ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 96 - (hide annotations) (download) (as text)
Fri Mar 2 13:10:43 2007 UTC (8 years, 1 month ago) by nigel
File MIME type: text/html
File size: 83305 byte(s)
 r6896@hex:  nm | 2007-03-02 13:09:14 +0000
 Added EOL and keywork properties throughout

1 nigel 63 <html>
2     <head>
3     <title>pcreapi specification</title>
4     </head>
5     <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 nigel 75 <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 nigel 63 <ul>
16 nigel 75 <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17     <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18 nigel 91 <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19     <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
20     <li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
21     <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22     <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23     <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24     <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25     <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26     <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27     <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28     <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
29     <li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
30     <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
31     <li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
32     <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33     <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
34     <li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
35 nigel 93 <li><a name="TOC20" href="#SEC20">SEE ALSO</a>
36 nigel 63 </ul>
37 nigel 75 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
38 nigel 63 <P>
39     <b>#include &#60;pcre.h&#62;</b>
40     </P>
41     <P>
42     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
43     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
44     <b>const unsigned char *<i>tableptr</i>);</b>
45     </P>
46     <P>
47 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
48     <b>int *<i>errorcodeptr</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 nigel 63 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
54     <b>const char **<i>errptr</i>);</b>
55     </P>
56     <P>
57     <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
58     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
59     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
60     </P>
61     <P>
62 nigel 77 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
63     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
64     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
65     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
66     </P>
67     <P>
68 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
69     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
70     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
71     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
72     </P>
73     <P>
74     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
75     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
76     <b>int <i>buffersize</i>);</b>
77     </P>
78     <P>
79     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
80     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
81     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
82     <b>const char **<i>stringptr</i>);</b>
83     </P>
84     <P>
85     <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
86     <b>const char *<i>name</i>);</b>
87     </P>
88     <P>
89 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
90     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
91     </P>
92     <P>
93 nigel 63 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
94     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
95     <b>const char **<i>stringptr</i>);</b>
96     </P>
97     <P>
98     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
99     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
100     </P>
101     <P>
102     <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
103     </P>
104     <P>
105     <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
106     </P>
107     <P>
108     <b>const unsigned char *pcre_maketables(void);</b>
109     </P>
110     <P>
111     <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
112     <b>int <i>what</i>, void *<i>where</i>);</b>
113     </P>
114     <P>
115     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
116     <b>*<i>firstcharptr</i>);</b>
117     </P>
118     <P>
119 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
120     </P>
121     <P>
122 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
123     </P>
124     <P>
125     <b>char *pcre_version(void);</b>
126     </P>
127     <P>
128     <b>void *(*pcre_malloc)(size_t);</b>
129     </P>
130     <P>
131     <b>void (*pcre_free)(void *);</b>
132     </P>
133     <P>
134 nigel 73 <b>void *(*pcre_stack_malloc)(size_t);</b>
135     </P>
136     <P>
137     <b>void (*pcre_stack_free)(void *);</b>
138     </P>
139     <P>
140 nigel 63 <b>int (*pcre_callout)(pcre_callout_block *);</b>
141     </P>
142 nigel 75 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
143 nigel 63 <P>
144 nigel 93 PCRE has its own native API, which is described in this document. There are
145     also some wrapper functions that correspond to the POSIX regular expression
146 nigel 77 API. These are described in the
147 nigel 75 <a href="pcreposix.html"><b>pcreposix</b></a>
148 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
149     wrapper is distributed with PCRE. It is documented in the
150     <a href="pcrecpp.html"><b>pcrecpp</b></a>
151     page.
152 nigel 63 </P>
153     <P>
154 nigel 77 The native API C function prototypes are defined in the header file
155     <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
156     It can normally be accessed by adding <b>-lpcre</b> to the command for linking
157     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
158     and PCRE_MINOR to contain the major and minor release numbers for the library.
159 nigel 75 Applications can use these to include support for different releases of PCRE.
160 nigel 63 </P>
161     <P>
162 nigel 77 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
163     and <b>pcre_exec()</b> are used for compiling and matching regular expressions
164     in a Perl-compatible manner. A sample program that demonstrates the simplest
165     way of using them is provided in the file called <i>pcredemo.c</i> in the source
166     distribution. The
167 nigel 75 <a href="pcresample.html"><b>pcresample</b></a>
168     documentation describes how to run it.
169 nigel 63 </P>
170     <P>
171 nigel 77 A second matching function, <b>pcre_dfa_exec()</b>, which is not
172     Perl-compatible, is also provided. This uses a different algorithm for the
173 nigel 91 matching. The alternative algorithm finds all possible matches (at a given
174 nigel 93 point in the subject), and scans the subject just once. However, this algorithm
175     does not return captured substrings. A description of the two matching
176     algorithms and their advantages and disadvantages is given in the
177 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
178     documentation.
179     </P>
180     <P>
181 nigel 75 In addition to the main compiling and matching functions, there are convenience
182 nigel 77 functions for extracting captured substrings from a subject string that is
183     matched by <b>pcre_exec()</b>. They are:
184 nigel 63 <pre>
185     <b>pcre_copy_substring()</b>
186     <b>pcre_copy_named_substring()</b>
187     <b>pcre_get_substring()</b>
188     <b>pcre_get_named_substring()</b>
189     <b>pcre_get_substring_list()</b>
190 nigel 75 <b>pcre_get_stringnumber()</b>
191 nigel 91 <b>pcre_get_stringtable_entries()</b>
192 nigel 75 </pre>
193 nigel 63 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
194     provided, to free the memory used for extracted strings.
195     </P>
196     <P>
197 nigel 75 The function <b>pcre_maketables()</b> is used to build a set of character tables
198 nigel 77 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
199     or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
200     specialist use. Most commonly, no special tables are passed, in which case
201     internal tables that are generated when PCRE is built are used.
202 nigel 63 </P>
203     <P>
204     The function <b>pcre_fullinfo()</b> is used to find out information about a
205 nigel 75 compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
206 nigel 63 some of the available information, but is retained for backwards compatibility.
207     The function <b>pcre_version()</b> returns a pointer to a string containing the
208     version of PCRE and its date of release.
209     </P>
210     <P>
211 nigel 77 The function <b>pcre_refcount()</b> maintains a reference count in a data block
212     containing a compiled pattern. This is provided for the benefit of
213     object-oriented applications.
214     </P>
215     <P>
216 nigel 63 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
217 nigel 75 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
218 nigel 63 respectively. PCRE calls the memory management functions via these variables,
219     so a calling program can replace them if it wishes to intercept the calls. This
220     should be done before calling any PCRE functions.
221     </P>
222     <P>
223 nigel 73 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
224     indirections to memory management functions. These special functions are used
225     only when PCRE is compiled to use the heap for remembering data, instead of
226 nigel 91 recursive function calls, when running the <b>pcre_exec()</b> function. See the
227     <a href="pcrebuild.html"><b>pcrebuild</b></a>
228     documentation for details of how to do this. It is a non-standard way of
229     building PCRE, for use in environments that have limited stacks. Because of the
230     greater use of memory management, it runs more slowly. Separate functions are
231     provided so that special-purpose external code can be used for this case. When
232     used, these functions are always called in a stack-like manner (last obtained,
233     first freed), and always for memory blocks of the same size. There is a
234     discussion about PCRE's stack usage in the
235     <a href="pcrestack.html"><b>pcrestack</b></a>
236     documentation.
237 nigel 73 </P>
238     <P>
239 nigel 63 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
240     by the caller to a "callout" function, which PCRE will then call at specified
241 nigel 75 points during a matching operation. Details are given in the
242     <a href="pcrecallout.html"><b>pcrecallout</b></a>
243 nigel 63 documentation.
244     </P>
245 nigel 91 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
246 nigel 63 <P>
247 nigel 93 PCRE supports four different conventions for indicating line breaks in
248     strings: a single CR (carriage return) character, a single LF (linefeed)
249     character, the two-character sequence CRLF, or any Unicode newline sequence.
250     The Unicode newline sequences are the three just mentioned, plus the single
251     characters VT (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line,
252     U+0085), LS (line separator, U+2028), and PS (paragraph separator, U+2029).
253     </P>
254     <P>
255     Each of the first three conventions is used by at least one operating system as
256     its standard newline sequence. When PCRE is built, a default can be specified.
