited repeats.
1474 </P>
1475 <P>
1476 Internally, <b>pcre_exec()</b> uses a function called <b>match()</b>, which it
1477 calls repeatedly (sometimes recursively). The limit set by <i>match_limit</i> is
1478 imposed on the number of times this function is called during a match, which
1479 has the effect of limiting the amount of backtracking that can take place. For
1480 patterns that are not anchored, the count restarts from zero for each position
1481 in the subject string.
1482 </P>
1483 <P>
1484 When <b>pcre_exec()</b> is called with a pattern that was successfully studied
1485 with a JIT option, the way that the matching is executed is entirely different.
1486 However, there is still the possibility of runaway matching that goes on for a
1487 very long time, and so the <i>match_limit</i> value is also used in this case
1488 (but in a different way) to limit how long the matching can continue.
1489 </P>
1490 <P>
1491 The default value for the limit can be set when PCRE is built; the default
1492 default is 10 million, which handles all but the most extreme cases. You can
1493 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1494 block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1495 the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1496 PCRE_ERROR_MATCHLIMIT.
1497 </P>
1498 <P>
1499 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1500 instead of limiting the total number of times that <b>match()</b> is called, it
1501 limits the depth of recursion. The recursion depth is a smaller number than the
1502 total number of calls, because not all calls to <b>match()</b> are recursive.
1503 This limit is of use only if it is set smaller than <i>match_limit</i>.
1504 </P>
1505 <P>
1506 Limiting the recursion depth limits the amount of machine stack that can be
1507 used, or, when PCRE has been compiled to use memory on the heap instead of the
1508 stack, the amount of heap memory that can be used. This limit is not relevant,
1509 and is ignored, when matching is done using JIT compiled code.
1510 </P>
1511 <P>
1512 The default value for <i>match_limit_recursion</i> can be set when PCRE is
1513 built; the default default is the same value as the default for
1514 <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1515 with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1516 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1517 is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1518 </P>
1519 <P>
1520 The <i>callout_data</i> field is used in conjunction with the "callout" feature,
1521 and is described in the
1522 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1523 documentation.
1524 </P>
1525 <P>
1526 The <i>tables</i> field is used to pass a character tables pointer to
1527 <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1528 pattern. A non-NULL value is stored with the compiled pattern only if custom
1529 tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1530 If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1531 internal tables to be used. This facility is helpful when re-using patterns
1532 that have been saved after compiling with an external set of tables, because
1533 the external tables might be at a different address when <b>pcre_exec()</b> is
1534 called. See the
1535 <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1536 documentation for a discussion of saving compiled patterns for later use.
1537 </P>
1538 <P>
1539 If PCRE_EXTRA_MARK is set in the <i>flags</i> field, the <i>mark</i> field must
1540 be set to point to a suitable variable. If the pattern contains any
1541 backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1542 a name to pass back, a pointer to the name string (zero terminated) is placed
1543 in the variable pointed to by the <i>mark</i> field. The names are within the
1544 compiled pattern; if you wish to retain such a name you must copy it before
1545 freeing the memory of a compiled pattern. If there is no name to pass back, the
1546 variable pointed to by the <i>mark</i> field is set to NULL. For details of the
1547 backtracking control verbs, see the section entitled
1548 <a href="pcrepattern#backtrackcontrol">"Backtracking control"</a>
1549 in the
1550 <a href="pcrepattern.html"><b>pcrepattern</b></a>
1551 documentation.
1552 <a name="execoptions"></a></P>
1553 <br><b>
1554 Option bits for <b>pcre_exec()</b>
1555 </b><br>
1556 <P>
1557 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1558 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1559 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1560 PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
1561 PCRE_PARTIAL_SOFT.
1562 </P>
1563 <P>
1564 If the pattern was successfully studied with one of the just-in-time (JIT)
1565 compile options, the only supported options for JIT execution are
1566 PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
1567 PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
1568 unsupported option is used, JIT execution is disabled and the normal
1569 interpretive code in <b>pcre_exec()</b> is run.
1570 <pre>
1571 PCRE_ANCHORED
1572 </pre>
1573 The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1574 matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1575 to be anchored by virtue of its contents, it cannot be made unachored at
1576 matching time.
1577 <pre>
1578 PCRE_BSR_ANYCRLF
1579 PCRE_BSR_UNICODE
1580 </pre>
1581 These options (which are mutually exclusive) control what the \R escape
1582 sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1583 match any Unicode newline sequence. These options override the choice that was
1584 made or defaulted when the pattern was compiled.
1585 <pre>
1586 PCRE_NEWLINE_CR
1587 PCRE_NEWLINE_LF
1588 PCRE_NEWLINE_CRLF
1589 PCRE_NEWLINE_ANYCRLF
1590 PCRE_NEWLINE_ANY
1591 </pre>
1592 These options override the newline definition that was chosen or defaulted when
1593 the pattern was compiled. For details, see the description of
1594 <b>pcre_compile()</b> above. During matching, the newline choice affects the
1595 behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1596 the way the match position is advanced after a match failure for an unanchored
1597 pattern.
1598 </P>
1599 <P>
1600 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1601 match attempt for an unanchored pattern fails when the current position is at a
1602 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1603 characters, the match position is advanced by two characters instead of one, in
1604 other words, to after the CRLF.
1605 </P>
1606 <P>
1607 The above rule is a compromise that makes the most common cases work as
1608 expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1609 set), it does not match the string "\r\nA" because, after failing at the
1610 start, it skips both the CR and the LF before retrying. However, the pattern
1611 [\r\n]A does match that string, because it contains an explicit CR or LF
1612 reference, and so advances only by one character after the first failure.
1613 </P>
1614 <P>
1615 An explicit match for CR of LF is either a literal appearance of one of those
1616 characters, or one of the \r or \n escape sequences. Implicit matches such as
1617 [^X] do not count, nor does \s (which includes CR and LF in the characters
1618 that it matches).
