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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 416 - (hide annotations) (download) (as text)
Sat Apr 11 14:34:02 2009 UTC (5 years, 3 months ago) by ph10
File MIME type: text/html
File size: 90808 byte(s)
File tidies for 7.9 release.

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 ph10 313 documentation describes how to compile and 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 ph10 227 <a name="newlines"></a></P>
247 nigel 91 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248 nigel 63 <P>
249 ph10 150 PCRE supports five different conventions for indicating line breaks in
250 nigel 93 strings: a single CR (carriage return) character, a single LF (linefeed)
251 ph10 150 character, the two-character sequence CRLF, any of the three preceding, or any
252     Unicode newline sequence. The Unicode newline sequences are the three just
253     mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
254     U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
255     (paragraph separator, U+2029).
256 nigel 93 </P>
257     <P>
258     Each of the first three conventions is used by at least one operating system as
259     its standard newline sequence. When PCRE is built, a default can be specified.
260     The default default is LF, which is the Unix standard. When PCRE is run, the
261     default can be overridden, either when a pattern is compiled, or when it is
262     matched.
263     </P>
264     <P>
265 ph10 227 At compile time, the newline convention can be specified by the <i>options</i>
266     argument of <b>pcre_compile()</b>, or it can be specified by special text at the
267     start of the pattern itself; this overrides any other settings. See the
268     <a href="pcrepattern.html"><b>pcrepattern</b></a>
269     page for details of the special character sequences.
270     </P>
271     <P>
272 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
273 nigel 93 pair of characters that indicate a line break". The choice of newline
274     convention affects the handling of the dot, circumflex, and dollar
275     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
276     recognized line ending sequence, the match position advancement for a
277 ph10 227 non-anchored pattern. There is more detail about this in the
278     <a href="#execoptions">section on <b>pcre_exec()</b> options</a>
279 ph10 231 below.
280 nigel 91 </P>
281 ph10 231 <P>
282     The choice of newline convention does not affect the interpretation of
283     the \n or \r escape sequences, nor does it affect what \R matches, which is
284     controlled in a similar way, but by separate options.
285     </P>
286 nigel 91 <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
287     <P>
288 nigel 63 The PCRE functions can be used in multi-threading applications, with the
289 nigel 73 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
290     <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
291     callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
292 nigel 63 </P>
293     <P>
294     The compiled form of a regular expression is not altered during matching, so
295     the same compiled pattern can safely be used by several threads at once.
296     </P>
297 nigel 91 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
298 nigel 63 <P>
299 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
300     time, possibly by a different program, and even on a host other than the one on
301     which it was compiled. Details are given in the
302     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
303 ph10 155 documentation. However, compiling a regular expression with one version of PCRE
304     for use with a different version is not guaranteed to work and may cause
305     crashes.
306 nigel 75 </P>
307 nigel 91 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
308 nigel 75 <P>
309 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
310     </P>
311     <P>
312     The function <b>pcre_config()</b> makes it possible for a PCRE client to
313     discover which optional features have been compiled into the PCRE library. The
314     <a href="pcrebuild.html"><b>pcrebuild</b></a>
315     documentation has more details about these optional features.
316     </P>
317     <P>
318     The first argument for <b>pcre_config()</b> is an integer, specifying which
319     information is required; the second argument is a pointer to a variable into
320     which the information is placed. The following information is available:
321     <pre>
322     PCRE_CONFIG_UTF8
323 nigel 75 </pre>
324 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
325     otherwise it is set to zero.
326     <pre>
327 nigel 75 PCRE_CONFIG_UNICODE_PROPERTIES
328     </pre>
329     The output is an integer that is set to one if support for Unicode character
330     properties is available; otherwise it is set to zero.
331     <pre>
332 nigel 63 PCRE_CONFIG_NEWLINE
333 nigel 75 </pre>
334 nigel 91 The output is an integer whose value specifies the default character sequence
335 nigel 93 that is recognized as meaning "newline". The four values that are supported
336 ph10 392 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
337     Though they are derived from ASCII, the same values are returned in EBCDIC
338     environments. The default should normally correspond to the standard sequence
339     for your operating system.
340 nigel 63 <pre>
341 ph10 231 PCRE_CONFIG_BSR
342     </pre>
343     The output is an integer whose value indicates what character sequences the \R
344     escape sequence matches by default. A value of 0 means that \R matches any
345     Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
346     or CRLF. The default can be overridden when a pattern is compiled or matched.