257     The default default is LF, which is the Unix standard. When PCRE is run, the
258     default can be overridden, either when a pattern is compiled, or when it is
259     matched.
260     </P>
261     <P>
262 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
263 nigel 93 pair of characters that indicate a line break". The choice of newline
264     convention affects the handling of the dot, circumflex, and dollar
265     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
266     recognized line ending sequence, the match position advancement for a
267     non-anchored pattern. The choice of newline convention does not affect the
268     interpretation of the \n or \r escape sequences.
269 nigel 91 </P>
270     <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
271     <P>
272 nigel 63 The PCRE functions can be used in multi-threading applications, with the
273 nigel 73 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
274     <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
275     callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
276 nigel 63 </P>
277     <P>
278     The compiled form of a regular expression is not altered during matching, so
279     the same compiled pattern can safely be used by several threads at once.
280     </P>
281 nigel 91 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
282 nigel 63 <P>
283 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
284     time, possibly by a different program, and even on a host other than the one on
285     which it was compiled. Details are given in the
286     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
287     documentation.
288     </P>
289 nigel 91 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
290 nigel 75 <P>
291 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
292     </P>
293     <P>
294     The function <b>pcre_config()</b> makes it possible for a PCRE client to
295     discover which optional features have been compiled into the PCRE library. The
296     <a href="pcrebuild.html"><b>pcrebuild</b></a>
297     documentation has more details about these optional features.
298     </P>
299     <P>
300     The first argument for <b>pcre_config()</b> is an integer, specifying which
301     information is required; the second argument is a pointer to a variable into
302     which the information is placed. The following information is available:
303     <pre>
305 nigel 75 </pre>
306 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
307     otherwise it is set to zero.
308     <pre>
310     </pre>
311     The output is an integer that is set to one if support for Unicode character
312     properties is available; otherwise it is set to zero.
313     <pre>
315 nigel 75 </pre>
316 nigel 91 The output is an integer whose value specifies the default character sequence
317 nigel 93 that is recognized as meaning "newline". The four values that are supported
318     are: 10 for LF, 13 for CR, 3338 for CRLF, and -1 for ANY. The default should
319     normally be the standard sequence for your operating system.
320 nigel 63 <pre>
322 nigel 75 </pre>
323 nigel 63 The output is an integer that contains the number of bytes used for internal
324     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
325     allow larger regular expressions to be compiled, at the expense of slower
326     matching. The default value of 2 is sufficient for all but the most massive
327     patterns, since it allows the compiled pattern to be up to 64K in size.
328     <pre>
330 nigel 75 </pre>
331 nigel 63 The output is an integer that contains the threshold above which the POSIX
332     interface uses <b>malloc()</b> for output vectors. Further details are given in
333 nigel 75 the
334     <a href="pcreposix.html"><b>pcreposix</b></a>
335     documentation.
336 nigel 63 <pre>
338 nigel 75 </pre>
339 nigel 63 The output is an integer that gives the default limit for the number of
340     internal matching function calls in a <b>pcre_exec()</b> execution. Further
341     details are given with <b>pcre_exec()</b> below.
342 nigel 73 <pre>
344     </pre>
345     The output is an integer that gives the default limit for the depth of
346     recursion when calling the internal matching function in a <b>pcre_exec()</b>
347     execution. Further details are given with <b>pcre_exec()</b> below.
348     <pre>
350 nigel 75 </pre>
351 nigel 77 The output is an integer that is set to one if internal recursion when running
352     <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
353     to remember their state. This is the usual way that PCRE is compiled. The
354     output is zero if PCRE was compiled to use blocks of data on the heap instead
355     of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
356     <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
357     avoiding the use of the stack.
358 nigel 73 </P>
359 nigel 91 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
360 nigel 63 <P>
361     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
362     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
363     <b>const unsigned char *<i>tableptr</i>);</b>
364 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
365     <b>int *<i>errorcodeptr</i>,</b>
366     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
367     <b>const unsigned char *<i>tableptr</i>);</b>
368 nigel 63 </P>
369     <P>
370 nigel 77 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
371     called to compile a pattern into an internal form. The only difference between
372     the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
373     <i>errorcodeptr</i>, via which a numerical error code can be returned.
374 nigel 63 </P>
375     <P>
376 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
377     <i>pattern</i> argument. A pointer to a single block of memory that is obtained
378     via <b>pcre_malloc</b> is returned. This contains the compiled code and related
379     data. The <b>pcre</b> type is defined for the returned block; this is a typedef
380     for a structure whose contents are not externally defined. It is up to the
381 nigel 91 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
382 nigel 77 </P>
383     <P>
384 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
385     depend on memory location, the complete <b>pcre</b> data block is not
386 nigel 75 fully relocatable, because it may contain a copy of the <i>tableptr</i>
387     argument, which is an address (see below).
388 nigel 63 </P>
389     <P>
390 nigel 93 The <i>options</i> argument contains various bit settings that affect the
391 nigel 75 compilation. It should be zero if no options are required. The available
392     options are described below. Some of them, in particular, those that are
393     compatible with Perl, can also be set and unset from within the pattern (see
394     the detailed description in the
395     <a href="pcrepattern.html"><b>pcrepattern</b></a>
396     documentation). For these options, the contents of the <i>options</i> argument
397     specifies their initial settings at the start of compilation and execution. The
398 nigel 91 PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
399     matching as well as at compile time.
400 nigel 63 </P>
401     <P>
402     If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
403     Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
404     NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
405 nigel 87 error message. This is a static string that is part of the library. You must
406     not try to free it. The offset from the start of the pattern to the character
407     where the error was discovered is placed in the variable pointed to by
408 nigel 63 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
409     </P>
410     <P>
411 nigel 77 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
412     <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
413     returned via this argument in the event of an error. This is in addition to the
414     textual error message. Error codes and messages are listed below.
415     </P>
416     <P>
417 nigel 63 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
418 nigel 75 character tables that are built when PCRE is compiled, using the default C
419     locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
420     call to <b>pcre_maketables()</b>. This value is stored with the compiled
421     pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
422     passed to it. For more discussion, see the section on locale support below.
423 nigel 63 </P>
424     <P>
425     This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
426     <pre>
427     pcre *re;
428     const char *error;
429     int erroffset;
430     re = pcre_compile(
431     "^A.*Z", /* the pattern */
432     0, /* default options */
433     &error, /* for error message */
434     &erroffset, /* for error offset */
435     NULL); /* use default character tables */
436 nigel 75 </pre>
437     The following names for option bits are defined in the <b>pcre.h</b> header
438     file:
439 nigel 63 <pre>
441 nigel 75 </pre>
442 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
443 nigel 75 constrained to match only at the first matching point in the string that is
444 nigel 63 being searched (the "subject string"). This effect can also be achieved by
445     appropriate constructs in the pattern itself, which is the only way to do it in
446     Perl.
447     <pre>
448 nigel 75 PCRE_AUTO_CALLOUT
449     </pre>
450     If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
451     all with number 255, before each pattern item. For discussion of the callout
452     facility, see the
453     <a href="pcrecallout.html"><b>pcrecallout</b></a>
454     documentation.
455     <pre>
456 nigel 63 PCRE_CASELESS
457 nigel 75 </pre>
458 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
459     letters. It is equivalent to Perl's /i option, and it can be changed within a
460 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
461     concept of case for characters whose values are less than 128, so caseless
462     matching is always possible. For characters with higher values, the concept of
463     case is supported if PCRE is compiled with Unicode property support, but not
464     otherwise. If you want to use caseless matching for characters 128 and above,
465     you must ensure that PCRE is compiled with Unicode property support as well as
466     with UTF-8 support.
467 nigel 63 <pre>
469 nigel 75 </pre>
470 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
471     end of the subject string. Without this option, a dollar also matches
472 nigel 91 immediately before a newline at the end of the string (but not before any other
473     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
474     There is no equivalent to this option in Perl, and no way to set it within a
475     pattern.