1619 </P>
1620 <P>
1621 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1622 valid newline sequence and explicit \r or \n escapes appear in the pattern.
1623 <pre>
1624 PCRE_NOTBOL
1625 </pre>
1626 This option specifies that first character of the subject string is not the
1627 beginning of a line, so the circumflex metacharacter should not match before
1628 it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1629 never to match. This option affects only the behaviour of the circumflex
1630 metacharacter. It does not affect \A.
1631 <pre>
1632 PCRE_NOTEOL
1633 </pre>
1634 This option specifies that the end of the subject string is not the end of a
1635 line, so the dollar metacharacter should not match it nor (except in multiline
1636 mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1637 compile time) causes dollar never to match. This option affects only the
1638 behaviour of the dollar metacharacter. It does not affect \Z or \z.
1639 <pre>
1640 PCRE_NOTEMPTY
1641 </pre>
1642 An empty string is not considered to be a valid match if this option is set. If
1643 there are alternatives in the pattern, they are tried. If all the alternatives
1644 match the empty string, the entire match fails. For example, if the pattern
1645 <pre>
1646 a?b?
1647 </pre>
1648 is applied to a string not beginning with "a" or "b", it matches an empty
1649 string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1650 valid, so PCRE searches further into the string for occurrences of "a" or "b".
1651 <pre>
1652 PCRE_NOTEMPTY_ATSTART
1653 </pre>
1654 This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1655 the start of the subject is permitted. If the pattern is anchored, such a match
1656 can occur only if the pattern contains \K.
1657 </P>
1658 <P>
1659 Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1660 does make a special case of a pattern match of the empty string within its
1661 <b>split()</b> function, and when using the /g modifier. It is possible to
1662 emulate Perl's behaviour after matching a null string by first trying the match
1663 again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1664 if that fails, by advancing the starting offset (see below) and trying an
1665 ordinary match again. There is some code that demonstrates how to do this in
1666 the
1667 <a href="pcredemo.html"><b>pcredemo</b></a>
1668 sample program. In the most general case, you have to check to see if the
1669 newline convention recognizes CRLF as a newline, and if so, and the current
1670 character is CR followed by LF, advance the starting offset by two characters
1671 instead of one.
1672 <pre>
1673 PCRE_NO_START_OPTIMIZE
1674 </pre>
1675 There are a number of optimizations that <b>pcre_exec()</b> uses at the start of
1676 a match, in order to speed up the process. For example, if it is known that an
1677 unanchored match must start with a specific character, it searches the subject
1678 for that character, and fails immediately if it cannot find it, without
1679 actually running the main matching function. This means that a special item
1680 such as (*COMMIT) at the start of a pattern is not considered until after a
1681 suitable starting point for the match has been found. When callouts or (*MARK)
1682 items are in use, these "start-up" optimizations can cause them to be skipped
1683 if the pattern is never actually used. The start-up optimizations are in effect
1684 a pre-scan of the subject that takes place before the pattern is run.
1685 </P>
1686 <P>
1687 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1688 causing performance to suffer, but ensuring that in cases where the result is
1689 "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1690 are considered at every possible starting position in the subject string. If
1691 PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1692 time. The use of PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set,
1693 matching is always done using interpretively.
1694 </P>
1695 <P>
1696 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1697 Consider the pattern
1698 <pre>
1699 (*COMMIT)ABC
1700 </pre>
1701 When this is compiled, PCRE records the fact that a match must start with the
1702 character "A". Suppose the subject string is "DEFABC". The start-up
1703 optimization scans along the subject, finds "A" and runs the first match
1704 attempt from there. The (*COMMIT) item means that the pattern must match the
1705 current starting position, which in this case, it does. However, if the same
1706 match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1707 subject string does not happen. The first match attempt is run starting from
1708 "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1709 the overall result is "no match". If the pattern is studied, more start-up
1710 optimizations may be used. For example, a minimum length for the subject may be
1711 recorded. Consider the pattern
1712 <pre>
1713 (*MARK:A)(X|Y)
1714 </pre>
1715 The minimum length for a match is one character. If the subject is "ABC", there
1716 will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1717 If the pattern is studied, the final attempt does not take place, because PCRE
1718 knows that the subject is too short, and so the (*MARK) is never encountered.
1719 In this case, studying the pattern does not affect the overall match result,
1720 which is still "no match", but it does affect the auxiliary information that is
1721 returned.
1722 <pre>
1723 PCRE_NO_UTF8_CHECK
1724 </pre>
1725 When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1726 string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1727 The entire string is checked before any other processing takes place. The value
1728 of <i>startoffset</i> is also checked to ensure that it points to the start of a
1729 UTF-8 character. There is a discussion about the
1730 <a href="pcreunicode.html#utf8strings">validity of UTF-8 strings</a>
1731 in the
1732 <a href="pcreunicode.html"><b>pcreunicode</b></a>
1733 page. If an invalid sequence of bytes is found, <b>pcre_exec()</b> returns the
1734 error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1735 truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1736 cases, information about the precise nature of the error may also be returned
1737 (see the descriptions of these errors in the section entitled \fIError return
1738 values from\fP <b>pcre_exec()</b>
1739 <a href="#errorlist">below).</a>
1740 If <i>startoffset</i> contains a value that does not point to the start of a
1741 UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1742 returned.
1743 </P>
1744 <P>
1745 If you already know that your subject is valid, and you want to skip these
1746 checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1747 calling <b>pcre_exec()</b>. You might want to do this for the second and
1748 subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1749 all the matches in a single subject string. However, you should be sure that
1750 the value of <i>startoffset</i> points to the start of a character (or the end
1751 of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1752 invalid string as a subject or an invalid value of <i>startoffset</i> is
1753 undefined. Your program may crash.
