ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 111 - (hide annotations) (download) (as text)
Thu Mar 8 16:53:09 2007 UTC (8 years, 1 month ago) by ph10
File MIME type: text/html
File size: 83591 byte(s)
Create the PrepareRelease script to process the documentation and create the 
.generic files for distribution, also to remove trailing spaces. Update a lot 
more of the build-time documentation. Arrange for PrepareRelease and its 
sub-scripts to be distributed.

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 ph10 111 <p>
11 nigel 75 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 ph10 111 <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 ph10 99 <li><a name="TOC21" href="#SEC21">AUTHOR</a>
37     <li><a name="TOC22" href="#SEC22">REVISION</a>
38 nigel 63 </ul>
39 nigel 75 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
40 nigel 63 <P>
41     <b>#include &#60;pcre.h&#62;</b>
42     </P>
43     <P>
44     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
45     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
46     <b>const unsigned char *<i>tableptr</i>);</b>
47     </P>
48     <P>
49 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
50     <b>int *<i>errorcodeptr</i>,</b>
51     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
52     <b>const unsigned char *<i>tableptr</i>);</b>
53     </P>
54     <P>
55 nigel 63 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
56     <b>const char **<i>errptr</i>);</b>
57     </P>
58     <P>
59     <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
60     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
61     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
62     </P>
63     <P>
64 nigel 77 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
65     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
66     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
67     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
68     </P>
69     <P>
70 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
71     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
72     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
73     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
74     </P>
75     <P>
76     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
77     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
78     <b>int <i>buffersize</i>);</b>
79     </P>
80     <P>
81     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
82     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
83     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
84     <b>const char **<i>stringptr</i>);</b>
85     </P>
86     <P>
87     <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
88     <b>const char *<i>name</i>);</b>
89     </P>
90     <P>
91 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
92     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
93     </P>
94     <P>
95 nigel 63 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
96     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
97     <b>const char **<i>stringptr</i>);</b>
98     </P>
99     <P>
100     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
101     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
102     </P>
103     <P>
104     <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
105     </P>
106     <P>
107     <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
108     </P>
109     <P>
110     <b>const unsigned char *pcre_maketables(void);</b>
111     </P>
112     <P>
113     <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
114     <b>int <i>what</i>, void *<i>where</i>);</b>
115     </P>
116     <P>
117     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
118     <b>*<i>firstcharptr</i>);</b>
119     </P>
120     <P>
121 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
122     </P>
123     <P>
124 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
125     </P>
126     <P>
127     <b>char *pcre_version(void);</b>
128     </P>
129     <P>
130     <b>void *(*pcre_malloc)(size_t);</b>
131     </P>
132     <P>
133     <b>void (*pcre_free)(void *);</b>
134     </P>
135     <P>
136 nigel 73 <b>void *(*pcre_stack_malloc)(size_t);</b>
137     </P>
138     <P>
139     <b>void (*pcre_stack_free)(void *);</b>
140     </P>
141     <P>
142 nigel 63 <b>int (*pcre_callout)(pcre_callout_block *);</b>
143     </P>
144 nigel 75 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
145 nigel 63 <P>
146 nigel 93 PCRE has its own native API, which is described in this document. There are
147     also some wrapper functions that correspond to the POSIX regular expression
148 nigel 77 API. These are described in the
149 nigel 75 <a href="pcreposix.html"><b>pcreposix</b></a>
150 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
151     wrapper is distributed with PCRE. It is documented in the
152     <a href="pcrecpp.html"><b>pcrecpp</b></a>
153     page.
154 nigel 63 </P>
155     <P>
156 nigel 77 The native API C function prototypes are defined in the header file
157     <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
158     It can normally be accessed by adding <b>-lpcre</b> to the command for linking
159     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
160     and PCRE_MINOR to contain the major and minor release numbers for the library.
161 nigel 75 Applications can use these to include support for different releases of PCRE.
162 nigel 63 </P>
163     <P>
164 nigel 77 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
165     and <b>pcre_exec()</b> are used for compiling and matching regular expressions
166     in a Perl-compatible manner. A sample program that demonstrates the simplest
167     way of using them is provided in the file called <i>pcredemo.c</i> in the source
168     distribution. The
169 nigel 75 <a href="pcresample.html"><b>pcresample</b></a>
170     documentation describes how to run it.
171 nigel 63 </P>
172     <P>
173 nigel 77 A second matching function, <b>pcre_dfa_exec()</b>, which is not
174     Perl-compatible, is also provided. This uses a different algorithm for the
175 nigel 91 matching. The alternative algorithm finds all possible matches (at a given
176 nigel 93 point in the subject), and scans the subject just once. However, this algorithm
177     does not return captured substrings. A description of the two matching
178     algorithms and their advantages and disadvantages is given in the
179 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
180     documentation.
181     </P>
182     <P>
183 nigel 75 In addition to the main compiling and matching functions, there are convenience
184 nigel 77 functions for extracting captured substrings from a subject string that is
185     matched by <b>pcre_exec()</b>. They are:
186 nigel 63 <pre>
187     <b>pcre_copy_substring()</b>
188     <b>pcre_copy_named_substring()</b>
189     <b>pcre_get_substring()</b>
190     <b>pcre_get_named_substring()</b>
191     <b>pcre_get_substring_list()</b>
192 nigel 75 <b>pcre_get_stringnumber()</b>
193 nigel 91 <b>pcre_get_stringtable_entries()</b>
194 nigel 75 </pre>
195 nigel 63 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
196     provided, to free the memory used for extracted strings.
197     </P>
198     <P>
199 nigel 75 The function <b>pcre_maketables()</b> is used to build a set of character tables
200 nigel 77 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
201     or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
202     specialist use. Most commonly, no special tables are passed, in which case
203     internal tables that are generated when PCRE is built are used.
204 nigel 63 </P>
205     <P>
206     The function <b>pcre_fullinfo()</b> is used to find out information about a
207 nigel 75 compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
208 nigel 63 some of the available information, but is retained for backwards compatibility.
209     The function <b>pcre_version()</b> returns a pointer to a string containing the
210     version of PCRE and its date of release.
211     </P>
212     <P>
213 nigel 77 The function <b>pcre_refcount()</b> maintains a reference count in a data block
214     containing a compiled pattern. This is provided for the benefit of
215     object-oriented applications.
216     </P>
217     <P>
218 nigel 63 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
219 nigel 75 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
220 nigel 63 respectively. PCRE calls the memory management functions via these variables,
221     so a calling program can replace them if it wishes to intercept the calls. This
222     should be done before calling any PCRE functions.
223     </P>
224     <P>
225 nigel 73 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
226     indirections to memory management functions. These special functions are used
227     only when PCRE is compiled to use the heap for remembering data, instead of
228 nigel 91 recursive function calls, when running the <b>pcre_exec()</b> function. See the
229     <a href="pcrebuild.html"><b>pcrebuild</b></a>
230     documentation for details of how to do this. It is a non-standard way of
231     building PCRE, for use in environments that have limited stacks. Because of the
232     greater use of memory management, it runs more slowly. Separate functions are
233     provided so that special-purpose external code can be used for this case. When
234     used, these functions are always called in a stack-like manner (last obtained,
235     first freed), and always for memory blocks of the same size. There is a
236     discussion about PCRE's stack usage in the
237     <a href="pcrestack.html"><b>pcrestack</b></a>
238     documentation.
239 nigel 73 </P>
240     <P>
241 nigel 63 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
242     by the caller to a "callout" function, which PCRE will then call at specified
243 nigel 75 points during a matching operation. Details are given in the
244     <a href="pcrecallout.html"><b>pcrecallout</b></a>
245 nigel 63 documentation.
246     </P>
247 nigel 91 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248 nigel 63 <P>
249 nigel 93 PCRE supports four different conventions for indicating line breaks in
250     strings: a single CR (carriage return) character, a single LF (linefeed)
251     character, the two-character sequence CRLF, or any Unicode newline sequence.
252     The Unicode newline sequences are the three just mentioned, plus the single
253     characters VT (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line,
254     U+0085), LS (line separator, U+2028), and PS (paragraph separator, U+2029).
255     </P>
256     <P>
257     Each of the first three conventions is used by at least one operating system as
258     its standard newline sequence. When PCRE is built, a default can be specified.