347     <pre>
348 nigel 63 PCRE_CONFIG_LINK_SIZE
349 nigel 75 </pre>
350 nigel 63 The output is an integer that contains the number of bytes used for internal
351     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
352     allow larger regular expressions to be compiled, at the expense of slower
353     matching. The default value of 2 is sufficient for all but the most massive
354     patterns, since it allows the compiled pattern to be up to 64K in size.
355     <pre>
356     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
357 nigel 75 </pre>
358 nigel 63 The output is an integer that contains the threshold above which the POSIX
359     interface uses <b>malloc()</b> for output vectors. Further details are given in
360 nigel 75 the
361     <a href="pcreposix.html"><b>pcreposix</b></a>
362     documentation.
363 nigel 63 <pre>
364     PCRE_CONFIG_MATCH_LIMIT
365 nigel 75 </pre>
366 ph10 392 The output is a long integer that gives the default limit for the number of
367 nigel 63 internal matching function calls in a <b>pcre_exec()</b> execution. Further
368     details are given with <b>pcre_exec()</b> below.
369 nigel 73 <pre>
370 nigel 87 PCRE_CONFIG_MATCH_LIMIT_RECURSION
371     </pre>
372 ph10 392 The output is a long integer that gives the default limit for the depth of
373 nigel 87 recursion when calling the internal matching function in a <b>pcre_exec()</b>
374     execution. Further details are given with <b>pcre_exec()</b> below.
375     <pre>
376 nigel 73 PCRE_CONFIG_STACKRECURSE
377 nigel 75 </pre>
378 nigel 77 The output is an integer that is set to one if internal recursion when running
379     <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
380     to remember their state. This is the usual way that PCRE is compiled. The
381     output is zero if PCRE was compiled to use blocks of data on the heap instead
382     of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
383     <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
384     avoiding the use of the stack.
385 nigel 73 </P>
386 nigel 91 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
387 nigel 63 <P>
388     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
389     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
390     <b>const unsigned char *<i>tableptr</i>);</b>
391 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
392     <b>int *<i>errorcodeptr</i>,</b>
393     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
394     <b>const unsigned char *<i>tableptr</i>);</b>
395 nigel 63 </P>
396     <P>
397 nigel 77 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
398     called to compile a pattern into an internal form. The only difference between
399     the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
400     <i>errorcodeptr</i>, via which a numerical error code can be returned.
401 nigel 63 </P>
402     <P>
403 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
404     <i>pattern</i> argument. A pointer to a single block of memory that is obtained
405     via <b>pcre_malloc</b> is returned. This contains the compiled code and related
406     data. The <b>pcre</b> type is defined for the returned block; this is a typedef
407     for a structure whose contents are not externally defined. It is up to the
408 nigel 91 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
409 nigel 77 </P>
410     <P>
411 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
412     depend on memory location, the complete <b>pcre</b> data block is not
413 nigel 75 fully relocatable, because it may contain a copy of the <i>tableptr</i>
414     argument, which is an address (see below).
415 nigel 63 </P>
416     <P>
417 nigel 93 The <i>options</i> argument contains various bit settings that affect the
418 nigel 75 compilation. It should be zero if no options are required. The available
419 ph10 416 options are described below. Some of them (in particular, those that are
420     compatible with Perl, but also some others) can also be set and unset from
421     within the pattern (see the detailed description in the
422 nigel 75 <a href="pcrepattern.html"><b>pcrepattern</b></a>
423 ph10 416 documentation). For those options that can be different in different parts of
424     the pattern, the contents of the <i>options</i> argument specifies their initial
425     settings at the start of compilation and execution. The PCRE_ANCHORED and
426     PCRE_NEWLINE_<i>xxx</i> options can be set at the time of matching as well as at
427     compile time.
428 nigel 63 </P>
429     <P>
430     If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
431     Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
432     NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
433 nigel 87 error message. This is a static string that is part of the library. You must
434     not try to free it. The offset from the start of the pattern to the character
435     where the error was discovered is placed in the variable pointed to by
436 nigel 63 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
437     </P>
438     <P>
439 nigel 77 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
440     <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
441     returned via this argument in the event of an error. This is in addition to the
442     textual error message. Error codes and messages are listed below.