1754 <pre>
1755 PCRE_PARTIAL_HARD
1756 PCRE_PARTIAL_SOFT
1757 </pre>
1758 These options turn on the partial matching feature. For backwards
1759 compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1760 occurs if the end of the subject string is reached successfully, but there are
1761 not enough subject characters to complete the match. If this happens when
1762 PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1763 testing any remaining alternatives. Only if no complete match can be found is
1764 PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1765 PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1766 but only if no complete match can be found.
1767 </P>
1768 <P>
1769 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1770 partial match is found, <b>pcre_exec()</b> immediately returns
1771 PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1772 when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1773 important that an alternative complete match.
1774 </P>
1775 <P>
1776 In both cases, the portion of the string that was inspected when the partial
1777 match was found is set as the first matching string. There is a more detailed
1778 discussion of partial and multi-segment matching, with examples, in the
1779 <a href="pcrepartial.html"><b>pcrepartial</b></a>
1780 documentation.
1781 </P>
1782 <br><b>
1783 The string to be matched by <b>pcre_exec()</b>
1784 </b><br>
1785 <P>
1786 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1787 <i>subject</i>, a length in bytes in <i>length</i>, and a starting byte offset
1788 in <i>startoffset</i>. If this is negative or greater than the length of the
1789 subject, <b>pcre_exec()</b> returns PCRE_ERROR_BADOFFSET. When the starting
1790 offset is zero, the search for a match starts at the beginning of the subject,
1791 and this is by far the most common case. In UTF-8 mode, the byte offset must
1792 point to the start of a UTF-8 character (or the end of the subject). Unlike the
1793 pattern string, the subject may contain binary zero bytes.
1794 </P>
1795 <P>
1796 A non-zero starting offset is useful when searching for another match in the
1797 same subject by calling <b>pcre_exec()</b> again after a previous success.
1798 Setting <i>startoffset</i> differs from just passing over a shortened string and
1799 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1800 lookbehind. For example, consider the pattern
1801 <pre>
1802 \Biss\B
1803 </pre>
1804 which finds occurrences of "iss" in the middle of words. (\B matches only if
1805 the current position in the subject is not a word boundary.) When applied to
1806 the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1807 occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1808 subject, namely "issipi", it does not match, because \B is always false at the
1809 start of the subject, which is deemed to be a word boundary. However, if
1810 <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1811 set to 4, it finds the second occurrence of "iss" because it is able to look
1812 behind the starting point to discover that it is preceded by a letter.
1813 </P>
1814 <P>
1815 Finding all the matches in a subject is tricky when the pattern can match an
1816 empty string. It is possible to emulate Perl's /g behaviour by first trying the
1817 match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1818 PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1819 and trying an ordinary match again. There is some code that demonstrates how to
1820 do this in the
1821 <a href="pcredemo.html"><b>pcredemo</b></a>
1822 sample program. In the most general case, you have to check to see if the
1823 newline convention recognizes CRLF as a newline, and if so, and the current
1824 character is CR followed by LF, advance the starting offset by two characters
1825 instead of one.
1826 </P>
1827 <P>
1828 If a non-zero starting offset is passed when the pattern is anchored, one
1829 attempt to match at the given offset is made. This can only succeed if the
1830 pattern does not require the match to be at the start of the subject.
1831 </P>
1832 <br><b>
1833 How <b>pcre_exec()</b> returns captured substrings
1834 </b><br>
1835 <P>
1836 In general, a pattern matches a certain portion of the subject, and in
1837 addition, further substrings from the subject may be picked out by parts of the
1838 pattern. Following the usage in Jeffrey Friedl's book, this is called
1839 "capturing" in what follows, and the phrase "capturing subpattern" is used for
1840 a fragment of a pattern that picks out a substring. PCRE supports several other
1841 kinds of parenthesized subpattern that do not cause substrings to be captured.
1842 </P>
1843 <P>
1844 Captured substrings are returned to the caller via a vector of integers whose
1845 address is passed in <i>ovector</i>. The number of elements in the vector is
1846 passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>: this
1847 argument is NOT the size of <i>ovector</i> in bytes.
1848 </P>
1849 <P>
1850 The first two-thirds of the vector is used to pass back captured substrings,
1851 each substring using a pair of integers. The remaining third of the vector is
1852 used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1853 and is not available for passing back information. The number passed in
1854 <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1855 rounded down.
1856 </P>
1857 <P>
1858 When a match is successful, information about captured substrings is returned
1859 in pairs of integers, starting at the beginning of <i>ovector</i>, and
1860 continuing up to two-thirds of its length at the most. The first element of
1861 each pair is set to the byte offset of the first character in a substring, and
1862 the second is set to the byte offset of the first character after the end of a
1863 substring. <b>Note</b>: these values are always byte offsets, even in UTF-8
1864 mode. They are not character counts.
1865 </P>
1866 <P>
1867 The first pair of integers, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the
1868 portion of the subject string matched by the entire pattern. The next pair is
1869 used for the first capturing subpattern, and so on. The value returned by
1870 <b>pcre_exec()</b> is one more than the highest numbered pair that has been set.
1871 For example, if two substrings have been captured, the returned value is 3. If
1872 there are no capturing subpatterns, the return value from a successful match is
1873 1, indicating that just the first pair of offsets has been set.
1874 </P>
1875 <P>
1876 If a capturing subpattern is matched repeatedly, it is the last portion of the
1877 string that it matched that is returned.
1878 </P>
1879 <P>
1880 If the vector is too small to hold all the captured substring offsets, it is
1881 used as far as possible (up to two-thirds of its length), and the function
1882 returns a value of zero. If neither the actual string matched nor any captured
1883 substrings are of interest, <b>pcre_exec()</b> may be called with <i>ovector</i>
1884 passed as NULL and <i>ovecsize</i> as zero. However, if the pattern contains
1885 back references and the <i>ovector</i> is not big enough to remember the related
1886 substrings, PCRE has to get additional memory for use during matching. Thus it
1887 is usually advisable to supply an <i>ovector</i> of reasonable size.