259     The default default is LF, which is the Unix standard. When PCRE is run, the
260     default can be overridden, either when a pattern is compiled, or when it is
261     matched.
262     </P>
263     <P>
264 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
265 nigel 93 pair of characters that indicate a line break". The choice of newline
266     convention affects the handling of the dot, circumflex, and dollar
267     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
268     recognized line ending sequence, the match position advancement for a
269     non-anchored pattern. The choice of newline convention does not affect the
270     interpretation of the \n or \r escape sequences.
271 nigel 91 </P>
272     <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
273     <P>
274 nigel 63 The PCRE functions can be used in multi-threading applications, with the
275 nigel 73 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
276     <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
277     callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
278 nigel 63 </P>
279     <P>
280     The compiled form of a regular expression is not altered during matching, so
281     the same compiled pattern can safely be used by several threads at once.
282     </P>
283 nigel 91 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
284 nigel 63 <P>
285 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
286     time, possibly by a different program, and even on a host other than the one on
287     which it was compiled. Details are given in the
288     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
289     documentation.
290     </P>
291 nigel 91 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
292 nigel 75 <P>
293 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
294     </P>
295     <P>
296     The function <b>pcre_config()</b> makes it possible for a PCRE client to
297     discover which optional features have been compiled into the PCRE library. The
298     <a href="pcrebuild.html"><b>pcrebuild</b></a>
299     documentation has more details about these optional features.
300     </P>
301     <P>
302     The first argument for <b>pcre_config()</b> is an integer, specifying which
303     information is required; the second argument is a pointer to a variable into
304     which the information is placed. The following information is available:
305     <pre>
307 nigel 75 </pre>
308 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
309     otherwise it is set to zero.
310     <pre>
312     </pre>
313     The output is an integer that is set to one if support for Unicode character
314     properties is available; otherwise it is set to zero.
315     <pre>
317 nigel 75 </pre>
318 nigel 91 The output is an integer whose value specifies the default character sequence
319 nigel 93 that is recognized as meaning "newline". The four values that are supported
320     are: 10 for LF, 13 for CR, 3338 for CRLF, and -1 for ANY. The default should
321     normally be the standard sequence for your operating system.
322 nigel 63 <pre>
324 nigel 75 </pre>
325 nigel 63 The output is an integer that contains the number of bytes used for internal
326     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
327     allow larger regular expressions to be compiled, at the expense of slower
328     matching. The default value of 2 is sufficient for all but the most massive
329     patterns, since it allows the compiled pattern to be up to 64K in size.
330     <pre>
332 nigel 75 </pre>
333 nigel 63 The output is an integer that contains the threshold above which the POSIX
334     interface uses <b>malloc()</b> for output vectors. Further details are given in
335 nigel 75 the
336     <a href="pcreposix.html"><b>pcreposix</b></a>
337     documentation.
338 nigel 63 <pre>
340 nigel 75 </pre>
341 nigel 63 The output is an integer that gives the default limit for the number of
342     internal matching function calls in a <b>pcre_exec()</b> execution. Further
343     details are given with <b>pcre_exec()</b> below.
344 nigel 73 <pre>
346     </pre>
347     The output is an integer that gives the default limit for the depth of
348     recursion when calling the internal matching function in a <b>pcre_exec()</b>
349     execution. Further details are given with <b>pcre_exec()</b> below.
350     <pre>
352 nigel 75 </pre>
353 nigel 77 The output is an integer that is set to one if internal recursion when running
354     <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
355     to remember their state. This is the usual way that PCRE is compiled. The
356     output is zero if PCRE was compiled to use blocks of data on the heap instead
357     of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
358     <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
359     avoiding the use of the stack.
360 nigel 73 </P>
361 nigel 91 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
362 nigel 63 <P>
363     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
364     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
365     <b>const unsigned char *<i>tableptr</i>);</b>
366 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
367     <b>int *<i>errorcodeptr</i>,</b>
368     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
369     <b>const unsigned char *<i>tableptr</i>);</b>
370 nigel 63 </P>
371     <P>
372 nigel 77 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
373     called to compile a pattern into an internal form. The only difference between
374     the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
375     <i>errorcodeptr</i>, via which a numerical error code can be returned.
376 nigel 63 </P>
377     <P>
378 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
379     <i>pattern</i> argument. A pointer to a single block of memory that is obtained
380     via <b>pcre_malloc</b> is returned. This contains the compiled code and related
381     data. The <b>pcre</b> type is defined for the returned block; this is a typedef
382     for a structure whose contents are not externally defined. It is up to the
383 nigel 91 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
384 nigel 77 </P>
385     <P>
386 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
387     depend on memory location, the complete <b>pcre</b> data block is not
388 nigel 75 fully relocatable, because it may contain a copy of the <i>tableptr</i>
389     argument, which is an address (see below).
390 nigel 63 </P>
391     <P>
392 nigel 93 The <i>options</i> argument contains various bit settings that affect the
393 nigel 75 compilation. It should be zero if no options are required. The available
394     options are described below. Some of them, in particular, those that are
395     compatible with Perl, can also be set and unset from within the pattern (see
396     the detailed description in the
397     <a href="pcrepattern.html"><b>pcrepattern</b></a>
398     documentation). For these options, the contents of the <i>options</i> argument
399     specifies their initial settings at the start of compilation and execution. The
400 nigel 91 PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
401     matching as well as at compile time.
402 nigel 63 </P>
403     <P>
404     If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
405     Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
406     NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
407 nigel 87 error message. This is a static string that is part of the library. You must
408     not try to free it. The offset from the start of the pattern to the character
409     where the error was discovered is placed in the variable pointed to by
410 nigel 63 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
411     </P>
412     <P>
413 nigel 77 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
414     <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
415     returned via this argument in the event of an error. This is in addition to the
416     textual error message. Error codes and messages are listed below.
417     </P>
418     <P>
419 nigel 63 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
420 nigel 75 character tables that are built when PCRE is compiled, using the default C
421     locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
422     call to <b>pcre_maketables()</b>. This value is stored with the compiled
423     pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
424     passed to it. For more discussion, see the section on locale support below.
425 nigel 63 </P>
426     <P>
427     This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
428     <pre>
429     pcre *re;
430     const char *error;
431     int erroffset;
432     re = pcre_compile(
433     "^A.*Z", /* the pattern */
434     0, /* default options */
435     &error, /* for error message */
436     &erroffset, /* for error offset */
437     NULL); /* use default character tables */
438 nigel 75 </pre>
439     The following names for option bits are defined in the <b>pcre.h</b> header
440     file:
441 nigel 63 <pre>
443 nigel 75 </pre>
444 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
445 nigel 75 constrained to match only at the first matching point in the string that is
446 nigel 63 being searched (the "subject string"). This effect can also be achieved by
447     appropriate constructs in the pattern itself, which is the only way to do it in
448     Perl.
449     <pre>
450 nigel 75 PCRE_AUTO_CALLOUT
451     </pre>
452     If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
453     all with number 255, before each pattern item. For discussion of the callout
454     facility, see the
455     <a href="pcrecallout.html"><b>pcrecallout</b></a>
456     documentation.
457     <pre>
458 nigel 63 PCRE_CASELESS
459 nigel 75 </pre>
460 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
461     letters. It is equivalent to Perl's /i option, and it can be changed within a
462 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
463     concept of case for characters whose values are less than 128, so caseless
464     matching is always possible. For characters with higher values, the concept of
465     case is supported if PCRE is compiled with Unicode property support, but not
466     otherwise. If you want to use caseless matching for characters 128 and above,
467     you must ensure that PCRE is compiled with Unicode property support as well as
468     with UTF-8 support.
469 nigel 63 <pre>
471 nigel 75 </pre>
472 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
473     end of the subject string. Without this option, a dollar also matches
474 nigel 91 immediately before a newline at the end of the string (but not before any other
475     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
476     There is no equivalent to this option in Perl, and no way to set it within a
477     pattern.
478 nigel 63 <pre>
480 nigel 75 </pre>
481 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
482 nigel 91 including those that indicate newline. Without it, a dot does not match when
483     the current position is at a newline. This option is equivalent to Perl's /s
484     option, and it can be changed within a pattern by a (?s) option setting. A
485 nigel 93 negative class such as [^a] always matches newline characters, independent of
486     the setting of this option.