443     </P>
444     <P>
445 nigel 63 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
446 nigel 75 character tables that are built when PCRE is compiled, using the default C
447     locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
448     call to <b>pcre_maketables()</b>. This value is stored with the compiled
449     pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
450     passed to it. For more discussion, see the section on locale support below.
451 nigel 63 </P>
452     <P>
453     This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
454     <pre>
455     pcre *re;
456     const char *error;
457     int erroffset;
458     re = pcre_compile(
459     "^A.*Z", /* the pattern */
460     0, /* default options */
461     &error, /* for error message */
462     &erroffset, /* for error offset */
463     NULL); /* use default character tables */
464 nigel 75 </pre>
465     The following names for option bits are defined in the <b>pcre.h</b> header
466     file:
467 nigel 63 <pre>
468     PCRE_ANCHORED
469 nigel 75 </pre>
470 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
471 nigel 75 constrained to match only at the first matching point in the string that is
472 nigel 63 being searched (the "subject string"). This effect can also be achieved by
473     appropriate constructs in the pattern itself, which is the only way to do it in
474     Perl.
475     <pre>
476 nigel 75 PCRE_AUTO_CALLOUT
477     </pre>
478     If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
479     all with number 255, before each pattern item. For discussion of the callout
480     facility, see the
481     <a href="pcrecallout.html"><b>pcrecallout</b></a>
482     documentation.
483     <pre>
484 ph10 231 PCRE_BSR_ANYCRLF
485     PCRE_BSR_UNICODE
486     </pre>
487     These options (which are mutually exclusive) control what the \R escape
488     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
489     match any Unicode newline sequence. The default is specified when PCRE is
490     built. It can be overridden from within the pattern, or by setting an option
491     when a compiled pattern is matched.
492     <pre>
493 nigel 63 PCRE_CASELESS
494 nigel 75 </pre>
495 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
496     letters. It is equivalent to Perl's /i option, and it can be changed within a
497 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
498     concept of case for characters whose values are less than 128, so caseless
499     matching is always possible. For characters with higher values, the concept of
500     case is supported if PCRE is compiled with Unicode property support, but not
501     otherwise. If you want to use caseless matching for characters 128 and above,
502     you must ensure that PCRE is compiled with Unicode property support as well as
503     with UTF-8 support.
504 nigel 63 <pre>
505     PCRE_DOLLAR_ENDONLY
506 nigel 75 </pre>
507 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
508     end of the subject string. Without this option, a dollar also matches
509 nigel 91 immediately before a newline at the end of the string (but not before any other
510     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
511     There is no equivalent to this option in Perl, and no way to set it within a
512     pattern.
513 nigel 63 <pre>
514     PCRE_DOTALL
515 nigel 75 </pre>
516 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
517 nigel 91 including those that indicate newline. Without it, a dot does not match when
518     the current position is at a newline. This option is equivalent to Perl's /s
519     option, and it can be changed within a pattern by a (?s) option setting. A
520 nigel 93 negative class such as [^a] always matches newline characters, independent of
521     the setting of this option.
522 nigel 63 <pre>
523 nigel 91 PCRE_DUPNAMES
524     </pre>
525     If this bit is set, names used to identify capturing subpatterns need not be
526     unique. This can be helpful for certain types of pattern when it is known that
527     only one instance of the named subpattern can ever be matched. There are more
528     details of named subpatterns below; see also the
529     <a href="pcrepattern.html"><b>pcrepattern</b></a>
530     documentation.
531     <pre>
532 nigel 63 PCRE_EXTENDED
533 nigel 75 </pre>
534 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
535     ignored except when escaped or inside a character class. Whitespace does not
536     include the VT character (code 11). In addition, characters between an
537 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
538     ignored. This is equivalent to Perl's /x option, and it can be changed within a
539     pattern by a (?x) option setting.
540 nigel 63 </P>
541     <P>
542     This option makes it possible to include comments inside complicated patterns.
543     Note, however, that this applies only to data characters. Whitespace characters
544     may never appear within special character sequences in a pattern, for example
545     within the sequence (?( which introduces a conditional subpattern.