1888 </P>
1889 <P>
1890 There are some cases where zero is returned (indicating vector overflow) when
1891 in fact the vector is exactly the right size for the final match. For example,
1892 consider the pattern
1893 <pre>
1894 (a)(?:(b)c|bd)
1895 </pre>
1896 If a vector of 6 elements (allowing for only 1 captured substring) is given
1897 with subject string "abd", <b>pcre_exec()</b> will try to set the second
1898 captured string, thereby recording a vector overflow, before failing to match
1899 "c" and backing up to try the second alternative. The zero return, however,
1900 does correctly indicate that the maximum number of slots (namely 2) have been
1901 filled. In similar cases where there is temporary overflow, but the final
1902 number of used slots is actually less than the maximum, a non-zero value is
1903 returned.
1904 </P>
1905 <P>
1906 The <b>pcre_fullinfo()</b> function can be used to find out how many capturing
1907 subpatterns there are in a compiled pattern. The smallest size for
1908 <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1909 the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1910 </P>
1911 <P>
1912 It is possible for capturing subpattern number <i>n+1</i> to match some part of
1913 the subject when subpattern <i>n</i> has not been used at all. For example, if
1914 the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1915 function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1916 happens, both values in the offset pairs corresponding to unused subpatterns
1917 are set to -1.
1918 </P>
1919 <P>
1920 Offset values that correspond to unused subpatterns at the end of the
1921 expression are also set to -1. For example, if the string "abc" is matched
1922 against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1923 return from the function is 2, because the highest used capturing subpattern
1924 number is 1, and the offsets for for the second and third capturing subpatterns
1925 (assuming the vector is large enough, of course) are set to -1.
1926 </P>
1927 <P>
1928 <b>Note</b>: Elements in the first two-thirds of <i>ovector</i> that do not
1929 correspond to capturing parentheses in the pattern are never changed. That is,
1930 if a pattern contains <i>n</i> capturing parentheses, no more than
1931 <i>ovector[0]</i> to <i>ovector[2n+1]</i> are set by <b>pcre_exec()</b>. The other
1932 elements (in the first two-thirds) retain whatever values they previously had.
1933 </P>
1934 <P>
1935 Some convenience functions are provided for extracting the captured substrings
1936 as separate strings. These are described below.
1937 <a name="errorlist"></a></P>
1938 <br><b>
1939 Error return values from <b>pcre_exec()</b>
1940 </b><br>
1941 <P>
1942 If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1943 defined in the header file:
1944 <pre>
1945 PCRE_ERROR_NOMATCH (-1)
1946 </pre>
1947 The subject string did not match the pattern.
1948 <pre>
1949 PCRE_ERROR_NULL (-2)
1950 </pre>
1951 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1952 NULL and <i>ovecsize</i> was not zero.
1953 <pre>
1954 PCRE_ERROR_BADOPTION (-3)
1955 </pre>
1956 An unrecognized bit was set in the <i>options</i> argument.
1957 <pre>
1958 PCRE_ERROR_BADMAGIC (-4)
1959 </pre>
1960 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1961 the case when it is passed a junk pointer and to detect when a pattern that was
1962 compiled in an environment of one endianness is run in an environment with the
1963 other endianness. This is the error that PCRE gives when the magic number is
1964 not present.
1965 <pre>
1966 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1967 </pre>
1968 While running the pattern match, an unknown item was encountered in the
1969 compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1970 of the compiled pattern.
1971 <pre>
1972 PCRE_ERROR_NOMEMORY (-6)
1973 </pre>
1974 If a pattern contains back references, but the <i>ovector</i> that is passed to
1975 <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1976 gets a block of memory at the start of matching to use for this purpose. If the
1977 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1978 automatically freed at the end of matching.
1979 </P>
1980 <P>
1981 This error is also given if <b>pcre_stack_malloc()</b> fails in
1982 <b>pcre_exec()</b>. This can happen only when PCRE has been compiled with
1983 <b>--disable-stack-for-recursion</b>.
1984 <pre>
1985 PCRE_ERROR_NOSUBSTRING (-7)
1986 </pre>
1987 This error is used by the <b>pcre_copy_substring()</b>,
1988 <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1989 below). It is never returned by <b>pcre_exec()</b>.
1990 <pre>
1991 PCRE_ERROR_MATCHLIMIT (-8)
1992 </pre>
1993 The backtracking limit, as specified by the <i>match_limit</i> field in a
1994 <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1995 above.
1996 <pre>
1997 PCRE_ERROR_CALLOUT (-9)
1998 </pre>
1999 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
2000 use by callout functions that want to yield a distinctive error code. See the
2001 <a href="pcrecallout.html"><b>pcrecallout</b></a>
2002 documentation for details.
2003 <pre>
2004 PCRE_ERROR_BADUTF8 (-10)
2005 </pre>
2006 A string that contains an invalid UTF-8 byte sequence was passed as a subject,
2007 and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2008 (<i>ovecsize</i>) is at least 2, the byte offset to the start of the the invalid
2009 UTF-8 character is placed in the first element, and a reason code is placed in
2010 the second element. The reason codes are listed in the
2011 <a href="#badutf8reasons">following section.</a>
2012 For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2013 truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2014 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2015 <pre>
2016 PCRE_ERROR_BADUTF8_OFFSET (-11)
2017 </pre>
2018 The UTF-8 byte sequence that was passed as a subject was checked and found to
2019 be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
2020 <i>startoffset</i> did not point to the beginning of a UTF-8 character or the
2021 end of the subject.
2022 <pre>
2023 PCRE_ERROR_PARTIAL (-12)
2024 </pre>
2025 The subject string did not match, but it did match partially. See the
2026 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2027 documentation for details of partial matching.