487 nigel 63 <pre>
488 nigel 91 PCRE_DUPNAMES
489     </pre>
490     If this bit is set, names used to identify capturing subpatterns need not be
491     unique. This can be helpful for certain types of pattern when it is known that
492     only one instance of the named subpattern can ever be matched. There are more
493     details of named subpatterns below; see also the
494     <a href="pcrepattern.html"><b>pcrepattern</b></a>
495     documentation.
496     <pre>
497 nigel 63 PCRE_EXTENDED
498 nigel 75 </pre>
499 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
500     ignored except when escaped or inside a character class. Whitespace does not
501     include the VT character (code 11). In addition, characters between an
502 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
503     ignored. This is equivalent to Perl's /x option, and it can be changed within a
504     pattern by a (?x) option setting.
505 nigel 63 </P>
506     <P>
507     This option makes it possible to include comments inside complicated patterns.
508     Note, however, that this applies only to data characters. Whitespace characters
509     may never appear within special character sequences in a pattern, for example
510     within the sequence (?( which introduces a conditional subpattern.
511     <pre>
512     PCRE_EXTRA
513 nigel 75 </pre>
514 nigel 63 This option was invented in order to turn on additional functionality of PCRE
515     that is incompatible with Perl, but it is currently of very little use. When
516     set, any backslash in a pattern that is followed by a letter that has no
517     special meaning causes an error, thus reserving these combinations for future
518     expansion. By default, as in Perl, a backslash followed by a letter with no
519 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
520     give a warning for this.) There are at present no other features controlled by
521     this option. It can also be set by a (?X) option setting within a pattern.
522 nigel 63 <pre>
523 nigel 77 PCRE_FIRSTLINE
524     </pre>
525     If this option is set, an unanchored pattern is required to match before or at
526 nigel 91 the first newline in the subject string, though the matched text may continue
527     over the newline.
528 nigel 77 <pre>
529 nigel 63 PCRE_MULTILINE
530 nigel 75 </pre>
531     By default, PCRE treats the subject string as consisting of a single line of
532     characters (even if it actually contains newlines). The "start of line"
533 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
534     line" metacharacter ($) matches only at the end of the string, or before a
535     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
536     Perl.
537     </P>
538     <P>
539     When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
540 nigel 91 match immediately following or immediately before internal newlines in the
541     subject string, respectively, as well as at the very start and end. This is
542     equivalent to Perl's /m option, and it can be changed within a pattern by a
543     (?m) option setting. If there are no newlines in a subject string, or no
544 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
545     <pre>
546 nigel 91 PCRE_NEWLINE_CR
549 nigel 93 PCRE_NEWLINE_ANY
550 nigel 91 </pre>
551     These options override the default newline definition that was chosen when PCRE
552     was built. Setting the first or the second specifies that a newline is
553 nigel 93 indicated by a single character (CR or LF, respectively). Setting
554     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
555     CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that any Unicode newline
556     sequence should be recognized. The Unicode newline sequences are the three just
557     mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
558     U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
559     (paragraph separator, U+2029). The last two are recognized only in UTF-8 mode.
560 nigel 91 </P>
561     <P>
562 nigel 93 The newline setting in the options word uses three bits that are treated
563     as a number, giving eight possibilities. Currently only five are used (default
564     plus the four values above). This means that if you set more than one newline
565     option, the combination may or may not be sensible. For example,
566     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
567     other combinations yield unused numbers and cause an error.
568     </P>
569     <P>
570     The only time that a line break is specially recognized when compiling a
571     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
572     class is encountered. This indicates a comment that lasts until after the next
573     line break sequence. In other circumstances, line break sequences are treated
574     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
575     as whitespace characters and are therefore ignored.
576     </P>
577     <P>
578     The newline option that is set at compile time becomes the default that is used
579     for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
580 nigel 91 <pre>
582 nigel 75 </pre>
583 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
584     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
585     were followed by ?: but named parentheses can still be used for capturing (and
586     they acquire numbers in the usual way). There is no equivalent of this option
587     in Perl.
588     <pre>
590 nigel 75 </pre>
591 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
592     greedy by default, but become greedy if followed by "?". It is not compatible
593     with Perl. It can also be set by a (?U) option setting within the pattern.
594     <pre>
595     PCRE_UTF8
596 nigel 75 </pre>
597 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
598     of UTF-8 characters instead of single-byte character strings. However, it is
599 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
600 nigel 63 of this option provokes an error. Details of how this option changes the
601     behaviour of PCRE are given in the
602     <a href="pcre.html#utf8support">section on UTF-8 support</a>
603     in the main
604     <a href="pcre.html"><b>pcre</b></a>
605     page.
606 nigel 71 <pre>
608 nigel 75 </pre>
609 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
610     automatically checked. If an invalid UTF-8 sequence of bytes is found,
611     <b>pcre_compile()</b> returns an error. If you already know that your pattern is
612     valid, and you want to skip this check for performance reasons, you can set the
613     PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
614     UTF-8 string as a pattern is undefined. It may cause your program to crash.
615 nigel 77 Note that this option can also be passed to <b>pcre_exec()</b> and
616     <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
617     strings.
618 nigel 71 </P>
619 nigel 91 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
620 nigel 63 <P>
621 nigel 77 The following table lists the error codes than may be returned by
622     <b>pcre_compile2()</b>, along with the error messages that may be returned by
623 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
624     out of use. To avoid confusion, they have not been re-used.
625 nigel 77 <pre>
626     0 no error
627     1 \ at end of pattern
628     2 \c at end of pattern
629     3 unrecognized character follows \
630     4 numbers out of order in {} quantifier
631     5 number too big in {} quantifier
632     6 missing terminating ] for character class
633     7 invalid escape sequence in character class
634     8 range out of order in character class
635     9 nothing to repeat
636 nigel 93 10 [this code is not in use]
637 nigel 77 11 internal error: unexpected repeat
638     12 unrecognized character after (?
639     13 POSIX named classes are supported only within a class
640     14 missing )
641     15 reference to non-existent subpattern
642     16 erroffset passed as NULL
643     17 unknown option bit(s) set
644     18 missing ) after comment
645 nigel 93 19 [this code is not in use]
646 nigel 77 20 regular expression too large
647     21 failed to get memory
648     22 unmatched parentheses
649     23 internal error: code overflow
650     24 unrecognized character after (?&#60;
651     25 lookbehind assertion is not fixed length
652 nigel 91 26 malformed number or name after (?(
653 nigel 77 27 conditional group contains more than two branches
654     28 assertion expected after (?(
655     29 (?R or (?digits must be followed by )
656     30 unknown POSIX class name
657     31 POSIX collating elements are not supported
658     32 this version of PCRE is not compiled with PCRE_UTF8 support
659 nigel 93 33 [this code is not in use]
660 nigel 77 34 character value in \x{...} sequence is too large
661     35 invalid condition (?(0)
662     36 \C not allowed in lookbehind assertion
663     37 PCRE does not support \L, \l, \N, \U, or \u
664     38 number after (?C is &#62; 255
665     39 closing ) for (?C expected
666     40 recursive call could loop indefinitely
667     41 unrecognized character after (?P
668 nigel 93 42 syntax error in subpattern name (missing terminator)
669 nigel 91 43 two named subpatterns have the same name
670 nigel 77 44 invalid UTF-8 string
671     45 support for \P, \p, and \X has not been compiled
672     46 malformed \P or \p sequence
673     47 unknown property name after \P or \p
674 nigel 91 48 subpattern name is too long (maximum 32 characters)
675     49 too many named subpatterns (maximum 10,000)
676     50 repeated subpattern is too long
677     51 octal value is greater than \377 (not in UTF-8 mode)
678 nigel 93 52 internal error: overran compiling workspace
679     53 internal error: previously-checked referenced subpattern not found
680     54 DEFINE group contains more than one branch
681     55 repeating a DEFINE group is not allowed
682     56 inconsistent NEWLINE options"
683 nigel 77 </PRE>
684     </P>
685 nigel 91 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
686 nigel 77 <P>
687     <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
688 nigel 63 <b>const char **<i>errptr</i>);</b>
689     </P>
690     <P>
691 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
692     more time analyzing it in order to speed up the time taken for matching. The
693 nigel 63 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
694 nigel 75 argument. If studying the pattern produces additional information that will
695     help speed up matching, <b>pcre_study()</b> returns a pointer to a
696     <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
697     results of the study.