546     <pre>
547     PCRE_EXTRA
548 nigel 75 </pre>
549 nigel 63 This option was invented in order to turn on additional functionality of PCRE
550     that is incompatible with Perl, but it is currently of very little use. When
551     set, any backslash in a pattern that is followed by a letter that has no
552     special meaning causes an error, thus reserving these combinations for future
553     expansion. By default, as in Perl, a backslash followed by a letter with no
554 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
555     give a warning for this.) There are at present no other features controlled by
556     this option. It can also be set by a (?X) option setting within a pattern.
557 nigel 63 <pre>
558 nigel 77 PCRE_FIRSTLINE
559     </pre>
560     If this option is set, an unanchored pattern is required to match before or at
561 nigel 91 the first newline in the subject string, though the matched text may continue
562     over the newline.
563 nigel 77 <pre>
564 ph10 345 PCRE_JAVASCRIPT_COMPAT
565     </pre>
566     If this option is set, PCRE's behaviour is changed in some ways so that it is
567     compatible with JavaScript rather than Perl. The changes are as follows:
568     </P>
569     <P>
570     (1) A lone closing square bracket in a pattern causes a compile-time error,
571     because this is illegal in JavaScript (by default it is treated as a data
572     character). Thus, the pattern AB]CD becomes illegal when this option is set.
573     </P>
574     <P>
575     (2) At run time, a back reference to an unset subpattern group matches an empty
576     string (by default this causes the current matching alternative to fail). A
577     pattern such as (\1)(a) succeeds when this option is set (assuming it can find
578     an "a" in the subject), whereas it fails by default, for Perl compatibility.
579     <pre>
580 nigel 63 PCRE_MULTILINE
581 nigel 75 </pre>
582     By default, PCRE treats the subject string as consisting of a single line of
583     characters (even if it actually contains newlines). The "start of line"
584 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
585     line" metacharacter ($) matches only at the end of the string, or before a
586     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
587     Perl.
588     </P>
589     <P>
590     When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
591 nigel 91 match immediately following or immediately before internal newlines in the
592     subject string, respectively, as well as at the very start and end. This is
593     equivalent to Perl's /m option, and it can be changed within a pattern by a
594     (?m) option setting. If there are no newlines in a subject string, or no
595 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
596     <pre>
597 nigel 91 PCRE_NEWLINE_CR
598     PCRE_NEWLINE_LF
599     PCRE_NEWLINE_CRLF
600 ph10 150 PCRE_NEWLINE_ANYCRLF
601 nigel 93 PCRE_NEWLINE_ANY
602 nigel 91 </pre>
603     These options override the default newline definition that was chosen when PCRE
604     was built. Setting the first or the second specifies that a newline is
605 nigel 93 indicated by a single character (CR or LF, respectively). Setting
606     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
607 ph10 150 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
608     preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
609     that any Unicode newline sequence should be recognized. The Unicode newline
610     sequences are the three just mentioned, plus the single characters VT (vertical
611     tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
612     separator, U+2028), and PS (paragraph separator, U+2029). The last two are
613     recognized only in UTF-8 mode.
614 nigel 91 </P>
615     <P>
616 nigel 93 The newline setting in the options word uses three bits that are treated
617 ph10 150 as a number, giving eight possibilities. Currently only six are used (default
618     plus the five values above). This means that if you set more than one newline
619 nigel 93 option, the combination may or may not be sensible. For example,
620     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
621 ph10 150 other combinations may yield unused numbers and cause an error.
622 nigel 93 </P>
623     <P>
624     The only time that a line break is specially recognized when compiling a
625     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
626     class is encountered. This indicates a comment that lasts until after the next
627     line break sequence. In other circumstances, line break sequences are treated
628     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
629     as whitespace characters and are therefore ignored.
630     </P>
631     <P>
632     The newline option that is set at compile time becomes the default that is used
633     for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
634 nigel 91 <pre>
635 nigel 63 PCRE_NO_AUTO_CAPTURE
636 nigel 75 </pre>
637 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
638     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
639     were followed by ?: but named parentheses can still be used for capturing (and
640     they acquire numbers in the usual way). There is no equivalent of this option
641     in Perl.
642     <pre>
643     PCRE_UNGREEDY
644 nigel 75 </pre>
645 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
646     greedy by default, but become greedy if followed by "?". It is not compatible
647     with Perl. It can also be set by a (?U) option setting within the pattern.