2028 <pre>
2029 PCRE_ERROR_BADPARTIAL (-13)
2030 </pre>
2031 This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2032 option was used with a compiled pattern containing items that were not
2033 supported for partial matching. From release 8.00 onwards, there are no
2034 restrictions on partial matching.
2035 <pre>
2036 PCRE_ERROR_INTERNAL (-14)
2037 </pre>
2038 An unexpected internal error has occurred. This error could be caused by a bug
2039 in PCRE or by overwriting of the compiled pattern.
2040 <pre>
2041 PCRE_ERROR_BADCOUNT (-15)
2042 </pre>
2043 This error is given if the value of the <i>ovecsize</i> argument is negative.
2044 <pre>
2045 PCRE_ERROR_RECURSIONLIMIT (-21)
2046 </pre>
2047 The internal recursion limit, as specified by the <i>match_limit_recursion</i>
2048 field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
2049 description above.
2050 <pre>
2051 PCRE_ERROR_BADNEWLINE (-23)
2052 </pre>
2053 An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
2054 <pre>
2055 PCRE_ERROR_BADOFFSET (-24)
2056 </pre>
2057 The value of <i>startoffset</i> was negative or greater than the length of the
2058 subject, that is, the value in <i>length</i>.
2059 <pre>
2060 PCRE_ERROR_SHORTUTF8 (-25)
2061 </pre>
2062 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2063 ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2064 Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2065 fact sufficient to detect this case, but this special error code for
2066 PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2067 retained for backwards compatibility.
2068 <pre>
2069 PCRE_ERROR_RECURSELOOP (-26)
2070 </pre>
2071 This error is returned when <b>pcre_exec()</b> detects a recursion loop within
2072 the pattern. Specifically, it means that either the whole pattern or a
2073 subpattern has been called recursively for the second time at the same position
2074 in the subject string. Some simple patterns that might do this are detected and
2075 faulted at compile time, but more complicated cases, in particular mutual
2076 recursions between two different subpatterns, cannot be detected until run
2077 time.
2078 <pre>
2079 PCRE_ERROR_JIT_STACKLIMIT (-27)
2080 </pre>
2081 This error is returned when a pattern that was successfully studied using a
2082 JIT compile option is being matched, but the memory available for the
2083 just-in-time processing stack is not large enough. See the
2084 <a href="pcrejit.html"><b>pcrejit</b></a>
2085 documentation for more details.
2086 <pre>
2087 PCRE_ERROR_BADMODE (-28)
2088 </pre>
2089 This error is given if a pattern that was compiled by the 8-bit library is
2090 passed to a 16-bit library function, or vice versa.
2091 <pre>
2092 PCRE_ERROR_BADENDIANNESS (-29)
2093 </pre>
2094 This error is given if a pattern that was compiled and saved is reloaded on a
2095 host with different endianness. The utility function
2096 <b>pcre_pattern_to_host_byte_order()</b> can be used to convert such a pattern
2097 so that it runs on the new host.
2098 </P>
2099 <P>
2100 Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
2101 <a name="badutf8reasons"></a></P>
2102 <br><b>
2103 Reason codes for invalid UTF-8 strings
2104 </b><br>
2105 <P>
2106 This section applies only to the 8-bit library. The corresponding information
2107 for the 16-bit library is given in the
2108 <a href="pcre16.html"><b>pcre16</b></a>
2109 page.
2110 </P>
2111 <P>
2112 When <b>pcre_exec()</b> returns either PCRE_ERROR_BADUTF8 or
2113 PCRE_ERROR_SHORTUTF8, and the size of the output vector (<i>ovecsize</i>) is at
2114 least 2, the offset of the start of the invalid UTF-8 character is placed in
2115 the first output vector element (<i>ovector[0]</i>) and a reason code is placed
2116 in the second element (<i>ovector[1]</i>). The reason codes are given names in
2117 the <b>pcre.h</b> header file:
2118 <pre>
2119 PCRE_UTF8_ERR1
2120 PCRE_UTF8_ERR2
2121 PCRE_UTF8_ERR3
2122 PCRE_UTF8_ERR4
2123 PCRE_UTF8_ERR5
2124 </pre>
2125 The string ends with a truncated UTF-8 character; the code specifies how many
2126 bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2127 no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2128 allows for up to 6 bytes, and this is checked first; hence the possibility of
2129 4 or 5 missing bytes.
2130 <pre>
2131 PCRE_UTF8_ERR6
2132 PCRE_UTF8_ERR7
2133 PCRE_UTF8_ERR8
2134 PCRE_UTF8_ERR9
2135 PCRE_UTF8_ERR10
2136 </pre>
2137 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2138 character do not have the binary value 0b10 (that is, either the most
2139 significant bit is 0, or the next bit is 1).
2140 <pre>
2141 PCRE_UTF8_ERR11
2142 PCRE_UTF8_ERR12
2143 </pre>
2144 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2145 these code points are excluded by RFC 3629.
2146 <pre>
2147 PCRE_UTF8_ERR13
2148 </pre>
2149 A 4-byte character has a value greater than 0x10fff; these code points are
2150 excluded by RFC 3629.
2151 <pre>
2152 PCRE_UTF8_ERR14
2153 </pre>
2154 A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2155 code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2156 from UTF-8.
2157 <pre>
2158 PCRE_UTF8_ERR15
2159 PCRE_UTF8_ERR16
2160 PCRE_UTF8_ERR17
2161 PCRE_UTF8_ERR18
2162 PCRE_UTF8_ERR19
2163 </pre>
2164 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2165 value that can be represented by fewer bytes, which is invalid. For example,
2166 the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2167 one byte.