698 nigel 63 </P>
699     <P>
700 nigel 75 The returned value from <b>pcre_study()</b> can be passed directly to
701     <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
702 nigel 63 fields that can be set by the caller before the block is passed; these are
703 nigel 75 described
704     <a href="#extradata">below</a>
705     in the section on matching a pattern.
706 nigel 63 </P>
707     <P>
708 nigel 77 If studying the pattern does not produce any additional information
709 nigel 75 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
710     wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
711     own <b>pcre_extra</b> block.
712 nigel 63 </P>
713     <P>
714 nigel 75 The second argument of <b>pcre_study()</b> contains option bits. At present, no
715     options are defined, and this argument should always be zero.
716     </P>
717     <P>
718 nigel 63 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
719     studying succeeds (even if no data is returned), the variable it points to is
720 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
721     static string that is part of the library. You must not try to free it. You
722     should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
723     sure that it has run successfully.
724 nigel 63 </P>
725     <P>
726     This is a typical call to <b>pcre_study</b>():
727     <pre>
728     pcre_extra *pe;
729     pe = pcre_study(
730     re, /* result of pcre_compile() */
731     0, /* no options exist */
732     &error); /* set to NULL or points to a message */
733 nigel 75 </pre>
734 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
735     not have a single fixed starting character. A bitmap of possible starting
736 nigel 75 bytes is created.
737     <a name="localesupport"></a></P>
738 nigel 91 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
739 nigel 63 <P>
740 nigel 77 PCRE handles caseless matching, and determines whether characters are letters
741 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
742 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
743 nigel 75 less than 128. Higher-valued codes never match escapes such as \w or \d, but
744     can be tested with \p if PCRE is built with Unicode character property
745 nigel 87 support. The use of locales with Unicode is discouraged.
746 nigel 63 </P>
747     <P>
748 nigel 75 An internal set of tables is created in the default C locale when PCRE is
749     built. This is used when the final argument of <b>pcre_compile()</b> is NULL,
750     and is sufficient for many applications. An alternative set of tables can,
751     however, be supplied. These may be created in a different locale from the
752     default. As more and more applications change to using Unicode, the need for
753     this locale support is expected to die away.
754 nigel 63 </P>
755     <P>
756 nigel 75 External tables are built by calling the <b>pcre_maketables()</b> function,
757     which has no arguments, in the relevant locale. The result can then be passed
758     to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
759     example, to build and use tables that are appropriate for the French locale
760     (where accented characters with values greater than 128 are treated as letters),
761     the following code could be used:
762 nigel 63 <pre>
763 nigel 75 setlocale(LC_CTYPE, "fr_FR");
764 nigel 63 tables = pcre_maketables();
765     re = pcre_compile(..., tables);
766 nigel 75 </pre>
767     When <b>pcre_maketables()</b> runs, the tables are built in memory that is
768     obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
769     that the memory containing the tables remains available for as long as it is
770     needed.
771 nigel 63 </P>
772     <P>
773 nigel 75 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
774 nigel 63 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
775 nigel 75 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
776     pattern, compilation, studying and matching all happen in the same locale, but
777     different patterns can be compiled in different locales.
778 nigel 63 </P>
779     <P>
780 nigel 75 It is possible to pass a table pointer or NULL (indicating the use of the
781     internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
782     this facility could be used to match a pattern in a different locale from the
783     one in which it was compiled. Passing table pointers at run time is discussed
784     below in the section on matching a pattern.
785     </P>
786 nigel 91 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
787 nigel 75 <P>
788 nigel 63 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
789     <b>int <i>what</i>, void *<i>where</i>);</b>
790     </P>
791     <P>
792     The <b>pcre_fullinfo()</b> function returns information about a compiled
793     pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
794     nevertheless retained for backwards compability (and is documented below).
795     </P>
796     <P>
797     The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
798     pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
799     the pattern was not studied. The third argument specifies which piece of
800     information is required, and the fourth argument is a pointer to a variable
801     to receive the data. The yield of the function is zero for success, or one of
802     the following negative numbers:
803     <pre>
804     PCRE_ERROR_NULL the argument <i>code</i> was NULL
805     the argument <i>where</i> was NULL
806     PCRE_ERROR_BADMAGIC the "magic number" was not found
807     PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
808 nigel 75 </pre>
809     The "magic number" is placed at the start of each compiled pattern as an simple
810     check against passing an arbitrary memory pointer. Here is a typical call of
811     <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
812 nigel 63 <pre>
813     int rc;
814 nigel 91 size_t length;
815 nigel 63 rc = pcre_fullinfo(
816     re, /* result of pcre_compile() */
817     pe, /* result of pcre_study(), or NULL */
818     PCRE_INFO_SIZE, /* what is required */
819     &length); /* where to put the data */
820 nigel 75 </pre>
821 nigel 63 The possible values for the third argument are defined in <b>pcre.h</b>, and are
822     as follows:
823     <pre>
825 nigel 75 </pre>
826 nigel 63 Return the number of the highest back reference in the pattern. The fourth
827     argument should point to an <b>int</b> variable. Zero is returned if there are
828     no back references.
829     <pre>
831 nigel 75 </pre>
832 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
833 nigel 75 should point to an <b>int</b> variable.
834 nigel 63 <pre>
836 nigel 75 </pre>
837     Return a pointer to the internal default character tables within PCRE. The
838     fourth argument should point to an <b>unsigned char *</b> variable. This
839     information call is provided for internal use by the <b>pcre_study()</b>
840     function. External callers can cause PCRE to use its internal tables by passing
841     a NULL table pointer.
842     <pre>
844 nigel 75 </pre>
845 nigel 63 Return information about the first byte of any matched string, for a
846 nigel 91 non-anchored pattern. The fourth argument should point to an <b>int</b>
847     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
848     still recognized for backwards compatibility.)
849 nigel 63 </P>
850     <P>
851 nigel 75 If there is a fixed first byte, for example, from a pattern such as
852 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
853 nigel 75 <br>
854     <br>
855 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
856     starts with "^", or
857 nigel 75 <br>
858     <br>
859 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
860     (if it were set, the pattern would be anchored),
861 nigel 75 <br>
862     <br>
863 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
864     subject string or after any newline within the string. Otherwise -2 is
865     returned. For anchored patterns, -2 is returned.
866     <pre>
868 nigel 75 </pre>
869 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
870     table indicating a fixed set of bytes for the first byte in any matching
871     string, a pointer to the table is returned. Otherwise NULL is returned. The
872     fourth argument should point to an <b>unsigned char *</b> variable.
873     <pre>
875 nigel 75 </pre>
876 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
877     string, other than at its start, if such a byte has been recorded. The fourth
878     argument should point to an <b>int</b> variable. If there is no such byte, -1 is
879     returned. For anchored patterns, a last literal byte is recorded only if it
880     follows something of variable length. For example, for the pattern
881     /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
882     is -1.
883 nigel 63 <pre>
887 nigel 75 </pre>
888 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
889     names are just an additional way of identifying the parentheses, which still
890 nigel 91 acquire numbers. Several convenience functions such as
891     <b>pcre_get_named_substring()</b> are provided for extracting captured
892     substrings by name. It is also possible to extract the data directly, by first
893     converting the name to a number in order to access the correct pointers in the
894     output vector (described with <b>pcre_exec()</b> below). To do the conversion,
895     you need to use the name-to-number map, which is described by these three
896     values.
897 nigel 63 </P>
898     <P>
899     The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
900     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
901     entry; both of these return an <b>int</b> value. The entry size depends on the
902     length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
903     entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
904     are the number of the capturing parenthesis, most significant byte first. The
905     rest of the entry is the corresponding name, zero terminated. The names are in
906 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
907     their parentheses numbers. For example, consider the following pattern (assume
908 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
909     <pre>
910 nigel 93 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
911 nigel 75 </pre>
912 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
913     in the table is eight bytes long. The table is as follows, with non-printing
914 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
915 nigel 63 <pre>
916     00 01 d a t e 00 ??