648     <pre>
649     PCRE_UTF8
650 nigel 75 </pre>
651 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
652     of UTF-8 characters instead of single-byte character strings. However, it is
653 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
654 nigel 63 of this option provokes an error. Details of how this option changes the
655     behaviour of PCRE are given in the
656     <a href="pcre.html#utf8support">section on UTF-8 support</a>
657     in the main
658     <a href="pcre.html"><b>pcre</b></a>
659     page.
660 nigel 71 <pre>
661     PCRE_NO_UTF8_CHECK
662 nigel 75 </pre>
663 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
664 ph10 211 automatically checked. There is a discussion about the
665     <a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
666     in the main
667     <a href="pcre.html"><b>pcre</b></a>
668     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
669     returns an error. If you already know that your pattern is valid, and you want
670     to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
671     option. When it is set, the effect of passing an invalid UTF-8 string as a
672     pattern is undefined. It may cause your program to crash. Note that this option
673     can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
674     the UTF-8 validity checking of subject strings.
675 nigel 71 </P>
676 nigel 91 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
677 nigel 63 <P>
678 nigel 77 The following table lists the error codes than may be returned by
679     <b>pcre_compile2()</b>, along with the error messages that may be returned by
680 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
681     out of use. To avoid confusion, they have not been re-used.
682 nigel 77 <pre>
683     0 no error
684     1 \ at end of pattern
685     2 \c at end of pattern
686     3 unrecognized character follows \
687     4 numbers out of order in {} quantifier
688     5 number too big in {} quantifier
689     6 missing terminating ] for character class
690     7 invalid escape sequence in character class
691     8 range out of order in character class
692     9 nothing to repeat
693 nigel 93 10 [this code is not in use]
694 nigel 77 11 internal error: unexpected repeat
695 ph10 292 12 unrecognized character after (? or (?-
696 nigel 77 13 POSIX named classes are supported only within a class
697     14 missing )
698     15 reference to non-existent subpattern
699     16 erroffset passed as NULL
700     17 unknown option bit(s) set
701     18 missing ) after comment
702 nigel 93 19 [this code is not in use]
703 ph10 292 20 regular expression is too large
704 nigel 77 21 failed to get memory
705     22 unmatched parentheses
706     23 internal error: code overflow
707     24 unrecognized character after (?&#60;
708     25 lookbehind assertion is not fixed length
709 nigel 91 26 malformed number or name after (?(
710 nigel 77 27 conditional group contains more than two branches
711     28 assertion expected after (?(
712 ph10 182 29 (?R or (?[+-]digits must be followed by )
713 nigel 77 30 unknown POSIX class name
714     31 POSIX collating elements are not supported
715     32 this version of PCRE is not compiled with PCRE_UTF8 support
716 nigel 93 33 [this code is not in use]
717 nigel 77 34 character value in \x{...} sequence is too large
718     35 invalid condition (?(0)
719     36 \C not allowed in lookbehind assertion
720     37 PCRE does not support \L, \l, \N, \U, or \u
721     38 number after (?C is &#62; 255
722     39 closing ) for (?C expected
723     40 recursive call could loop indefinitely
724     41 unrecognized character after (?P
725 nigel 93 42 syntax error in subpattern name (missing terminator)
726 nigel 91 43 two named subpatterns have the same name
727 nigel 77 44 invalid UTF-8 string
728     45 support for \P, \p, and \X has not been compiled
729     46 malformed \P or \p sequence
730     47 unknown property name after \P or \p
731 nigel 91 48 subpattern name is too long (maximum 32 characters)
732 ph10 292 49 too many named subpatterns (maximum 10000)
733 ph10 202 50 [this code is not in use]
734 nigel 91 51 octal value is greater than \377 (not in UTF-8 mode)
735 nigel 93 52 internal error: overran compiling workspace
736     53 internal error: previously-checked referenced subpattern not found
737     54 DEFINE group contains more than one branch
738     55 repeating a DEFINE group is not allowed
739 ph10 231 56 inconsistent NEWLINE options
740 ph10 345 57 \g is not followed by a braced, angle-bracketed, or quoted
741     name/number or by a plain number
742     58 a numbered reference must not be zero
743 ph10 292 59 (*VERB) with an argument is not supported
744     60 (*VERB) not recognized
745     61 number is too big
746     62 subpattern name expected
747     63 digit expected after (?+
748 ph10 345 64 ] is an invalid data character in JavaScript compatibility mode
749 ph10 292 </pre>
750     The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
751     be used if the limits were changed when PCRE was built.