476 nigel 63 <pre>
478 nigel 75 </pre>
479 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
480 nigel 91 including those that indicate newline. Without it, a dot does not match when
481     the current position is at a newline. This option is equivalent to Perl's /s
482     option, and it can be changed within a pattern by a (?s) option setting. A
483 nigel 93 negative class such as [^a] always matches newline characters, independent of
484     the setting of this option.
485 nigel 63 <pre>
486 nigel 91 PCRE_DUPNAMES
487     </pre>
488     If this bit is set, names used to identify capturing subpatterns need not be
489     unique. This can be helpful for certain types of pattern when it is known that
490     only one instance of the named subpattern can ever be matched. There are more
491     details of named subpatterns below; see also the
492     <a href="pcrepattern.html"><b>pcrepattern</b></a>
493     documentation.
494     <pre>
495 nigel 63 PCRE_EXTENDED
496 nigel 75 </pre>
497 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
498     ignored except when escaped or inside a character class. Whitespace does not
499     include the VT character (code 11). In addition, characters between an
500 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
501     ignored. This is equivalent to Perl's /x option, and it can be changed within a
502     pattern by a (?x) option setting.
503 nigel 63 </P>
504     <P>
505     This option makes it possible to include comments inside complicated patterns.
506     Note, however, that this applies only to data characters. Whitespace characters
507     may never appear within special character sequences in a pattern, for example
508     within the sequence (?( which introduces a conditional subpattern.
509     <pre>
510     PCRE_EXTRA
511 nigel 75 </pre>
512 nigel 63 This option was invented in order to turn on additional functionality of PCRE
513     that is incompatible with Perl, but it is currently of very little use. When
514     set, any backslash in a pattern that is followed by a letter that has no
515     special meaning causes an error, thus reserving these combinations for future
516     expansion. By default, as in Perl, a backslash followed by a letter with no
517 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
518     give a warning for this.) There are at present no other features controlled by
519     this option. It can also be set by a (?X) option setting within a pattern.
520 nigel 63 <pre>
521 nigel 77 PCRE_FIRSTLINE
522     </pre>
523     If this option is set, an unanchored pattern is required to match before or at
524 nigel 91 the first newline in the subject string, though the matched text may continue
525     over the newline.
526 nigel 77 <pre>
527 nigel 63 PCRE_MULTILINE
528 nigel 75 </pre>
529     By default, PCRE treats the subject string as consisting of a single line of
530     characters (even if it actually contains newlines). The "start of line"
531 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
532     line" metacharacter ($) matches only at the end of the string, or before a
533     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
534     Perl.
535     </P>
536     <P>
537     When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
538 nigel 91 match immediately following or immediately before internal newlines in the
539     subject string, respectively, as well as at the very start and end. This is
540     equivalent to Perl's /m option, and it can be changed within a pattern by a
541     (?m) option setting. If there are no newlines in a subject string, or no
542 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
543     <pre>
544 nigel 91 PCRE_NEWLINE_CR
547 nigel 93 PCRE_NEWLINE_ANY
548 nigel 91 </pre>
549     These options override the default newline definition that was chosen when PCRE
550     was built. Setting the first or the second specifies that a newline is
551 nigel 93 indicated by a single character (CR or LF, respectively). Setting
552     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
553     CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that any Unicode newline
554     sequence should be recognized. The Unicode newline sequences are the three just
555     mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
556     U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
557     (paragraph separator, U+2029). The last two are recognized only in UTF-8 mode.
558 nigel 91 </P>
559     <P>
560 nigel 93 The newline setting in the options word uses three bits that are treated
561     as a number, giving eight possibilities. Currently only five are used (default
562     plus the four values above). This means that if you set more than one newline
563     option, the combination may or may not be sensible. For example,
564     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
565     other combinations yield unused numbers and cause an error.
566     </P>
567     <P>
568     The only time that a line break is specially recognized when compiling a
569     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
570     class is encountered. This indicates a comment that lasts until after the next
571     line break sequence. In other circumstances, line break sequences are treated
572     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
573     as whitespace characters and are therefore ignored.
574     </P>
575     <P>
576     The newline option that is set at compile time becomes the default that is used
577     for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
578 nigel 91 <pre>
580 nigel 75 </pre>
581 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
582     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
583     were followed by ?: but named parentheses can still be used for capturing (and
584     they acquire numbers in the usual way). There is no equivalent of this option
585     in Perl.
586     <pre>
588 nigel 75 </pre>
589 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
590     greedy by default, but become greedy if followed by "?". It is not compatible
591     with Perl. It can also be set by a (?U) option setting within the pattern.
592     <pre>
593     PCRE_UTF8
594 nigel 75 </pre>
595 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
596     of UTF-8 characters instead of single-byte character strings. However, it is
597 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
598 nigel 63 of this option provokes an error. Details of how this option changes the
599     behaviour of PCRE are given in the
600     <a href="pcre.html#utf8support">section on UTF-8 support</a>
601     in the main
602     <a href="pcre.html"><b>pcre</b></a>
603     page.
604 nigel 71 <pre>
606 nigel 75 </pre>
607 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
608     automatically checked. If an invalid UTF-8 sequence of bytes is found,
609     <b>pcre_compile()</b> returns an error. If you already know that your pattern is
610     valid, and you want to skip this check for performance reasons, you can set the
611     PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
612     UTF-8 string as a pattern is undefined. It may cause your program to crash.
613 nigel 77 Note that this option can also be passed to <b>pcre_exec()</b> and
614     <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
615     strings.
616 nigel 71 </P>
617 nigel 91 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
618 nigel 63 <P>
619 nigel 77 The following table lists the error codes than may be returned by
620     <b>pcre_compile2()</b>, along with the error messages that may be returned by
621 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
622     out of use. To avoid confusion, they have not been re-used.
623 nigel 77 <pre>
624     0 no error
625     1 \ at end of pattern
626     2 \c at end of pattern
627     3 unrecognized character follows \
628     4 numbers out of order in {} quantifier
629     5 number too big in {} quantifier
630     6 missing terminating ] for character class
631     7 invalid escape sequence in character class
632     8 range out of order in character class
633     9 nothing to repeat
634 nigel 93 10 [this code is not in use]
635 nigel 77 11 internal error: unexpected repeat
636     12 unrecognized character after (?
637     13 POSIX named classes are supported only within a class
638     14 missing )
639     15 reference to non-existent subpattern
640     16 erroffset passed as NULL
641     17 unknown option bit(s) set
642     18 missing ) after comment
643 nigel 93 19 [this code is not in use]
644 nigel 77 20 regular expression too large
645     21 failed to get memory
646     22 unmatched parentheses
647     23 internal error: code overflow
648     24 unrecognized character after (?&#60;
649     25 lookbehind assertion is not fixed length
650 nigel 91 26 malformed number or name after (?(
651 nigel 77 27 conditional group contains more than two branches
652     28 assertion expected after (?(
653     29 (?R or (?digits must be followed by )
654     30 unknown POSIX class name
655     31 POSIX collating elements are not supported
656     32 this version of PCRE is not compiled with PCRE_UTF8 support
657 nigel 93 33 [this code is not in use]
658 nigel 77 34 character value in \x{...} sequence is too large
659     35 invalid condition (?(0)
660     36 \C not allowed in lookbehind assertion
661     37 PCRE does not support \L, \l, \N, \U, or \u
662     38 number after (?C is &#62; 255
663     39 closing ) for (?C expected
664     40 recursive call could loop indefinitely
665     41 unrecognized character after (?P
666 nigel 93 42 syntax error in subpattern name (missing terminator)
667 nigel 91 43 two named subpatterns have the same name
668 nigel 77 44 invalid UTF-8 string
669     45 support for \P, \p, and \X has not been compiled
670     46 malformed \P or \p sequence
671     47 unknown property name after \P or \p
672 nigel 91 48 subpattern name is too long (maximum 32 characters)
673     49 too many named subpatterns (maximum 10,000)
674     50 repeated subpattern is too long
675     51 octal value is greater than \377 (not in UTF-8 mode)
676 nigel 93 52 internal error: overran compiling workspace
677     53 internal error: previously-checked referenced subpattern not found
678     54 DEFINE group contains more than one branch
679     55 repeating a DEFINE group is not allowed
680     56 inconsistent NEWLINE options"
681 nigel 77 </PRE>
682     </P>
683 nigel 91 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
684 nigel 77 <P>
685     <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
686 nigel 63 <b>const char **<i>errptr</i>);</b>
687     </P>
688     <P>
689 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
690     more time analyzing it in order to speed up the time taken for matching. The
691 nigel 63 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
692 nigel 75 argument. If studying the pattern produces additional information that will
693     help speed up matching, <b>pcre_study()</b> returns a pointer to a
694     <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
695     results of the study.