2168 <pre>
2169 PCRE_UTF8_ERR20
2170 </pre>
2171 The two most significant bits of the first byte of a character have the binary
2172 value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2173 byte can only validly occur as the second or subsequent byte of a multi-byte
2174 character.
2175 <pre>
2176 PCRE_UTF8_ERR21
2177 </pre>
2178 The first byte of a character has the value 0xfe or 0xff. These values can
2179 never occur in a valid UTF-8 string.
2180 </P>
2181 <br><a name="SEC18" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
2182 <P>
2183 <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
2184 <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
2185 <b>int <i>buffersize</i>);</b>
2186 </P>
2187 <P>
2188 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
2189 <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
2190 <b>const char **<i>stringptr</i>);</b>
2191 </P>
2192 <P>
2193 <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
2194 <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
2195 </P>
2196 <P>
2197 Captured substrings can be accessed directly by using the offsets returned by
2198 <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
2199 <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
2200 <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
2201 as new, separate, zero-terminated strings. These functions identify substrings
2202 by number. The next section describes functions for extracting named
2203 substrings.
2204 </P>
2205 <P>
2206 A substring that contains a binary zero is correctly extracted and has a
2207 further zero added on the end, but the result is not, of course, a C string.
2208 However, you can process such a string by referring to the length that is
2209 returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
2210 Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
2211 for handling strings containing binary zeros, because the end of the final
2212 string is not independently indicated.
2213 </P>
2214 <P>
2215 The first three arguments are the same for all three of these functions:
2216 <i>subject</i> is the subject string that has just been successfully matched,
2217 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
2218 <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
2219 captured by the match, including the substring that matched the entire regular
2220 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
2221 than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
2222 space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
2223 number of elements in the vector divided by three.
2224 </P>
2225 <P>
2226 The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
2227 extract a single substring, whose number is given as <i>stringnumber</i>. A
2228 value of zero extracts the substring that matched the entire pattern, whereas
2229 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
2230 the string is placed in <i>buffer</i>, whose length is given by
2231 <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
2232 obtained via <b>pcre_malloc</b>, and its address is returned via
2233 <i>stringptr</i>. The yield of the function is the length of the string, not
2234 including the terminating zero, or one of these error codes:
2235 <pre>
2236 PCRE_ERROR_NOMEMORY (-6)
2237 </pre>
2238 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
2239 memory failed for <b>pcre_get_substring()</b>.
2240 <pre>
2241 PCRE_ERROR_NOSUBSTRING (-7)
2242 </pre>
2243 There is no substring whose number is <i>stringnumber</i>.
2244 </P>
2245 <P>
2246 The <b>pcre_get_substring_list()</b> function extracts all available substrings
2247 and builds a list of pointers to them. All this is done in a single block of
2248 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
2249 is returned via <i>listptr</i>, which is also the start of the list of string
2250 pointers. The end of the list is marked by a NULL pointer. The yield of the
2251 function is zero if all went well, or the error code
2252 <pre>
2253 PCRE_ERROR_NOMEMORY (-6)
2254 </pre>
2255 if the attempt to get the memory block failed.
2256 </P>
2257 <P>
2258 When any of these functions encounter a substring that is unset, which can
2259 happen when capturing subpattern number <i>n+1</i> matches some part of the
2260 subject, but subpattern <i>n</i> has not been used at all, they return an empty
2261 string. This can be distinguished from a genuine zero-length substring by
2262 inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
2263 substrings.
2264 </P>
2265 <P>
2266 The two convenience functions <b>pcre_free_substring()</b> and
2267 <b>pcre_free_substring_list()</b> can be used to free the memory returned by
2268 a previous call of <b>pcre_get_substring()</b> or
2269 <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
2270 the function pointed to by <b>pcre_free</b>, which of course could be called
2271 directly from a C program. However, PCRE is used in some situations where it is
2272 linked via a special interface to another programming language that cannot use
2273 <b>pcre_free</b> directly; it is for these cases that the functions are
2274 provided.
2275 </P>
2276 <br><a name="SEC19" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
2277 <P>
2278 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
2279 <b>const char *<i>name</i>);</b>
2280 </P>
2281 <P>
2282 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
2283 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
2284 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
2285 <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
2286 </P>
2287 <P>
2288 <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
2289 <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
2290 <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
2291 <b>const char **<i>stringptr</i>);</b>
2292 </P>
2293 <P>
2294 To extract a substring by name, you first have to find associated number.
2295 For example, for this pattern
2296 <pre>
2297 (a+)b(?&#60;xxx&#62;\d+)...
2298 </pre>
2299 the number of the subpattern called "xxx" is 2. If the name is known to be
2300 unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2301 calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
2302 pattern, and the second is the name. The yield of the function is the
2303 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2304 that name.
2305 </P>
2306 <P>
2307 Given the number, you can extract the substring directly, or use one of the
2308 functions described in the previous section. For convenience, there are also
2309 two functions that do the whole job.
2310 </P>
2311 <P>
2312 Most of the arguments of <b>pcre_copy_named_substring()</b> and
2313 <b>pcre_get_named_substring()</b> are the same as those for the similarly named
2314 functions that extract by number. As these are described in the previous
2315 section, they are not re-described here. There are just two differences:
2316 </P>
2317 <P>
2318 First, instead of a substring number, a substring name is given. Second, there
2319 is an extra argument, given at the start, which is a pointer to the compiled
2320 pattern. This is needed in order to gain access to the name-to-number
2321 translation table.
2322 </P>
2323 <P>
2324 These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
2325 then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
2326 appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
2327 the behaviour may not be what you want (see the next section).
2328 </P>
2329 <P>
2330 <b>Warning:</b> If the pattern uses the (?| feature to set up multiple
2331 subpatterns with the same number, as described in the
2332 <a href="pcrepattern.html#dupsubpatternnumber">section on duplicate subpattern numbers</a>
2333 in the
2334 <a href="pcrepattern.html"><b>pcrepattern</b></a>
2335 page, you cannot use names to distinguish the different subpatterns, because
2336 names are not included in the compiled code. The matching process uses only
2337 numbers. For this reason, the use of different names for subpatterns of the
2338 same number causes an error at compile time.