752 nigel 77 </P>
753 nigel 91 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
754 nigel 77 <P>
755     <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
756 nigel 63 <b>const char **<i>errptr</i>);</b>
757     </P>
758     <P>
759 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
760     more time analyzing it in order to speed up the time taken for matching. The
761 nigel 63 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
762 nigel 75 argument. If studying the pattern produces additional information that will
763     help speed up matching, <b>pcre_study()</b> returns a pointer to a
764     <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
765     results of the study.
766 nigel 63 </P>
767     <P>
768 nigel 75 The returned value from <b>pcre_study()</b> can be passed directly to
769     <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
770 nigel 63 fields that can be set by the caller before the block is passed; these are
771 nigel 75 described
772     <a href="#extradata">below</a>
773     in the section on matching a pattern.
774 nigel 63 </P>
775     <P>
776 nigel 77 If studying the pattern does not produce any additional information
777 nigel 75 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
778     wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
779     own <b>pcre_extra</b> block.
780 nigel 63 </P>
781     <P>
782 nigel 75 The second argument of <b>pcre_study()</b> contains option bits. At present, no
783     options are defined, and this argument should always be zero.
784     </P>
785     <P>
786 nigel 63 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
787     studying succeeds (even if no data is returned), the variable it points to is
788 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
789     static string that is part of the library. You must not try to free it. You
790     should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
791     sure that it has run successfully.
792 nigel 63 </P>
793     <P>
794     This is a typical call to <b>pcre_study</b>():
795     <pre>
796     pcre_extra *pe;
797     pe = pcre_study(
798     re, /* result of pcre_compile() */
799     0, /* no options exist */
800     &error); /* set to NULL or points to a message */
801 nigel 75 </pre>
802 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
803     not have a single fixed starting character. A bitmap of possible starting
804 nigel 75 bytes is created.
805     <a name="localesupport"></a></P>
806 nigel 91 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
807 nigel 63 <P>
808 ph10 142 PCRE handles caseless matching, and determines whether characters are letters,
809 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
810 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
811 nigel 75 less than 128. Higher-valued codes never match escapes such as \w or \d, but
812     can be tested with \p if PCRE is built with Unicode character property
813 ph10 142 support. The use of locales with Unicode is discouraged. If you are handling
814     characters with codes greater than 128, you should either use UTF-8 and
815     Unicode, or use locales, but not try to mix the two.
816 nigel 63 </P>
817     <P>
818 ph10 142 PCRE contains an internal set of tables that are used when the final argument
819     of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
820     Normally, the internal tables recognize only ASCII characters. However, when
821     PCRE is built, it is possible to cause the internal tables to be rebuilt in the
822     default "C" locale of the local system, which may cause them to be different.
823 nigel 63 </P>
824     <P>
825 ph10 142 The internal tables can always be overridden by tables supplied by the
826     application that calls PCRE. These may be created in a different locale from
827     the default. As more and more applications change to using Unicode, the need
828     for this locale support is expected to die away.
829     </P>
830     <P>
831 nigel 75 External tables are built by calling the <b>pcre_maketables()</b> function,
832     which has no arguments, in the relevant locale. The result can then be passed
833     to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
834     example, to build and use tables that are appropriate for the French locale
835     (where accented characters with values greater than 128 are treated as letters),
836     the following code could be used:
837 nigel 63 <pre>
838 nigel 75 setlocale(LC_CTYPE, "fr_FR");
839 nigel 63 tables = pcre_maketables();
840     re = pcre_compile(..., tables);
841 nigel 75 </pre>
842 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
843     are using Windows, the name for the French locale is "french".
844     </P>
845     <P>
846 nigel 75 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
847     obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
848     that the memory containing the tables remains available for as long as it is
849     needed.
850 nigel 63 </P>
851     <P>
852 nigel 75 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
853 nigel 63 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
854 nigel 75 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
855     pattern, compilation, studying and matching all happen in the same locale, but
856     different patterns can be compiled in different locales.
857 nigel 63 </P>
858     <P>
859 nigel 75 It is possible to pass a table pointer or NULL (indicating the use of the
860     internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
861     this facility could be used to match a pattern in a different locale from the
862     one in which it was compiled. Passing table pointers at run time is discussed
863     below in the section on matching a pattern.