696 nigel 63 </P>
697     <P>
698 nigel 75 The returned value from <b>pcre_study()</b> can be passed directly to
699     <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
700 nigel 63 fields that can be set by the caller before the block is passed; these are
701 nigel 75 described
702     <a href="#extradata">below</a>
703     in the section on matching a pattern.
704 nigel 63 </P>
705     <P>
706 nigel 77 If studying the pattern does not produce any additional information
707 nigel 75 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
708     wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
709     own <b>pcre_extra</b> block.
710 nigel 63 </P>
711     <P>
712 nigel 75 The second argument of <b>pcre_study()</b> contains option bits. At present, no
713     options are defined, and this argument should always be zero.
714     </P>
715     <P>
716 nigel 63 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
717     studying succeeds (even if no data is returned), the variable it points to is
718 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
719     static string that is part of the library. You must not try to free it. You
720     should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
721     sure that it has run successfully.
722 nigel 63 </P>
723     <P>
724     This is a typical call to <b>pcre_study</b>():
725     <pre>
726     pcre_extra *pe;
727     pe = pcre_study(
728     re, /* result of pcre_compile() */
729     0, /* no options exist */
730     &error); /* set to NULL or points to a message */
731 nigel 75 </pre>
732 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
733     not have a single fixed starting character. A bitmap of possible starting
734 nigel 75 bytes is created.
735     <a name="localesupport"></a></P>
736 nigel 91 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
737 nigel 63 <P>
738 nigel 77 PCRE handles caseless matching, and determines whether characters are letters
739 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
740 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
741 nigel 75 less than 128. Higher-valued codes never match escapes such as \w or \d, but
742     can be tested with \p if PCRE is built with Unicode character property
743 nigel 87 support. The use of locales with Unicode is discouraged.
744 nigel 63 </P>
745     <P>
746 nigel 75 An internal set of tables is created in the default C locale when PCRE is
747     built. This is used when the final argument of <b>pcre_compile()</b> is NULL,
748     and is sufficient for many applications. An alternative set of tables can,
749     however, be supplied. These may be created in a different locale from the
750     default. As more and more applications change to using Unicode, the need for
751     this locale support is expected to die away.
752 nigel 63 </P>
753     <P>
754 nigel 75 External tables are built by calling the <b>pcre_maketables()</b> function,
755     which has no arguments, in the relevant locale. The result can then be passed
756     to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
757     example, to build and use tables that are appropriate for the French locale
758     (where accented characters with values greater than 128 are treated as letters),
759     the following code could be used:
760 nigel 63 <pre>
761 nigel 75 setlocale(LC_CTYPE, "fr_FR");
762 nigel 63 tables = pcre_maketables();
763     re = pcre_compile(..., tables);
764 nigel 75 </pre>
765     When <b>pcre_maketables()</b> runs, the tables are built in memory that is
766     obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
767     that the memory containing the tables remains available for as long as it is
768     needed.
769 nigel 63 </P>
770     <P>
771 nigel 75 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
772 nigel 63 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
773 nigel 75 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
774     pattern, compilation, studying and matching all happen in the same locale, but
775     different patterns can be compiled in different locales.
776 nigel 63 </P>
777     <P>
778 nigel 75 It is possible to pass a table pointer or NULL (indicating the use of the
779     internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
780     this facility could be used to match a pattern in a different locale from the
781     one in which it was compiled. Passing table pointers at run time is discussed
782     below in the section on matching a pattern.
783     </P>
784 nigel 91 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
785 nigel 75 <P>
786 nigel 63 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
787     <b>int <i>what</i>, void *<i>where</i>);</b>
788     </P>
789     <P>
790     The <b>pcre_fullinfo()</b> function returns information about a compiled
791     pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
792     nevertheless retained for backwards compability (and is documented below).
793     </P>
794     <P>
795     The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
796     pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
797     the pattern was not studied. The third argument specifies which piece of
798     information is required, and the fourth argument is a pointer to a variable
799     to receive the data. The yield of the function is zero for success, or one of
800     the following negative numbers:
801     <pre>
802     PCRE_ERROR_NULL the argument <i>code</i> was NULL
803     the argument <i>where</i> was NULL
804     PCRE_ERROR_BADMAGIC the "magic number" was not found
805     PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
806 nigel 75 </pre>
807     The "magic number" is placed at the start of each compiled pattern as an simple
808     check against passing an arbitrary memory pointer. Here is a typical call of
809     <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
810 nigel 63 <pre>
811     int rc;
812 nigel 91 size_t length;
813 nigel 63 rc = pcre_fullinfo(
814     re, /* result of pcre_compile() */
815     pe, /* result of pcre_study(), or NULL */
816     PCRE_INFO_SIZE, /* what is required */
817     &length); /* where to put the data */
818 nigel 75 </pre>
819 nigel 63 The possible values for the third argument are defined in <b>pcre.h</b>, and are
820     as follows:
821     <pre>
823 nigel 75 </pre>
824 nigel 63 Return the number of the highest back reference in the pattern. The fourth
825     argument should point to an <b>int</b> variable. Zero is returned if there are
826     no back references.
827     <pre>
829 nigel 75 </pre>
830 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
831 nigel 75 should point to an <b>int</b> variable.
832 nigel 63 <pre>
834 nigel 75 </pre>
835     Return a pointer to the internal default character tables within PCRE. The
836     fourth argument should point to an <b>unsigned char *</b> variable. This
837     information call is provided for internal use by the <b>pcre_study()</b>
838     function. External callers can cause PCRE to use its internal tables by passing
839     a NULL table pointer.
840     <pre>
842 nigel 75 </pre>
843 nigel 63 Return information about the first byte of any matched string, for a
844 nigel 91 non-anchored pattern. The fourth argument should point to an <b>int</b>
845     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
846     still recognized for backwards compatibility.)
847 nigel 63 </P>
848     <P>
849 nigel 75 If there is a fixed first byte, for example, from a pattern such as
850 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
851 nigel 75 <br>
852     <br>
853 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
854     starts with "^", or
855 nigel 75 <br>
856     <br>
857 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
858     (if it were set, the pattern would be anchored),
859 nigel 75 <br>
860     <br>
861 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
862     subject string or after any newline within the string. Otherwise -2 is
863     returned. For anchored patterns, -2 is returned.
864     <pre>
866 nigel 75 </pre>
867 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
868     table indicating a fixed set of bytes for the first byte in any matching
869     string, a pointer to the table is returned. Otherwise NULL is returned. The
870     fourth argument should point to an <b>unsigned char *</b> variable.
871     <pre>
873 nigel 75 </pre>
874 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
875     string, other than at its start, if such a byte has been recorded. The fourth
876     argument should point to an <b>int</b> variable. If there is no such byte, -1 is
877     returned. For anchored patterns, a last literal byte is recorded only if it
878     follows something of variable length. For example, for the pattern
879     /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
880     is -1.
881 nigel 63 <pre>
885 nigel 75 </pre>
886 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
887     names are just an additional way of identifying the parentheses, which still
888 nigel 91 acquire numbers. Several convenience functions such as
889     <b>pcre_get_named_substring()</b> are provided for extracting captured
890     substrings by name. It is also possible to extract the data directly, by first
891     converting the name to a number in order to access the correct pointers in the
892     output vector (described with <b>pcre_exec()</b> below). To do the conversion,
893     you need to use the name-to-number map, which is described by these three
894     values.
895 nigel 63 </P>
896     <P>
897     The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
898     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
899     entry; both of these return an <b>int</b> value. The entry size depends on the
900     length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
901     entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
902     are the number of the capturing parenthesis, most significant byte first. The
903     rest of the entry is the corresponding name, zero terminated. The names are in
904 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
905     their parentheses numbers. For example, consider the following pattern (assume
906 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
907     <pre>
908 nigel 93 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
909 nigel 75 </pre>
910 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
911     in the table is eight bytes long. The table is as follows, with non-printing
912 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
913 nigel 63 <pre>
914     00 01 d a t e 00 ??