917     00 05 d a y 00 ?? ??
918     00 04 m o n t h 00
919     00 02 y e a r 00 ??
920 nigel 75 </pre>
921     When writing code to extract data from named subpatterns using the
922 nigel 91 name-to-number map, remember that the length of the entries is likely to be
923 nigel 75 different for each compiled pattern.
924 nigel 63 <pre>
926 nigel 75 </pre>
927 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
928     argument should point to an <b>unsigned long int</b> variable. These option bits
929     are those specified in the call to <b>pcre_compile()</b>, modified by any
930     top-level option settings within the pattern itself.
931     </P>
932     <P>
933     A pattern is automatically anchored by PCRE if all of its top-level
934     alternatives begin with one of the following:
935     <pre>
936     ^ unless PCRE_MULTILINE is set
937     \A always
938     \G always
939 nigel 75 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
940     </pre>
941 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
942     <b>pcre_fullinfo()</b>.
943     <pre>
945 nigel 75 </pre>
946 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
947     the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
948     place the compiled data. The fourth argument should point to a <b>size_t</b>
949     variable.
950     <pre>
952 nigel 75 </pre>
953     Return the size of the data block pointed to by the <i>study_data</i> field in
954 nigel 63 a <b>pcre_extra</b> block. That is, it is the value that was passed to
955     <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
956     created by <b>pcre_study()</b>. The fourth argument should point to a
957     <b>size_t</b> variable.
958     </P>
959 nigel 91 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
960 nigel 63 <P>
961     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
962     <b>*<i>firstcharptr</i>);</b>
963     </P>
964     <P>
965     The <b>pcre_info()</b> function is now obsolete because its interface is too
966     restrictive to return all the available data about a compiled pattern. New
967     programs should use <b>pcre_fullinfo()</b> instead. The yield of
968     <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
969     following negative numbers:
970     <pre>
971     PCRE_ERROR_NULL the argument <i>code</i> was NULL
972     PCRE_ERROR_BADMAGIC the "magic number" was not found
973 nigel 75 </pre>
974 nigel 63 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
975     pattern was compiled is placed in the integer it points to (see
976     PCRE_INFO_OPTIONS above).
977     </P>
978     <P>
979     If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
980     it is used to pass back information about the first character of any matched
981     string (see PCRE_INFO_FIRSTBYTE above).
982     </P>
983 nigel 91 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
984 nigel 63 <P>
985 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
986     </P>
987     <P>
988     The <b>pcre_refcount()</b> function is used to maintain a reference count in the
989     data block that contains a compiled pattern. It is provided for the benefit of
990     applications that operate in an object-oriented manner, where different parts
991     of the application may be using the same compiled pattern, but you want to free
992     the block when they are all done.
993     </P>
994     <P>
995     When a pattern is compiled, the reference count field is initialized to zero.
996     It is changed only by calling this function, whose action is to add the
997     <i>adjust</i> value (which may be positive or negative) to it. The yield of the
998     function is the new value. However, the value of the count is constrained to
999     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1000     it is forced to the appropriate limit value.
1001     </P>
1002     <P>
1003     Except when it is zero, the reference count is not correctly preserved if a
1004     pattern is compiled on one host and then transferred to a host whose byte-order
1005     is different. (This seems a highly unlikely scenario.)
1006     </P>
1007 nigel 91 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1008 nigel 77 <P>
1009 nigel 63 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1010     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1011     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1012     </P>
1013     <P>
1014     The function <b>pcre_exec()</b> is called to match a subject string against a
1015 nigel 75 compiled pattern, which is passed in the <i>code</i> argument. If the
1016 nigel 63 pattern has been studied, the result of the study should be passed in the
1017 nigel 77 <i>extra</i> argument. This function is the main matching facility of the
1018     library, and it operates in a Perl-like manner. For specialist use there is
1019     also an alternative matching function, which is described
1020     <a href="#dfamatch">below</a>
1021     in the section about the <b>pcre_dfa_exec()</b> function.
1022 nigel 63 </P>
1023     <P>
1024 nigel 75 In most applications, the pattern will have been compiled (and optionally
1025     studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1026     possible to save compiled patterns and study data, and then use them later
1027     in different processes, possibly even on different hosts. For a discussion
1028     about this, see the
1029     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1030     documentation.
1031 nigel 63 </P>
1032     <P>
1033 nigel 75 Here is an example of a simple call to <b>pcre_exec()</b>:
1034 nigel 63 <pre>
1035     int rc;
1036     int ovector[30];
1037     rc = pcre_exec(
1038     re, /* result of pcre_compile() */
1039     NULL, /* we didn't study the pattern */
1040     "some string", /* the subject string */
1041     11, /* the length of the subject string */
1042     0, /* start at offset 0 in the subject */
1043     0, /* default options */
1044 nigel 75 ovector, /* vector of integers for substring information */
1045 nigel 77 30); /* number of elements (NOT size in bytes) */
1046 nigel 75 <a name="extradata"></a></PRE>
1047 nigel 63 </P>
1048 nigel 75 <br><b>
1049     Extra data for <b>pcre_exec()</b>
1050     </b><br>
1051 nigel 63 <P>
1052     If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1053     data block. The <b>pcre_study()</b> function returns such a block (when it
1054     doesn't return NULL), but you can also create one for yourself, and pass
1055 nigel 87 additional information in it. The <b>pcre_extra</b> block contains the following
1056     fields (not necessarily in this order):
1057 nigel 63 <pre>
1058     unsigned long int <i>flags</i>;
1059     void *<i>study_data</i>;
1060     unsigned long int <i>match_limit</i>;
1061 nigel 87 unsigned long int <i>match_limit_recursion</i>;
1062 nigel 63 void *<i>callout_data</i>;
1063 nigel 75 const unsigned char *<i>tables</i>;
1064     </pre>
1065 nigel 63 The <i>flags</i> field is a bitmap that specifies which of the other fields
1066     are set. The flag bits are:
1067     <pre>
1072 nigel 75 PCRE_EXTRA_TABLES
1073     </pre>
1074 nigel 63 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1075     <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1076 nigel 75 the appropriate flag bit. You should not set this yourself, but you may add to
1077     the block by setting the other fields and their corresponding flag bits.
1078 nigel 63 </P>
1079     <P>
1080     The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1081     vast amount of resources when running patterns that are not going to match,
1082     but which have a very large number of possibilities in their search trees. The
1083 nigel 75 classic example is the use of nested unlimited repeats.
1084 nigel 63 </P>
1085     <P>
1086 nigel 75 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1087 nigel 87 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1088     number of times this function is called during a match, which has the effect of
1089     limiting the amount of backtracking that can take place. For patterns that are
1090     not anchored, the count restarts from zero for each position in the subject
1091     string.
1092 nigel 75 </P>
1093     <P>
1094 nigel 87 The default value for the limit can be set when PCRE is built; the default
1095 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1096 nigel 87 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1097     block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1098     the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1100 nigel 63 </P>
1101     <P>
1102 nigel 87 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1103     instead of limiting the total number of times that <b>match()</b> is called, it
1104     limits the depth of recursion. The recursion depth is a smaller number than the
1105     total number of calls, because not all calls to <b>match()</b> are recursive.
1106     This limit is of use only if it is set smaller than <i>match_limit</i>.
1107     </P>
1108     <P>
1109     Limiting the recursion depth limits the amount of stack that can be used, or,
1110     when PCRE has been compiled to use memory on the heap instead of the stack, the
1111     amount of heap memory that can be used.
1112     </P>
1113     <P>
1114     The default value for <i>match_limit_recursion</i> can be set when PCRE is
1115     built; the default default is the same value as the default for
1116     <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1117     with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1118     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1119     is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1120     </P>
1121     <P>
1122 nigel 63 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1123 nigel 75 which is described in the
1124     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1125     documentation.
1126 nigel 63 </P>
1127     <P>
1128 nigel 75 The <i>tables</i> field is used to pass a character tables pointer to
1129     <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1130     pattern. A non-NULL value is stored with the compiled pattern only if custom
1131     tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1132     If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1133     internal tables to be used. This facility is helpful when re-using patterns
1134     that have been saved after compiling with an external set of tables, because
1135     the external tables might be at a different address when <b>pcre_exec()</b> is
1136     called. See the
1137     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1138     documentation for a discussion of saving compiled patterns for later use.