864     </P>
865 nigel 91 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
866 nigel 75 <P>
867 nigel 63 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
868     <b>int <i>what</i>, void *<i>where</i>);</b>
869     </P>
870     <P>
871     The <b>pcre_fullinfo()</b> function returns information about a compiled
872     pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
873     nevertheless retained for backwards compability (and is documented below).
874     </P>
875     <P>
876     The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
877     pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
878     the pattern was not studied. The third argument specifies which piece of
879     information is required, and the fourth argument is a pointer to a variable
880     to receive the data. The yield of the function is zero for success, or one of
881     the following negative numbers:
882     <pre>
883     PCRE_ERROR_NULL the argument <i>code</i> was NULL
884     the argument <i>where</i> was NULL
885     PCRE_ERROR_BADMAGIC the "magic number" was not found
886     PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
887 nigel 75 </pre>
888     The "magic number" is placed at the start of each compiled pattern as an simple
889     check against passing an arbitrary memory pointer. Here is a typical call of
890     <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
891 nigel 63 <pre>
892     int rc;
893 nigel 91 size_t length;
894 nigel 63 rc = pcre_fullinfo(
895     re, /* result of pcre_compile() */
896     pe, /* result of pcre_study(), or NULL */
897     PCRE_INFO_SIZE, /* what is required */
898     &length); /* where to put the data */
899 nigel 75 </pre>
900 nigel 63 The possible values for the third argument are defined in <b>pcre.h</b>, and are
901     as follows:
902     <pre>
903     PCRE_INFO_BACKREFMAX
904 nigel 75 </pre>
905 nigel 63 Return the number of the highest back reference in the pattern. The fourth
906     argument should point to an <b>int</b> variable. Zero is returned if there are
907     no back references.
908     <pre>
909     PCRE_INFO_CAPTURECOUNT
910 nigel 75 </pre>
911 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
912 nigel 75 should point to an <b>int</b> variable.
913 nigel 63 <pre>
914 nigel 77 PCRE_INFO_DEFAULT_TABLES
915 nigel 75 </pre>
916     Return a pointer to the internal default character tables within PCRE. The
917     fourth argument should point to an <b>unsigned char *</b> variable. This
918     information call is provided for internal use by the <b>pcre_study()</b>
919     function. External callers can cause PCRE to use its internal tables by passing
920     a NULL table pointer.
921     <pre>
922 nigel 63 PCRE_INFO_FIRSTBYTE
923 nigel 75 </pre>
924 nigel 63 Return information about the first byte of any matched string, for a
925 nigel 91 non-anchored pattern. The fourth argument should point to an <b>int</b>
926     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
927     still recognized for backwards compatibility.)
928 nigel 63 </P>
929     <P>
930 nigel 75 If there is a fixed first byte, for example, from a pattern such as
931 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
932 nigel 75 <br>
933     <br>
934 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
935     starts with "^", or
936 nigel 75 <br>
937     <br>
938 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
939     (if it were set, the pattern would be anchored),
940 nigel 75 <br>
941     <br>
942 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
943     subject string or after any newline within the string. Otherwise -2 is
944     returned. For anchored patterns, -2 is returned.
945     <pre>
946     PCRE_INFO_FIRSTTABLE
947 nigel 75 </pre>
948 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
949     table indicating a fixed set of bytes for the first byte in any matching
950     string, a pointer to the table is returned. Otherwise NULL is returned. The
951     fourth argument should point to an <b>unsigned char *</b> variable.
952     <pre>
953 ph10 227 PCRE_INFO_HASCRORLF
954     </pre>
955     Return 1 if the pattern contains any explicit matches for CR or LF characters,
956 ph10 243 otherwise 0. The fourth argument should point to an <b>int</b> variable. An
957     explicit match is either a literal CR or LF character, or \r or \n.
958 ph10 227 <pre>
959 ph10 172 PCRE_INFO_JCHANGED
960     </pre>
961 ph10 286 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
962     0. The fourth argument should point to an <b>int</b> variable. (?J) and
963     (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
964 ph10 172 <pre>
965 nigel 63 PCRE_INFO_LASTLITERAL
966 nigel 75 </pre>
967 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
968     string, other than at its start, if such a byte has been recorded. The fourth
969     argument should point to an <b>int</b> variable. If there is no such byte, -1 is
970     returned. For anchored patterns, a last literal byte is recorded only if it
971     follows something of variable length. For example, for the pattern
972     /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
973     is -1.