915     00 05 d a y 00 ?? ??
916     00 04 m o n t h 00
917     00 02 y e a r 00 ??
918 nigel 75 </pre>
919     When writing code to extract data from named subpatterns using the
920 nigel 91 name-to-number map, remember that the length of the entries is likely to be
921 nigel 75 different for each compiled pattern.
922 nigel 63 <pre>
924 nigel 75 </pre>
925 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
926     argument should point to an <b>unsigned long int</b> variable. These option bits
927     are those specified in the call to <b>pcre_compile()</b>, modified by any
928     top-level option settings within the pattern itself.
929     </P>
930     <P>
931     A pattern is automatically anchored by PCRE if all of its top-level
932     alternatives begin with one of the following:
933     <pre>
934     ^ unless PCRE_MULTILINE is set
935     \A always
936     \G always
937 nigel 75 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
938     </pre>
939 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
940     <b>pcre_fullinfo()</b>.
941     <pre>
943 nigel 75 </pre>
944 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
945     the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
946     place the compiled data. The fourth argument should point to a <b>size_t</b>
947     variable.
948     <pre>
950 nigel 75 </pre>
951     Return the size of the data block pointed to by the <i>study_data</i> field in
952 nigel 63 a <b>pcre_extra</b> block. That is, it is the value that was passed to
953     <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
954     created by <b>pcre_study()</b>. The fourth argument should point to a
955     <b>size_t</b> variable.
956     </P>
957 nigel 91 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
958 nigel 63 <P>
959     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
960     <b>*<i>firstcharptr</i>);</b>
961     </P>
962     <P>
963     The <b>pcre_info()</b> function is now obsolete because its interface is too
964     restrictive to return all the available data about a compiled pattern. New
965     programs should use <b>pcre_fullinfo()</b> instead. The yield of
966     <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
967     following negative numbers:
968     <pre>
969     PCRE_ERROR_NULL the argument <i>code</i> was NULL
970     PCRE_ERROR_BADMAGIC the "magic number" was not found
971 nigel 75 </pre>
972 nigel 63 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
973     pattern was compiled is placed in the integer it points to (see
974     PCRE_INFO_OPTIONS above).
975     </P>
976     <P>
977     If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
978     it is used to pass back information about the first character of any matched
979     string (see PCRE_INFO_FIRSTBYTE above).
980     </P>
981 nigel 91 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
982 nigel 63 <P>
983 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
984     </P>
985     <P>
986     The <b>pcre_refcount()</b> function is used to maintain a reference count in the
987     data block that contains a compiled pattern. It is provided for the benefit of
988     applications that operate in an object-oriented manner, where different parts
989     of the application may be using the same compiled pattern, but you want to free
990     the block when they are all done.
991     </P>
992     <P>
993     When a pattern is compiled, the reference count field is initialized to zero.
994     It is changed only by calling this function, whose action is to add the
995     <i>adjust</i> value (which may be positive or negative) to it. The yield of the
996     function is the new value. However, the value of the count is constrained to
997     lie between 0 and 65535, inclusive. If the new value is outside these limits,
998     it is forced to the appropriate limit value.
999     </P>
1000     <P>
1001     Except when it is zero, the reference count is not correctly preserved if a
1002     pattern is compiled on one host and then transferred to a host whose byte-order
1003     is different. (This seems a highly unlikely scenario.)
1004     </P>
1005 nigel 91 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1006 nigel 77 <P>
1007 nigel 63 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1008     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1009     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1010     </P>
1011     <P>
1012     The function <b>pcre_exec()</b> is called to match a subject string against a
1013 nigel 75 compiled pattern, which is passed in the <i>code</i> argument. If the
1014 nigel 63 pattern has been studied, the result of the study should be passed in the
1015 nigel 77 <i>extra</i> argument. This function is the main matching facility of the
1016     library, and it operates in a Perl-like manner. For specialist use there is
1017     also an alternative matching function, which is described
1018     <a href="#dfamatch">below</a>
1019     in the section about the <b>pcre_dfa_exec()</b> function.
1020 nigel 63 </P>
1021     <P>
1022 nigel 75 In most applications, the pattern will have been compiled (and optionally
1023     studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1024     possible to save compiled patterns and study data, and then use them later
1025     in different processes, possibly even on different hosts. For a discussion
1026     about this, see the
1027     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1028     documentation.
1029 nigel 63 </P>
1030     <P>
1031 nigel 75 Here is an example of a simple call to <b>pcre_exec()</b>:
1032 nigel 63 <pre>
1033     int rc;
1034     int ovector[30];
1035     rc = pcre_exec(
1036     re, /* result of pcre_compile() */
1037     NULL, /* we didn't study the pattern */
1038     "some string", /* the subject string */
1039     11, /* the length of the subject string */
1040     0, /* start at offset 0 in the subject */
1041     0, /* default options */
1042 nigel 75 ovector, /* vector of integers for substring information */
1043 nigel 77 30); /* number of elements (NOT size in bytes) */
1044 nigel 75 <a name="extradata"></a></PRE>
1045 nigel 63 </P>
1046 nigel 75 <br><b>
1047     Extra data for <b>pcre_exec()</b>
1048     </b><br>
1049 nigel 63 <P>
1050     If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1051     data block. The <b>pcre_study()</b> function returns such a block (when it
1052     doesn't return NULL), but you can also create one for yourself, and pass
1053 nigel 87 additional information in it. The <b>pcre_extra</b> block contains the following
1054     fields (not necessarily in this order):
1055 nigel 63 <pre>
1056     unsigned long int <i>flags</i>;
1057     void *<i>study_data</i>;
1058     unsigned long int <i>match_limit</i>;
1059 nigel 87 unsigned long int <i>match_limit_recursion</i>;
1060 nigel 63 void *<i>callout_data</i>;
1061 nigel 75 const unsigned char *<i>tables</i>;
1062     </pre>
1063 nigel 63 The <i>flags</i> field is a bitmap that specifies which of the other fields
1064     are set. The flag bits are:
1065     <pre>
1070 nigel 75 PCRE_EXTRA_TABLES
1071     </pre>
1072 nigel 63 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1073     <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1074 nigel 75 the appropriate flag bit. You should not set this yourself, but you may add to
1075     the block by setting the other fields and their corresponding flag bits.
1076 nigel 63 </P>
1077     <P>
1078     The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1079     vast amount of resources when running patterns that are not going to match,
1080     but which have a very large number of possibilities in their search trees. The
1081 nigel 75 classic example is the use of nested unlimited repeats.
1082 nigel 63 </P>
1083     <P>
1084 nigel 75 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1085 nigel 87 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1086     number of times this function is called during a match, which has the effect of
1087     limiting the amount of backtracking that can take place. For patterns that are
1088     not anchored, the count restarts from zero for each position in the subject
1089     string.
1090 nigel 75 </P>
1091     <P>
1092 nigel 87 The default value for the limit can be set when PCRE is built; the default
1093 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1094 nigel 87 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1095     block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1096     the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1098 nigel 63 </P>
1099     <P>
1100 nigel 87 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1101     instead of limiting the total number of times that <b>match()</b> is called, it
1102     limits the depth of recursion. The recursion depth is a smaller number than the
1103     total number of calls, because not all calls to <b>match()</b> are recursive.
1104     This limit is of use only if it is set smaller than <i>match_limit</i>.
1105     </P>
1106     <P>
1107     Limiting the recursion depth limits the amount of stack that can be used, or,
1108     when PCRE has been compiled to use memory on the heap instead of the stack, the
1109     amount of heap memory that can be used.
1110     </P>
1111     <P>
1112     The default value for <i>match_limit_recursion</i> can be set when PCRE is
1113     built; the default default is the same value as the default for
1114     <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1115     with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1116     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1117     is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1118     </P>
1119     <P>
1120 nigel 63 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1121 nigel 75 which is described in the
1122     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1123     documentation.