2339 </P>
2340 <br><a name="SEC20" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
2341 <P>
2342 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
2343 <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
2344 </P>
2345 <P>
2346 When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2347 are not required to be unique. (Duplicate names are always allowed for
2348 subpatterns with the same number, created by using the (?| feature. Indeed, if
2349 such subpatterns are named, they are required to use the same names.)
2350 </P>
2351 <P>
2352 Normally, patterns with duplicate names are such that in any one match, only
2353 one of the named subpatterns participates. An example is shown in the
2354 <a href="pcrepattern.html"><b>pcrepattern</b></a>
2355 documentation.
2356 </P>
2357 <P>
2358 When duplicates are present, <b>pcre_copy_named_substring()</b> and
2359 <b>pcre_get_named_substring()</b> return the first substring corresponding to
2360 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2361 returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
2362 returns one of the numbers that are associated with the name, but it is not
2363 defined which it is.
2364 </P>
2365 <P>
2366 If you want to get full details of all captured substrings for a given name,
2367 you must use the <b>pcre_get_stringtable_entries()</b> function. The first
2368 argument is the compiled pattern, and the second is the name. The third and
2369 fourth are pointers to variables which are updated by the function. After it
2370 has run, they point to the first and last entries in the name-to-number table
2371 for the given name. The function itself returns the length of each entry, or
2372 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2373 described above in the section entitled <i>Information about a pattern</i>
2374 <a href="#infoaboutpattern">above.</a>
2375 Given all the relevant entries for the name, you can extract each of their
2376 numbers, and hence the captured data, if any.
2377 </P>
2378 <br><a name="SEC21" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
2379 <P>
2380 The traditional matching function uses a similar algorithm to Perl, which stops
2381 when it finds the first match, starting at a given point in the subject. If you
2382 want to find all possible matches, or the longest possible match, consider
2383 using the alternative matching function (see below) instead. If you cannot use
2384 the alternative function, but still need to find all possible matches, you
2385 can kludge it up by making use of the callout facility, which is described in
2386 the
2387 <a href="pcrecallout.html"><b>pcrecallout</b></a>
2388 documentation.
2389 </P>
2390 <P>
2391 What you have to do is to insert a callout right at the end of the pattern.
2392 When your callout function is called, extract and save the current matched
2393 substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
2394 other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
2395 will yield PCRE_ERROR_NOMATCH.
2396 </P>
2397 <br><a name="SEC22" href="#TOC1">OBTAINING AN ESTIMATE OF STACK USAGE</a><br>
2398 <P>
2399 Matching certain patterns using <b>pcre_exec()</b> can use a lot of process
2400 stack, which in certain environments can be rather limited in size. Some users
2401 find it helpful to have an estimate of the amount of stack that is used by
2402 <b>pcre_exec()</b>, to help them set recursion limits, as described in the
2403 <a href="pcrestack.html"><b>pcrestack</b></a>
2404 documentation. The estimate that is output by <b>pcretest</b> when called with
2405 the <b>-m</b> and <b>-C</b> options is obtained by calling <b>pcre_exec</b> with
2406 the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
2407 </P>
2408 <P>
2409 Normally, if its first argument is NULL, <b>pcre_exec()</b> immediately returns
2410 the negative error code PCRE_ERROR_NULL, but with this special combination of
2411 arguments, it returns instead a negative number whose absolute value is the
2412 approximate stack frame size in bytes. (A negative number is used so that it is
2413 clear that no match has happened.) The value is approximate because in some
2414 cases, recursive calls to <b>pcre_exec()</b> occur when there are one or two
2415 additional variables on the stack.
2416 </P>
2417 <P>
2418 If PCRE has been compiled to use the heap instead of the stack for recursion,
2419 the value returned is the size of each block that is obtained from the heap.
2420 <a name="dfamatch"></a></P>
2421 <br><a name="SEC23" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
2422 <P>
2423 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
2424 <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
2425 <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
2426 <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
2427 </P>
2428 <P>
2429 The function <b>pcre_dfa_exec()</b> is called to match a subject string against
2430 a compiled pattern, using a matching algorithm that scans the subject string
2431 just once, and does not backtrack. This has different characteristics to the
2432 normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2433 patterns are not supported. Nevertheless, there are times when this kind of
2434 matching can be useful. For a discussion of the two matching algorithms, and a
2435 list of features that <b>pcre_dfa_exec()</b> does not support, see the
2436 <a href="pcrematching.html"><b>pcrematching</b></a>
2437 documentation.
2438 </P>
2439 <P>
2440 The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
2441 <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
2442 different way, and this is described below. The other common arguments are used
2443 in the same way as for <b>pcre_exec()</b>, so their description is not repeated
2444 here.
2445 </P>
2446 <P>
2447 The two additional arguments provide workspace for the function. The workspace
2448 vector should contain at least 20 elements. It is used for keeping track of
2449 multiple paths through the pattern tree. More workspace will be needed for
2450 patterns and subjects where there are a lot of potential matches.
2451 </P>
2452 <P>
2453 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
2454 <pre>
2455 int rc;
2456 int ovector[10];
2457 int wspace[20];
2458 rc = pcre_dfa_exec(
2459 re, /* result of pcre_compile() */
2460 NULL, /* we didn't study the pattern */
2461 "some string", /* the subject string */
2462 11, /* the length of the subject string */
2463 0, /* start at offset 0 in the subject */
2464 0, /* default options */
2465 ovector, /* vector of integers for substring information */
2466 10, /* number of elements (NOT size in bytes) */
2467 wspace, /* working space vector */
2468 20); /* number of elements (NOT size in bytes) */
2469 </PRE>
2470 </P>
2471 <br><b>
2472 Option bits for <b>pcre_dfa_exec()</b>
2473 </b><br>
2474 <P>
2475 The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
2476 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
2477 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2478 PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2479 PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2480 All but the last four of these are exactly the same as for <b>pcre_exec()</b>,
2481 so their description is not repeated here.