1139 nigel 63 </P>
1140 nigel 75 <br><b>
1141     Option bits for <b>pcre_exec()</b>
1142     </b><br>
1143 nigel 63 <P>
1144 nigel 75 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1145 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1147 nigel 63 <pre>
1148 nigel 75 PCRE_ANCHORED
1149     </pre>
1150     The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1151     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1152     to be anchored by virtue of its contents, it cannot be made unachored at
1153     matching time.
1154     <pre>
1155 nigel 91 PCRE_NEWLINE_CR
1158 nigel 93 PCRE_NEWLINE_ANY
1159 nigel 91 </pre>
1160     These options override the newline definition that was chosen or defaulted when
1161 nigel 93 the pattern was compiled. For details, see the description of
1162     <b>pcre_compile()</b> above. During matching, the newline choice affects the
1163     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1164     the way the match position is advanced after a match failure for an unanchored
1165     pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1166     fails when the current position is at a CRLF sequence, the match position is
1167     advanced by two characters instead of one, in other words, to after the CRLF.
1168 nigel 91 <pre>
1169 nigel 63 PCRE_NOTBOL
1170 nigel 75 </pre>
1171     This option specifies that first character of the subject string is not the
1172     beginning of a line, so the circumflex metacharacter should not match before
1173     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1174     never to match. This option affects only the behaviour of the circumflex
1175     metacharacter. It does not affect \A.
1176 nigel 63 <pre>
1177     PCRE_NOTEOL
1178 nigel 75 </pre>
1179     This option specifies that the end of the subject string is not the end of a
1180     line, so the dollar metacharacter should not match it nor (except in multiline
1181     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1182     compile time) causes dollar never to match. This option affects only the
1183     behaviour of the dollar metacharacter. It does not affect \Z or \z.
1184 nigel 63 <pre>
1186 nigel 75 </pre>
1187 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1188     there are alternatives in the pattern, they are tried. If all the alternatives
1189     match the empty string, the entire match fails. For example, if the pattern
1190     <pre>
1191     a?b?
1192 nigel 75 </pre>
1193 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1194     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1195     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1196     </P>
1197     <P>
1198     Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1199     of a pattern match of the empty string within its <b>split()</b> function, and
1200     when using the /g modifier. It is possible to emulate Perl's behaviour after
1201     matching a null string by first trying the match again at the same offset with
1202 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1203     starting offset (see below) and trying an ordinary match again. There is some
1204     code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1205     <pre>
1207     </pre>
1208     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1209     string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1210     The value of <i>startoffset</i> is also checked to ensure that it points to the
1211     start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1212     <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1213     contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1214 nigel 63 </P>
1215     <P>
1216 nigel 75 If you already know that your subject is valid, and you want to skip these
1217     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1218     calling <b>pcre_exec()</b>. You might want to do this for the second and
1219     subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1220     all the matches in a single subject string. However, you should be sure that
1221     the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1222     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1223     subject, or a value of <i>startoffset</i> that does not point to the start of a
1224     UTF-8 character, is undefined. Your program may crash.
1225     <pre>
1227     </pre>
1228     This option turns on the partial matching feature. If the subject string fails
1229     to match the pattern, but at some point during the matching process the end of
1230     the subject was reached (that is, the subject partially matches the pattern and
1231     the failure to match occurred only because there were not enough subject
1232     characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1233     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1234     may appear in the pattern. These are discussed in the
1235     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1236     documentation.
1237     </P>
1238     <br><b>
1239     The string to be matched by <b>pcre_exec()</b>
1240     </b><br>
1241     <P>
1242 nigel 63 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1243 nigel 73 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1244 nigel 75 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1245     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1246     bytes. When the starting offset is zero, the search for a match starts at the
1247     beginning of the subject, and this is by far the most common case.
1248 nigel 63 </P>
1249     <P>
1250     A non-zero starting offset is useful when searching for another match in the
1251     same subject by calling <b>pcre_exec()</b> again after a previous success.
1252     Setting <i>startoffset</i> differs from just passing over a shortened string and
1253     setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1254     lookbehind. For example, consider the pattern
1255     <pre>
1256     \Biss\B
1257 nigel 75 </pre>
1258 nigel 63 which finds occurrences of "iss" in the middle of words. (\B matches only if
1259     the current position in the subject is not a word boundary.) When applied to
1260     the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1261     occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1262     subject, namely "issipi", it does not match, because \B is always false at the
1263     start of the subject, which is deemed to be a word boundary. However, if
1264     <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1265     set to 4, it finds the second occurrence of "iss" because it is able to look
1266     behind the starting point to discover that it is preceded by a letter.
1267     </P>
1268     <P>
1269     If a non-zero starting offset is passed when the pattern is anchored, one
1270 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1271 nigel 63 pattern does not require the match to be at the start of the subject.
1272     </P>
1273 nigel 75 <br><b>
1274     How <b>pcre_exec()</b> returns captured substrings
1275     </b><br>
1276 nigel 63 <P>
1277     In general, a pattern matches a certain portion of the subject, and in
1278     addition, further substrings from the subject may be picked out by parts of the
1279     pattern. Following the usage in Jeffrey Friedl's book, this is called
1280     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1281     a fragment of a pattern that picks out a substring. PCRE supports several other
1282     kinds of parenthesized subpattern that do not cause substrings to be captured.
1283     </P>
1284     <P>
1285     Captured substrings are returned to the caller via a vector of integer offsets
1286     whose address is passed in <i>ovector</i>. The number of elements in the vector
1287 nigel 75 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1288     this argument is NOT the size of <i>ovector</i> in bytes.
1289 nigel 63 </P>
1290     <P>
1291 nigel 75 The first two-thirds of the vector is used to pass back captured substrings,
1292     each substring using a pair of integers. The remaining third of the vector is
1293     used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1294     and is not available for passing back information. The length passed in
1295     <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1296     rounded down.
1297     </P>
1298     <P>
1299     When a match is successful, information about captured substrings is returned
1300     in pairs of integers, starting at the beginning of <i>ovector</i>, and
1301 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1302     pair is set to the offset of the first character in a substring, and the second
1303     is set to the offset of the first character after the end of a substring. The
1304     first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1305     subject string matched by the entire pattern. The next pair is used for the
1306     first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1307 nigel 91 is one more than the highest numbered pair that has been set. For example, if
1308     two substrings have been captured, the returned value is 3. If there are no
1309     capturing subpatterns, the return value from a successful match is 1,
1310     indicating that just the first pair of offsets has been set.
1311 nigel 63 </P>
1312     <P>
1313     If a capturing subpattern is matched repeatedly, it is the last portion of the
1314 nigel 75 string that it matched that is returned.
1315 nigel 63 </P>
1316     <P>
1317 nigel 75 If the vector is too small to hold all the captured substring offsets, it is
1318     used as far as possible (up to two-thirds of its length), and the function
1319     returns a value of zero. In particular, if the substring offsets are not of
1320     interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1321 nigel 63 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1322 nigel 75 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1323     has to get additional memory for use during matching. Thus it is usually
1324     advisable to supply an <i>ovector</i>.
1325 nigel 63 </P>
1326     <P>
1327 nigel 91 The <b>pcre_info()</b> function can be used to find out how many capturing
1328 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1329     <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1330     the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1331 nigel 91 </P>
1332     <P>
1333     It is possible for capturing subpattern number <i>n+1</i> to match some part of
1334     the subject when subpattern <i>n</i> has not been used at all. For example, if
1335     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1336     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1337     happens, both values in the offset pairs corresponding to unused subpatterns
1338     are set to -1.
1339     </P>
1340     <P>
1341     Offset values that correspond to unused subpatterns at the end of the
1342     expression are also set to -1. For example, if the string "abc" is matched
1343     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1344     return from the function is 2, because the highest used capturing subpattern
1345     number is 1. However, you can refer to the offsets for the second and third
1346     capturing subpatterns if you wish (assuming the vector is large enough, of
1347     course).
1348     </P>
1349     <P>
1350     Some convenience functions are provided for extracting the captured substrings
1351     as separate strings. These are described below.