1124 nigel 63 </P>
1125     <P>
1126 nigel 75 The <i>tables</i> field is used to pass a character tables pointer to
1127     <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1128     pattern. A non-NULL value is stored with the compiled pattern only if custom
1129     tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1130     If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1131     internal tables to be used. This facility is helpful when re-using patterns
1132     that have been saved after compiling with an external set of tables, because
1133     the external tables might be at a different address when <b>pcre_exec()</b> is
1134     called. See the
1135     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1136     documentation for a discussion of saving compiled patterns for later use.
1137 nigel 63 </P>
1138 nigel 75 <br><b>
1139     Option bits for <b>pcre_exec()</b>
1140     </b><br>
1141 nigel 63 <P>
1142 nigel 75 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1143 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1145 nigel 63 <pre>
1146 nigel 75 PCRE_ANCHORED
1147     </pre>
1148     The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1149     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1150     to be anchored by virtue of its contents, it cannot be made unachored at
1151     matching time.
1152     <pre>
1153 nigel 91 PCRE_NEWLINE_CR
1156 nigel 93 PCRE_NEWLINE_ANY
1157 nigel 91 </pre>
1158     These options override the newline definition that was chosen or defaulted when
1159 nigel 93 the pattern was compiled. For details, see the description of
1160     <b>pcre_compile()</b> above. During matching, the newline choice affects the
1161     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1162     the way the match position is advanced after a match failure for an unanchored
1163     pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1164     fails when the current position is at a CRLF sequence, the match position is
1165     advanced by two characters instead of one, in other words, to after the CRLF.
1166 nigel 91 <pre>
1167 nigel 63 PCRE_NOTBOL
1168 nigel 75 </pre>
1169     This option specifies that first character of the subject string is not the
1170     beginning of a line, so the circumflex metacharacter should not match before
1171     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1172     never to match. This option affects only the behaviour of the circumflex
1173     metacharacter. It does not affect \A.
1174 nigel 63 <pre>
1175     PCRE_NOTEOL
1176 nigel 75 </pre>
1177     This option specifies that the end of the subject string is not the end of a
1178     line, so the dollar metacharacter should not match it nor (except in multiline
1179     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1180     compile time) causes dollar never to match. This option affects only the
1181     behaviour of the dollar metacharacter. It does not affect \Z or \z.
1182 nigel 63 <pre>
1184 nigel 75 </pre>
1185 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1186     there are alternatives in the pattern, they are tried. If all the alternatives
1187     match the empty string, the entire match fails. For example, if the pattern
1188     <pre>
1189     a?b?
1190 nigel 75 </pre>
1191 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1192     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1193     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1194     </P>
1195     <P>
1196     Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1197     of a pattern match of the empty string within its <b>split()</b> function, and
1198     when using the /g modifier. It is possible to emulate Perl's behaviour after
1199     matching a null string by first trying the match again at the same offset with
1200 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1201     starting offset (see below) and trying an ordinary match again. There is some
1202     code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1203     <pre>
1205     </pre>
1206     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1207     string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1208     The value of <i>startoffset</i> is also checked to ensure that it points to the
1209     start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1210     <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1211     contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1212 nigel 63 </P>
1213     <P>
1214 nigel 75 If you already know that your subject is valid, and you want to skip these
1215     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1216     calling <b>pcre_exec()</b>. You might want to do this for the second and
1217     subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1218     all the matches in a single subject string. However, you should be sure that
1219     the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1220     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1221     subject, or a value of <i>startoffset</i> that does not point to the start of a
1222     UTF-8 character, is undefined. Your program may crash.
1223     <pre>
1225     </pre>
1226     This option turns on the partial matching feature. If the subject string fails
1227     to match the pattern, but at some point during the matching process the end of
1228     the subject was reached (that is, the subject partially matches the pattern and
1229     the failure to match occurred only because there were not enough subject
1230     characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1231     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1232     may appear in the pattern. These are discussed in the
1233     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1234     documentation.
1235     </P>
1236     <br><b>
1237     The string to be matched by <b>pcre_exec()</b>
1238     </b><br>
1239     <P>
1240 nigel 63 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1241 nigel 73 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1242 nigel 75 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1243     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1244     bytes. When the starting offset is zero, the search for a match starts at the
1245     beginning of the subject, and this is by far the most common case.
1246 nigel 63 </P>
1247     <P>
1248     A non-zero starting offset is useful when searching for another match in the
1249     same subject by calling <b>pcre_exec()</b> again after a previous success.
1250     Setting <i>startoffset</i> differs from just passing over a shortened string and
1251     setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1252     lookbehind. For example, consider the pattern
1253     <pre>
1254     \Biss\B
1255 nigel 75 </pre>
1256 nigel 63 which finds occurrences of "iss" in the middle of words. (\B matches only if
1257     the current position in the subject is not a word boundary.) When applied to
1258     the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1259     occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1260     subject, namely "issipi", it does not match, because \B is always false at the
1261     start of the subject, which is deemed to be a word boundary. However, if
1262     <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1263     set to 4, it finds the second occurrence of "iss" because it is able to look
1264     behind the starting point to discover that it is preceded by a letter.
1265     </P>
1266     <P>
1267     If a non-zero starting offset is passed when the pattern is anchored, one
1268 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1269 nigel 63 pattern does not require the match to be at the start of the subject.
1270     </P>
1271 nigel 75 <br><b>
1272     How <b>pcre_exec()</b> returns captured substrings
1273     </b><br>
1274 nigel 63 <P>
1275     In general, a pattern matches a certain portion of the subject, and in
1276     addition, further substrings from the subject may be picked out by parts of the
1277     pattern. Following the usage in Jeffrey Friedl's book, this is called
1278     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1279     a fragment of a pattern that picks out a substring. PCRE supports several other
1280     kinds of parenthesized subpattern that do not cause substrings to be captured.
1281     </P>
1282     <P>
1283     Captured substrings are returned to the caller via a vector of integer offsets
1284     whose address is passed in <i>ovector</i>. The number of elements in the vector
1285 nigel 75 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1286     this argument is NOT the size of <i>ovector</i> in bytes.
1287 nigel 63 </P>
1288     <P>
1289 nigel 75 The first two-thirds of the vector is used to pass back captured substrings,
1290     each substring using a pair of integers. The remaining third of the vector is
1291     used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1292     and is not available for passing back information. The length passed in
1293     <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1294     rounded down.
1295     </P>
1296     <P>
1297     When a match is successful, information about captured substrings is returned
1298     in pairs of integers, starting at the beginning of <i>ovector</i>, and
1299 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1300     pair is set to the offset of the first character in a substring, and the second
1301     is set to the offset of the first character after the end of a substring. The
1302     first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1303     subject string matched by the entire pattern. The next pair is used for the
1304     first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1305 nigel 91 is one more than the highest numbered pair that has been set. For example, if
1306     two substrings have been captured, the returned value is 3. If there are no
1307     capturing subpatterns, the return value from a successful match is 1,
1308     indicating that just the first pair of offsets has been set.
1309 nigel 63 </P>
1310     <P>
1311     If a capturing subpattern is matched repeatedly, it is the last portion of the
1312 nigel 75 string that it matched that is returned.
1313 nigel 63 </P>
1314     <P>
1315 nigel 75 If the vector is too small to hold all the captured substring offsets, it is
1316     used as far as possible (up to two-thirds of its length), and the function
1317     returns a value of zero. In particular, if the substring offsets are not of
1318     interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1319 nigel 63 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1320 nigel 75 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1321     has to get additional memory for use during matching. Thus it is usually
1322     advisable to supply an <i>ovector</i>.
1323 nigel 63 </P>
1324     <P>
1325 nigel 91 The <b>pcre_info()</b> function can be used to find out how many capturing
1326 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1327     <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1328     the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1329 nigel 91 </P>
1330     <P>
1331     It is possible for capturing subpattern number <i>n+1</i> to match some part of
1332     the subject when subpattern <i>n</i> has not been used at all. For example, if
1333     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1334     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1335     happens, both values in the offset pairs corresponding to unused subpatterns
1336     are set to -1.
1337     </P>
1338     <P>
1339     Offset values that correspond to unused subpatterns at the end of the
1340     expression are also set to -1. For example, if the string "abc" is matched
1341     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1342     return from the function is 2, because the highest used capturing subpattern
1343     number is 1. However, you can refer to the offsets for the second and third
1344     capturing subpatterns if you wish (assuming the vector is large enough, of
1345     course).