2482 <pre>
2483 PCRE_PARTIAL_HARD
2484 PCRE_PARTIAL_SOFT
2485 </pre>
2486 These have the same general effect as they do for <b>pcre_exec()</b>, but the
2487 details are slightly different. When PCRE_PARTIAL_HARD is set for
2488 <b>pcre_dfa_exec()</b>, it returns PCRE_ERROR_PARTIAL if the end of the subject
2489 is reached and there is still at least one matching possibility that requires
2490 additional characters. This happens even if some complete matches have also
2491 been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2492 is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2493 there have been no complete matches, but there is still at least one matching
2494 possibility. The portion of the string that was inspected when the longest
2495 partial match was found is set as the first matching string in both cases.
2496 There is a more detailed discussion of partial and multi-segment matching, with
2497 examples, in the
2498 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2499 documentation.
2500 <pre>
2501 PCRE_DFA_SHORTEST
2502 </pre>
2503 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2504 soon as it has found one match. Because of the way the alternative algorithm
2505 works, this is necessarily the shortest possible match at the first possible
2506 matching point in the subject string.
2507 <pre>
2508 PCRE_DFA_RESTART
2509 </pre>
2510 When <b>pcre_dfa_exec()</b> returns a partial match, it is possible to call it
2511 again, with additional subject characters, and have it continue with the same
2512 match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2513 <i>workspace</i> and <i>wscount</i> options must reference the same vector as
2514 before because data about the match so far is left in them after a partial
2515 match. There is more discussion of this facility in the
2516 <a href="pcrepartial.html"><b>pcrepartial</b></a>
2517 documentation.
2518 </P>
2519 <br><b>
2520 Successful returns from <b>pcre_dfa_exec()</b>
2521 </b><br>
2522 <P>
2523 When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
2524 substring in the subject. Note, however, that all the matches from one run of
2525 the function start at the same point in the subject. The shorter matches are
2526 all initial substrings of the longer matches. For example, if the pattern
2527 <pre>
2528 &#60;.*&#62;
2529 </pre>
2530 is matched against the string
2531 <pre>
2532 This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
2533 </pre>
2534 the three matched strings are
2535 <pre>
2536 &#60;something&#62;
2537 &#60;something&#62; &#60;something else&#62;
2538 &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
2539 </pre>
2540 On success, the yield of the function is a number greater than zero, which is
2541 the number of matched substrings. The substrings themselves are returned in
2542 <i>ovector</i>. Each string uses two elements; the first is the offset to the
2543 start, and the second is the offset to the end. In fact, all the strings have
2544 the same start offset. (Space could have been saved by giving this only once,
2545 but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
2546 returns data, even though the meaning of the strings is different.)
2547 </P>
2548 <P>
2549 The strings are returned in reverse order of length; that is, the longest
2550 matching string is given first. If there were too many matches to fit into
2551 <i>ovector</i>, the yield of the function is zero, and the vector is filled with
2552 the longest matches. Unlike <b>pcre_exec()</b>, <b>pcre_dfa_exec()</b> can use
2553 the entire <i>ovector</i> for returning matched strings.
2554 </P>
2555 <br><b>
2556 Error returns from <b>pcre_dfa_exec()</b>
2557 </b><br>
2558 <P>
2559 The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
2560 Many of the errors are the same as for <b>pcre_exec()</b>, and these are
2561 described
2562 <a href="#errorlist">above.</a>
2563 There are in addition the following errors that are specific to
2564 <b>pcre_dfa_exec()</b>:
2565 <pre>
2566 PCRE_ERROR_DFA_UITEM (-16)
2567 </pre>
2568 This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
2569 that it does not support, for instance, the use of \C or a back reference.
2570 <pre>
2571 PCRE_ERROR_DFA_UCOND (-17)
2572 </pre>
2573 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
2574 uses a back reference for the condition, or a test for recursion in a specific
2575 group. These are not supported.
2576 <pre>
2577 PCRE_ERROR_DFA_UMLIMIT (-18)
2578 </pre>
2579 This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
2580 block that contains a setting of the <i>match_limit</i> or
2581 <i>match_limit_recursion</i> fields. This is not supported (these fields are
2582 meaningless for DFA matching).
2583 <pre>
2584 PCRE_ERROR_DFA_WSSIZE (-19)
2585 </pre>
2586 This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
2587 <i>workspace</i> vector.
2588 <pre>
2589 PCRE_ERROR_DFA_RECURSE (-20)
2590 </pre>
2591 When a recursive subpattern is processed, the matching function calls itself
2592 recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
2593 error is given if the output vector is not large enough. This should be
2594 extremely rare, as a vector of size 1000 is used.
2595 </P>
2596 <br><a name="SEC24" href="#TOC1">SEE ALSO</a><br>
2597 <P>
2598 <b>pcre16</b>(3), <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
2599 <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
2600 <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
2601 </P>
2602 <br><a name="SEC25" href="#TOC1">AUTHOR</a><br>
2603 <P>
2604 Philip Hazel
2605 <br>
2606 University Computing Service
2607 <br>
2608 Cambridge CB2 3QH, England.
2609 <br>
2610 </P>
2611 <br><a name="SEC26" href="#TOC1">REVISION</a><br>
2612 <P>
2613 Last updated: 14 April 2012
2614 <br>
2615 Copyright &copy; 1997-2012 University of Cambridge.
2616 <br>
2617 <p>
2618 Return to the <a href="index.html">PCRE index page</a>.
2619 </p>

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12