1352 nigel 77 <a name="errorlist"></a></P>
1353 nigel 75 <br><b>
1354 nigel 91 Error return values from <b>pcre_exec()</b>
1355 nigel 75 </b><br>
1356 nigel 63 <P>
1357     If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1358     defined in the header file:
1359     <pre>
1360     PCRE_ERROR_NOMATCH (-1)
1361 nigel 75 </pre>
1362 nigel 63 The subject string did not match the pattern.
1363     <pre>
1364     PCRE_ERROR_NULL (-2)
1365 nigel 75 </pre>
1366 nigel 63 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1367     NULL and <i>ovecsize</i> was not zero.
1368     <pre>
1370 nigel 75 </pre>
1371 nigel 63 An unrecognized bit was set in the <i>options</i> argument.
1372     <pre>
1374 nigel 75 </pre>
1375 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1376 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1377     compiled in an environment of one endianness is run in an environment with the
1378     other endianness. This is the error that PCRE gives when the magic number is
1379     not present.
1380 nigel 63 <pre>
1381 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1382 nigel 75 </pre>
1383 nigel 63 While running the pattern match, an unknown item was encountered in the
1384     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1385     of the compiled pattern.
1386     <pre>
1388 nigel 75 </pre>
1389 nigel 63 If a pattern contains back references, but the <i>ovector</i> that is passed to
1390     <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1391     gets a block of memory at the start of matching to use for this purpose. If the
1392 nigel 75 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1393     automatically freed at the end of matching.
1394 nigel 63 <pre>
1396 nigel 75 </pre>
1397 nigel 63 This error is used by the <b>pcre_copy_substring()</b>,
1398     <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1399     below). It is never returned by <b>pcre_exec()</b>.
1400     <pre>
1402 nigel 75 </pre>
1403 nigel 87 The backtracking limit, as specified by the <i>match_limit</i> field in a
1404     <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1405     above.
1406     <pre>
1407 nigel 63 PCRE_ERROR_CALLOUT (-9)
1408 nigel 75 </pre>
1409 nigel 63 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1410     use by callout functions that want to yield a distinctive error code. See the
1411 nigel 75 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1412     documentation for details.
1413 nigel 71 <pre>
1414 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1415 nigel 75 </pre>
1416 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1417 nigel 73 <pre>
1419 nigel 75 </pre>
1420 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1421     of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1422 nigel 75 <pre>
1423 nigel 77 PCRE_ERROR_PARTIAL (-12)
1424 nigel 75 </pre>
1425     The subject string did not match, but it did match partially. See the
1426     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1427     documentation for details of partial matching.
1428     <pre>
1429 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1430 nigel 75 </pre>
1431     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1432     are not supported for partial matching. See the
1433     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1434     documentation for details of partial matching.
1435     <pre>
1436 nigel 77 PCRE_ERROR_INTERNAL (-14)
1437 nigel 75 </pre>
1438     An unexpected internal error has occurred. This error could be caused by a bug
1439     in PCRE or by overwriting of the compiled pattern.
1440     <pre>
1441 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1442 nigel 75 </pre>
1443     This error is given if the value of the <i>ovecsize</i> argument is negative.
1444 nigel 93 <pre>
1446     </pre>
1447     The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1448     field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1449     description above.
1450     <pre>
1452     </pre>
1453     When a group that can match an empty substring is repeated with an unbounded
1454     upper limit, the subject position at the start of the group must be remembered,
1455     so that a test for an empty string can be made when the end of the group is
1456     reached. Some workspace is required for this; if it runs out, this error is
1457     given.
1458     <pre>
1460     </pre>
1461     An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1462 nigel 73 </P>
1463 nigel 93 <P>
1464     Error numbers -16 to -20 are not used by <b>pcre_exec()</b>.
1465     </P>
1466 nigel 91 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1467 nigel 63 <P>
1468     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1469     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1470     <b>int <i>buffersize</i>);</b>
1471     </P>
1472     <P>
1473     <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1474     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1475     <b>const char **<i>stringptr</i>);</b>
1476     </P>
1477     <P>
1478     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1479     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1480     </P>
1481     <P>
1482     Captured substrings can be accessed directly by using the offsets returned by
1483     <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1484     <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1485     <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1486     as new, separate, zero-terminated strings. These functions identify substrings
1487     by number. The next section describes functions for extracting named
1488 nigel 91 substrings.
1489 nigel 63 </P>
1490     <P>
1491 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1492     further zero added on the end, but the result is not, of course, a C string.
1493     However, you can process such a string by referring to the length that is
1494     returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1495     Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1496     for handling strings containing binary zeros, because the end of the final
1497     string is not independently indicated.
1498     </P>
1499     <P>
1500 nigel 63 The first three arguments are the same for all three of these functions:
1501 nigel 75 <i>subject</i> is the subject string that has just been successfully matched,
1502 nigel 63 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1503     <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1504     captured by the match, including the substring that matched the entire regular
1505 nigel 75 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1506     than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1507     space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1508     number of elements in the vector divided by three.
1509 nigel 63 </P>
1510     <P>
1511     The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1512     extract a single substring, whose number is given as <i>stringnumber</i>. A
1513 nigel 75 value of zero extracts the substring that matched the entire pattern, whereas
1514 nigel 63 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1515     the string is placed in <i>buffer</i>, whose length is given by
1516     <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1517     obtained via <b>pcre_malloc</b>, and its address is returned via
1518     <i>stringptr</i>. The yield of the function is the length of the string, not
1519 nigel 93 including the terminating zero, or one of these error codes:
1520 nigel 63 <pre>
1522 nigel 75 </pre>
1523 nigel 63 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1524     memory failed for <b>pcre_get_substring()</b>.
1525     <pre>
1527 nigel 75 </pre>
1528 nigel 63 There is no substring whose number is <i>stringnumber</i>.
1529     </P>
1530     <P>
1531     The <b>pcre_get_substring_list()</b> function extracts all available substrings
1532     and builds a list of pointers to them. All this is done in a single block of
1533 nigel 75 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1534 nigel 63 is returned via <i>listptr</i>, which is also the start of the list of string
1535     pointers. The end of the list is marked by a NULL pointer. The yield of the
1536 nigel 93 function is zero if all went well, or the error code
1537 nigel 63 <pre>
1539 nigel 75 </pre>
1540 nigel 63 if the attempt to get the memory block failed.
1541     </P>
1542     <P>
1543     When any of these functions encounter a substring that is unset, which can
1544     happen when capturing subpattern number <i>n+1</i> matches some part of the
1545     subject, but subpattern <i>n</i> has not been used at all, they return an empty
1546     string. This can be distinguished from a genuine zero-length substring by
1547     inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1548     substrings.
1549     </P>
1550     <P>
1551     The two convenience functions <b>pcre_free_substring()</b> and
1552     <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1553     a previous call of <b>pcre_get_substring()</b> or
1554     <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1555     the function pointed to by <b>pcre_free</b>, which of course could be called
1556     directly from a C program. However, PCRE is used in some situations where it is
1557 nigel 91 linked via a special interface to another programming language that cannot use
1558 nigel 63 <b>pcre_free</b> directly; it is for these cases that the functions are
1559     provided.
1560     </P>
1561 nigel 91 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1562 nigel 63 <P>
1563 nigel 75 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1564     <b>const char *<i>name</i>);</b>
1565     </P>
1566     <P>
1567 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1568     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1569     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1570     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1571     </P>
1572     <P>
1573     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1574     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1575     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1576     <b>const char **<i>stringptr</i>);</b>
1577     </P>
1578     <P>
1579 nigel 75 To extract a substring by name, you first have to find associated number.
1580     For example, for this pattern
1581 nigel 63 <pre>
1582 nigel 93 (a+)b(?&#60;xxx&#62;\d+)...
1583 nigel 75 </pre>
1584 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1585     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1586     calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1587     pattern, and the second is the name. The yield of the function is the
1588 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1589     that name.
1590 nigel 63 </P>
1591     <P>
1592 nigel 75 Given the number, you can extract the substring directly, or use one of the
1593     functions described in the previous section. For convenience, there are also
1594     two functions that do the whole job.