1346     </P>
1347     <P>
1348     Some convenience functions are provided for extracting the captured substrings
1349     as separate strings. These are described below.
1350 nigel 77 <a name="errorlist"></a></P>
1351 nigel 75 <br><b>
1352 nigel 91 Error return values from <b>pcre_exec()</b>
1353 nigel 75 </b><br>
1354 nigel 63 <P>
1355     If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1356     defined in the header file:
1357     <pre>
1358     PCRE_ERROR_NOMATCH (-1)
1359 nigel 75 </pre>
1360 nigel 63 The subject string did not match the pattern.
1361     <pre>
1362     PCRE_ERROR_NULL (-2)
1363 nigel 75 </pre>
1364 nigel 63 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1365     NULL and <i>ovecsize</i> was not zero.
1366     <pre>
1368 nigel 75 </pre>
1369 nigel 63 An unrecognized bit was set in the <i>options</i> argument.
1370     <pre>
1372 nigel 75 </pre>
1373 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1374 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1375     compiled in an environment of one endianness is run in an environment with the
1376     other endianness. This is the error that PCRE gives when the magic number is
1377     not present.
1378 nigel 63 <pre>
1379 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1380 nigel 75 </pre>
1381 nigel 63 While running the pattern match, an unknown item was encountered in the
1382     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1383     of the compiled pattern.
1384     <pre>
1386 nigel 75 </pre>
1387 nigel 63 If a pattern contains back references, but the <i>ovector</i> that is passed to
1388     <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1389     gets a block of memory at the start of matching to use for this purpose. If the
1390 nigel 75 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1391     automatically freed at the end of matching.
1392 nigel 63 <pre>
1394 nigel 75 </pre>
1395 nigel 63 This error is used by the <b>pcre_copy_substring()</b>,
1396     <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1397     below). It is never returned by <b>pcre_exec()</b>.
1398     <pre>
1400 nigel 75 </pre>
1401 nigel 87 The backtracking limit, as specified by the <i>match_limit</i> field in a
1402     <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1403     above.
1404     <pre>
1405 nigel 63 PCRE_ERROR_CALLOUT (-9)
1406 nigel 75 </pre>
1407 nigel 63 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1408     use by callout functions that want to yield a distinctive error code. See the
1409 nigel 75 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1410     documentation for details.
1411 nigel 71 <pre>
1412 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1413 nigel 75 </pre>
1414 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1415 nigel 73 <pre>
1417 nigel 75 </pre>
1418 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1419     of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1420 nigel 75 <pre>
1421 nigel 77 PCRE_ERROR_PARTIAL (-12)
1422 nigel 75 </pre>
1423     The subject string did not match, but it did match partially. See the
1424     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1425     documentation for details of partial matching.
1426     <pre>
1427 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1428 nigel 75 </pre>
1429     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1430     are not supported for partial matching. See the
1431     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1432     documentation for details of partial matching.
1433     <pre>
1434 nigel 77 PCRE_ERROR_INTERNAL (-14)
1435 nigel 75 </pre>
1436     An unexpected internal error has occurred. This error could be caused by a bug
1437     in PCRE or by overwriting of the compiled pattern.
1438     <pre>
1439 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1440 nigel 75 </pre>
1441     This error is given if the value of the <i>ovecsize</i> argument is negative.
1442 nigel 93 <pre>
1444     </pre>
1445     The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1446     field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1447     description above.
1448     <pre>
1450     </pre>
1451     When a group that can match an empty substring is repeated with an unbounded
1452     upper limit, the subject position at the start of the group must be remembered,
1453     so that a test for an empty string can be made when the end of the group is
1454     reached. Some workspace is required for this; if it runs out, this error is
1455     given.
1456     <pre>
1458     </pre>
1459     An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1460 nigel 73 </P>
1461 nigel 93 <P>
1462     Error numbers -16 to -20 are not used by <b>pcre_exec()</b>.
1463     </P>
1464 nigel 91 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1465 nigel 63 <P>
1466     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1467     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1468     <b>int <i>buffersize</i>);</b>
1469     </P>
1470     <P>
1471     <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1472     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1473     <b>const char **<i>stringptr</i>);</b>
1474     </P>
1475     <P>
1476     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1477     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1478     </P>
1479     <P>
1480     Captured substrings can be accessed directly by using the offsets returned by
1481     <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1482     <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1483     <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1484     as new, separate, zero-terminated strings. These functions identify substrings
1485     by number. The next section describes functions for extracting named
1486 nigel 91 substrings.
1487 nigel 63 </P>
1488     <P>
1489 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1490     further zero added on the end, but the result is not, of course, a C string.
1491     However, you can process such a string by referring to the length that is
1492     returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1493     Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1494     for handling strings containing binary zeros, because the end of the final
1495     string is not independently indicated.
1496     </P>
1497     <P>
1498 nigel 63 The first three arguments are the same for all three of these functions:
1499 nigel 75 <i>subject</i> is the subject string that has just been successfully matched,
1500 nigel 63 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1501     <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1502     captured by the match, including the substring that matched the entire regular
1503 nigel 75 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1504     than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1505     space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1506     number of elements in the vector divided by three.
1507 nigel 63 </P>
1508     <P>
1509     The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1510     extract a single substring, whose number is given as <i>stringnumber</i>. A
1511 nigel 75 value of zero extracts the substring that matched the entire pattern, whereas
1512 nigel 63 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1513     the string is placed in <i>buffer</i>, whose length is given by
1514     <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1515     obtained via <b>pcre_malloc</b>, and its address is returned via
1516     <i>stringptr</i>. The yield of the function is the length of the string, not
1517 nigel 93 including the terminating zero, or one of these error codes:
1518 nigel 63 <pre>
1520 nigel 75 </pre>
1521 nigel 63 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1522     memory failed for <b>pcre_get_substring()</b>.
1523     <pre>
1525 nigel 75 </pre>
1526 nigel 63 There is no substring whose number is <i>stringnumber</i>.
1527     </P>
1528     <P>
1529     The <b>pcre_get_substring_list()</b> function extracts all available substrings
1530     and builds a list of pointers to them. All this is done in a single block of
1531 nigel 75 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1532 nigel 63 is returned via <i>listptr</i>, which is also the start of the list of string
1533     pointers. The end of the list is marked by a NULL pointer. The yield of the
1534 nigel 93 function is zero if all went well, or the error code
1535 nigel 63 <pre>
1537 nigel 75 </pre>
1538 nigel 63 if the attempt to get the memory block failed.
1539     </P>
1540     <P>
1541     When any of these functions encounter a substring that is unset, which can
1542     happen when capturing subpattern number <i>n+1</i> matches some part of the
1543     subject, but subpattern <i>n</i> has not been used at all, they return an empty
1544     string. This can be distinguished from a genuine zero-length substring by
1545     inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1546     substrings.
1547     </P>
1548     <P>
1549     The two convenience functions <b>pcre_free_substring()</b> and
1550     <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1551     a previous call of <b>pcre_get_substring()</b> or
1552     <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1553     the function pointed to by <b>pcre_free</b>, which of course could be called
1554     directly from a C program. However, PCRE is used in some situations where it is
1555 nigel 91 linked via a special interface to another programming language that cannot use
1556 nigel 63 <b>pcre_free</b> directly; it is for these cases that the functions are
1557     provided.
1558     </P>
1559 nigel 91 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1560 nigel 63 <P>
1561 nigel 75 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1562     <b>const char *<i>name</i>);</b>
1563     </P>
1564     <P>
1565 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1566     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1567     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1568     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1569     </P>
1570     <P>
1571     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1572     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1573     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1574     <b>const char **<i>stringptr</i>);</b>
1575     </P>
1576     <P>
1577 nigel 75 To extract a substring by name, you first have to find associated number.
1578     For example, for this pattern
1579 nigel 63 <pre>
1580 nigel 93 (a+)b(?&#60;xxx&#62;\d+)...
1581 nigel 75 </pre>
1582 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1583     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1584     calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1585     pattern, and the second is the name. The yield of the function is the
1586 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1587     that name.
1588 nigel 63 </P>
1589     <P>
1590 nigel 75 Given the number, you can extract the substring directly, or use one of the
1591     functions described in the previous section. For convenience, there are also
1592     two functions that do the whole job.
1593 nigel 63 </P>
1594     <P>
1595 nigel 91 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1596     <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1597 nigel 75 functions that extract by number. As these are described in the previous
1598     section, they are not re-described here. There are just two differences:
1599 nigel 63 </P>
1600     <P>
1601     First, instead of a substring number, a substring name is given. Second, there
1602     is an extra argument, given at the start, which is a pointer to the compiled
1603     pattern. This is needed in order to gain access to the name-to-number
1604     translation table.
1605     </P>
1606     <P>
1607     These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1608     then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as
1609     appropriate.
1610     </P>
1611 nigel 91 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1612 nigel 63 <P>
1613 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1614     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1615     </P>
1616     <P>
1617     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1618     are not required to be unique. Normally, patterns with duplicate names are such
1619     that in any one match, only one of the named subpatterns participates. An
1620     example is shown in the
1621     <a href="pcrepattern.html"><b>pcrepattern</b></a>
1622     documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1623     and <b>pcre_get_named_substring()</b> return the first substring corresponding
1624     to the given name that is set. If none are set, an empty string is returned.
1625     The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1626     associated with the name, but it is not defined which it is.
1627     <br>
1628     <br>
1629     If you want to get full details of all captured substrings for a given name,
1630     you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1631     argument is the compiled pattern, and the second is the name. The third and
1632     fourth are pointers to variables which are updated by the function. After it
1633     has run, they point to the first and last entries in the name-to-number table
1634     for the given name. The function itself returns the length of each entry, or
1635 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1636     described above in the section entitled <i>Information about a pattern</i>.
1637     Given all the relevant entries for the name, you can extract each of their
1638     numbers, and hence the captured data, if any.
1639 nigel 91 </P>
1640     <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1641     <P>
1642 nigel 77 The traditional matching function uses a similar algorithm to Perl, which stops
1643     when it finds the first match, starting at a given point in the subject. If you
1644     want to find all possible matches, or the longest possible match, consider
1645     using the alternative matching function (see below) instead. If you cannot use
1646     the alternative function, but still need to find all possible matches, you
1647     can kludge it up by making use of the callout facility, which is described in
1648     the
1649     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1650     documentation.
1651     </P>
1652     <P>
1653     What you have to do is to insert a callout right at the end of the pattern.
1654     When your callout function is called, extract and save the current matched
1655     substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1656     other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1657     will yield PCRE_ERROR_NOMATCH.
1658     <a name="dfamatch"></a></P>
1659 nigel 91 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1660 nigel 77 <P>
1661     <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1662     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1663     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1664     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1665     </P>
1666     <P>
1667     The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1668 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1669     just once, and does not backtrack. This has different characteristics to the
1670     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1671     patterns are not supported. Nevertheless, there are times when this kind of
1672     matching can be useful. For a discussion of the two matching algorithms, see
1673     the
1674 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
1675     documentation.
1676     </P>
1677     <P>
1678     The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1679     <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1680     different way, and this is described below. The other common arguments are used
1681     in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1682     here.
1683     </P>
1684     <P>
1685     The two additional arguments provide workspace for the function. The workspace
1686     vector should contain at least 20 elements. It is used for keeping track of
1687     multiple paths through the pattern tree. More workspace will be needed for
1688 nigel 91 patterns and subjects where there are a lot of potential matches.
1689 nigel 77 </P>
1690     <P>
1691 nigel 87 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1692 nigel 77 <pre>
1693     int rc;
1694     int ovector[10];
1695     int wspace[20];
1696 nigel 87 rc = pcre_dfa_exec(
1697 nigel 77 re, /* result of pcre_compile() */
1698     NULL, /* we didn't study the pattern */
1699     "some string", /* the subject string */
1700     11, /* the length of the subject string */
1701     0, /* start at offset 0 in the subject */
1702     0, /* default options */
1703     ovector, /* vector of integers for substring information */
1704     10, /* number of elements (NOT size in bytes) */
1705     wspace, /* working space vector */
1706     20); /* number of elements (NOT size in bytes) */
1707     </PRE>
1708     </P>
1709     <br><b>
1710     Option bits for <b>pcre_dfa_exec()</b>
1711     </b><br>
1712     <P>
1713     The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1714 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1716 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1717     the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1718     <pre>
1720     </pre>
1721     This has the same general effect as it does for <b>pcre_exec()</b>, but the
1722     details are slightly different. When PCRE_PARTIAL is set for
1723     <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1724     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1725     complete matches, but there is still at least one matching possibility. The
1726     portion of the string that provided the partial match is set as the first
1727     matching string.
1728     <pre>
1730     </pre>
1731     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1732 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1733     works, this is necessarily the shortest possible match at the first possible
1734     matching point in the subject string.
1735 nigel 77 <pre>
1737     </pre>
1738     When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1739     a partial match, it is possible to call it again, with additional subject
1740     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1741     option requests this action; when it is set, the <i>workspace</i> and
1742     <i>wscount</i> options must reference the same vector as before because data
1743     about the match so far is left in them after a partial match. There is more
1744     discussion of this facility in the
1745     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1746     documentation.
1747     </P>
1748     <br><b>
1749     Successful returns from <b>pcre_dfa_exec()</b>
1750     </b><br>
1751     <P>
1752     When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1753     substring in the subject. Note, however, that all the matches from one run of
1754     the function start at the same point in the subject. The shorter matches are
1755     all initial substrings of the longer matches. For example, if the pattern
1756     <pre>
1757     &#60;.*&#62;
1758     </pre>
1759     is matched against the string
1760     <pre>
1761     This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1762     </pre>
1763     the three matched strings are
1764     <pre>
1765     &#60;something&#62;
1766     &#60;something&#62; &#60;something else&#62;
1767     &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1768     </pre>
1769     On success, the yield of the function is a number greater than zero, which is
1770     the number of matched substrings. The substrings themselves are returned in
1771     <i>ovector</i>. Each string uses two elements; the first is the offset to the
1772 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1773     the same start offset. (Space could have been saved by giving this only once,
1774     but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1775     returns data, even though the meaning of the strings is different.)
1776 nigel 77 </P>
1777     <P>
1778     The strings are returned in reverse order of length; that is, the longest
1779     matching string is given first. If there were too many matches to fit into
1780     <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1781     the longest matches.
1782     </P>
1783     <br><b>
1784     Error returns from <b>pcre_dfa_exec()</b>
1785     </b><br>
1786     <P>
1787     The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1788     Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1789     described
1790     <a href="#errorlist">above.</a>
1791     There are in addition the following errors that are specific to
1792     <b>pcre_dfa_exec()</b>:
1793     <pre>
1794     PCRE_ERROR_DFA_UITEM (-16)
1795     </pre>
1796     This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1797     that it does not support, for instance, the use of \C or a back reference.
1798     <pre>
1799     PCRE_ERROR_DFA_UCOND (-17)
1800     </pre>
1801 nigel 93 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1802     uses a back reference for the condition, or a test for recursion in a specific
1803     group. These are not supported.
1804 nigel 77 <pre>
1806     </pre>
1807     This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1808     block that contains a setting of the <i>match_limit</i> field. This is not
1809     supported (it is meaningless).
1810     <pre>
1811     PCRE_ERROR_DFA_WSSIZE (-19)
1812     </pre>
1813     This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1814     <i>workspace</i> vector.
1815     <pre>
1817     </pre>
1818     When a recursive subpattern is processed, the matching function calls itself
1819     recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1820     error is given if the output vector is not large enough. This should be
1821     extremely rare, as a vector of size 1000 is used.
1822     </P>
1823 nigel 93 <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1824 nigel 77 <P>
1825 nigel 93 <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1826     <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1827     <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1828     </P>
1829     <P>
1830     Last updated: 30 November 2006
1831 nigel 63 <br>
1832 nigel 87 Copyright &copy; 1997-2006 University of Cambridge.
1833 nigel 75 <p>
1834     Return to the <a href="index.html">PCRE index page</a>.
1835     </p>


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

ViewVC Help
Powered by ViewVC 1.1.12