1595 nigel 63 </P>
1596     <P>
1597 nigel 91 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1598     <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1599 nigel 75 functions that extract by number. As these are described in the previous
1600     section, they are not re-described here. There are just two differences:
1601 nigel 63 </P>
1602     <P>
1603     First, instead of a substring number, a substring name is given. Second, there
1604     is an extra argument, given at the start, which is a pointer to the compiled
1605     pattern. This is needed in order to gain access to the name-to-number
1606     translation table.
1607     </P>
1608     <P>
1609     These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1610     then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as
1611     appropriate.
1612     </P>
1613 nigel 91 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1614 nigel 63 <P>
1615 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1616     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1617     </P>
1618     <P>
1619     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1620     are not required to be unique. Normally, patterns with duplicate names are such
1621     that in any one match, only one of the named subpatterns participates. An
1622     example is shown in the
1623     <a href="pcrepattern.html"><b>pcrepattern</b></a>
1624     documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1625     and <b>pcre_get_named_substring()</b> return the first substring corresponding
1626     to the given name that is set. If none are set, an empty string is returned.
1627     The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1628     associated with the name, but it is not defined which it is.
1629     <br>
1630     <br>
1631     If you want to get full details of all captured substrings for a given name,
1632     you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1633     argument is the compiled pattern, and the second is the name. The third and
1634     fourth are pointers to variables which are updated by the function. After it
1635     has run, they point to the first and last entries in the name-to-number table
1636     for the given name. The function itself returns the length of each entry, or
1637 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1638     described above in the section entitled <i>Information about a pattern</i>.
1639     Given all the relevant entries for the name, you can extract each of their
1640     numbers, and hence the captured data, if any.
1641 nigel 91 </P>
1642     <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1643     <P>
1644 nigel 77 The traditional matching function uses a similar algorithm to Perl, which stops
1645     when it finds the first match, starting at a given point in the subject. If you
1646     want to find all possible matches, or the longest possible match, consider
1647     using the alternative matching function (see below) instead. If you cannot use
1648     the alternative function, but still need to find all possible matches, you
1649     can kludge it up by making use of the callout facility, which is described in
1650     the
1651     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1652     documentation.
1653     </P>
1654     <P>
1655     What you have to do is to insert a callout right at the end of the pattern.
1656     When your callout function is called, extract and save the current matched
1657     substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1658     other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1659     will yield PCRE_ERROR_NOMATCH.
1660     <a name="dfamatch"></a></P>
1661 nigel 91 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1662 nigel 77 <P>
1663     <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1664     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1665     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1666     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1667     </P>
1668     <P>
1669     The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1670 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1671     just once, and does not backtrack. This has different characteristics to the
1672     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1673     patterns are not supported. Nevertheless, there are times when this kind of
1674     matching can be useful. For a discussion of the two matching algorithms, see
1675     the
1676 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
1677     documentation.
1678     </P>
1679     <P>
1680     The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1681     <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1682     different way, and this is described below. The other common arguments are used
1683     in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1684     here.
1685     </P>
1686     <P>
1687     The two additional arguments provide workspace for the function. The workspace
1688     vector should contain at least 20 elements. It is used for keeping track of
1689     multiple paths through the pattern tree. More workspace will be needed for
1690 nigel 91 patterns and subjects where there are a lot of potential matches.
1691 nigel 77 </P>
1692     <P>
1693 nigel 87 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1694 nigel 77 <pre>
1695     int rc;
1696     int ovector[10];
1697     int wspace[20];
1698 nigel 87 rc = pcre_dfa_exec(
1699 nigel 77 re, /* result of pcre_compile() */
1700     NULL, /* we didn't study the pattern */
1701     "some string", /* the subject string */
1702     11, /* the length of the subject string */
1703     0, /* start at offset 0 in the subject */
1704     0, /* default options */
1705     ovector, /* vector of integers for substring information */
1706     10, /* number of elements (NOT size in bytes) */
1707     wspace, /* working space vector */
1708     20); /* number of elements (NOT size in bytes) */
1709     </PRE>
1710     </P>
1711     <br><b>
1712     Option bits for <b>pcre_dfa_exec()</b>
1713     </b><br>
1714     <P>
1715     The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1716 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1718 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1719     the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1720     <pre>
1722     </pre>
1723     This has the same general effect as it does for <b>pcre_exec()</b>, but the
1724     details are slightly different. When PCRE_PARTIAL is set for
1725     <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1726     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1727     complete matches, but there is still at least one matching possibility. The
1728     portion of the string that provided the partial match is set as the first
1729     matching string.
1730     <pre>
1732     </pre>
1733     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1734 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1735     works, this is necessarily the shortest possible match at the first possible
1736     matching point in the subject string.
1737 nigel 77 <pre>
1739     </pre>
1740     When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1741     a partial match, it is possible to call it again, with additional subject
1742     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1743     option requests this action; when it is set, the <i>workspace</i> and
1744     <i>wscount</i> options must reference the same vector as before because data
1745     about the match so far is left in them after a partial match. There is more
1746     discussion of this facility in the
1747     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1748     documentation.
1749     </P>
1750     <br><b>
1751     Successful returns from <b>pcre_dfa_exec()</b>
1752     </b><br>
1753     <P>
1754     When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1755     substring in the subject. Note, however, that all the matches from one run of
1756     the function start at the same point in the subject. The shorter matches are
1757     all initial substrings of the longer matches. For example, if the pattern
1758     <pre>
1759     &#60;.*&#62;
1760     </pre>
1761     is matched against the string
1762     <pre>
1763     This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1764     </pre>
1765     the three matched strings are
1766     <pre>
1767     &#60;something&#62;
1768     &#60;something&#62; &#60;something else&#62;
1769     &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1770     </pre>
1771     On success, the yield of the function is a number greater than zero, which is
1772     the number of matched substrings. The substrings themselves are returned in
1773     <i>ovector</i>. Each string uses two elements; the first is the offset to the
1774 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1775     the same start offset. (Space could have been saved by giving this only once,
1776     but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1777     returns data, even though the meaning of the strings is different.)
1778 nigel 77 </P>
1779     <P>
1780     The strings are returned in reverse order of length; that is, the longest
1781     matching string is given first. If there were too many matches to fit into
1782     <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1783     the longest matches.
1784     </P>
1785     <br><b>
1786     Error returns from <b>pcre_dfa_exec()</b>
1787     </b><br>
1788     <P>
1789     The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1790     Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1791     described
1792     <a href="#errorlist">above.</a>
1793     There are in addition the following errors that are specific to
1794     <b>pcre_dfa_exec()</b>:
1795     <pre>
1796     PCRE_ERROR_DFA_UITEM (-16)
1797     </pre>
1798     This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1799     that it does not support, for instance, the use of \C or a back reference.
1800     <pre>
1801     PCRE_ERROR_DFA_UCOND (-17)
1802     </pre>
1803 nigel 93 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1804     uses a back reference for the condition, or a test for recursion in a specific
1805     group. These are not supported.
1806 nigel 77 <pre>
1808     </pre>
1809     This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1810     block that contains a setting of the <i>match_limit</i> field. This is not
1811     supported (it is meaningless).
1812     <pre>
1813     PCRE_ERROR_DFA_WSSIZE (-19)
1814     </pre>
1815     This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1816     <i>workspace</i> vector.
1817     <pre>
1819     </pre>
1820     When a recursive subpattern is processed, the matching function calls itself
1821     recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1822     error is given if the output vector is not large enough. This should be
1823     extremely rare, as a vector of size 1000 is used.
1824     </P>
1825 nigel 93 <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1826 nigel 77 <P>
1827 nigel 93 <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1828     <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1829     <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1830     </P>
1831 ph10 99 <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1832 nigel 93 <P>
1833 ph10 99 Philip Hazel
1834 nigel 63 <br>
1835 ph10 99 University Computing Service
1836     <br>
1837     Cambridge CB2 3QH, England.
1838     <br>
1839     </P>
1840     <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1841     <P>
1842     Last updated: 06 March 2007
1843     <br>
1844     Copyright &copy; 1997-2007 University of Cambridge.
1845     <br>
1846 nigel 75 <p>
1847     Return to the <a href="index.html">PCRE index page</a>.
1848     </p>


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

ViewVC Help
Powered by ViewVC 1.1.12