974 nigel 63 <pre>
975     PCRE_INFO_NAMECOUNT
976     PCRE_INFO_NAMEENTRYSIZE
977     PCRE_INFO_NAMETABLE
978 nigel 75 </pre>
979 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
980     names are just an additional way of identifying the parentheses, which still
981 nigel 91 acquire numbers. Several convenience functions such as
982     <b>pcre_get_named_substring()</b> are provided for extracting captured
983     substrings by name. It is also possible to extract the data directly, by first
984     converting the name to a number in order to access the correct pointers in the
985     output vector (described with <b>pcre_exec()</b> below). To do the conversion,
986     you need to use the name-to-number map, which is described by these three
987     values.
988 nigel 63 </P>
989     <P>
990     The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
991     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
992     entry; both of these return an <b>int</b> value. The entry size depends on the
993     length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
994     entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
995     are the number of the capturing parenthesis, most significant byte first. The
996     rest of the entry is the corresponding name, zero terminated. The names are in
997 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
998     their parentheses numbers. For example, consider the following pattern (assume
999 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
1000     <pre>
1001 nigel 93 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
1002 nigel 75 </pre>
1003 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
1004     in the table is eight bytes long. The table is as follows, with non-printing
1005 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
1006 nigel 63 <pre>
1007     00 01 d a t e 00 ??
1008     00 05 d a y 00 ?? ??
1009     00 04 m o n t h 00
1010     00 02 y e a r 00 ??
1011 nigel 75 </pre>
1012     When writing code to extract data from named subpatterns using the
1013 nigel 91 name-to-number map, remember that the length of the entries is likely to be
1014 nigel 75 different for each compiled pattern.
1015 nigel 63 <pre>
1016 ph10 172 PCRE_INFO_OKPARTIAL
1017     </pre>
1018     Return 1 if the pattern can be used for partial matching, otherwise 0. The
1019     fourth argument should point to an <b>int</b> variable. The
1020     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1021     documentation lists the restrictions that apply to patterns when partial
1022     matching is used.
1023     <pre>
1024 nigel 63 PCRE_INFO_OPTIONS
1025 nigel 75 </pre>
1026 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
1027     argument should point to an <b>unsigned long int</b> variable. These option bits
1028     are those specified in the call to <b>pcre_compile()</b>, modified by any
1029 ph10 197 top-level option settings at the start of the pattern itself. In other words,
1030     they are the options that will be in force when matching starts. For example,
1031     if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1032     result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1033 nigel 63 </P>
1034     <P>
1035     A pattern is automatically anchored by PCRE if all of its top-level
1036     alternatives begin with one of the following:
1037     <pre>
1038     ^ unless PCRE_MULTILINE is set
1039     \A always
1040     \G always
1041 nigel 75 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
1042     </pre>
1043 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1044     <b>pcre_fullinfo()</b>.
1045     <pre>
1046     PCRE_INFO_SIZE
1047 nigel 75 </pre>
1048 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
1049     the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
1050     place the compiled data. The fourth argument should point to a <b>size_t</b>
1051     variable.
1052     <pre>
1053     PCRE_INFO_STUDYSIZE
1054 nigel 75 </pre>
1055     Return the size of the data block pointed to by the <i>study_data</i> field in
1056 nigel 63 a <b>pcre_extra</b> block. That is, it is the value that was passed to
1057     <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
1058     created by <b>pcre_study()</b>. The fourth argument should point to a
1059     <b>size_t</b> variable.
1060     </P>
1061 nigel 91 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
1062 nigel 63 <P>
1063     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
1064     <b>*<i>firstcharptr</i>);</b>
1065     </P>
1066     <P>
1067     The <b>pcre_info()</b> function is now obsolete because its interface is too
1068     restrictive to return all the available data about a compiled pattern. New
1069     programs should use <b>pcre_fullinfo()</b> instead. The yield of
1070     <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
1071     following negative numbers:
1072     <pre>
1073     PCRE_ERROR_NULL the argument <i>code</i> was NULL
1074     PCRE_ERROR_BADMAGIC the "magic number" was not found
1075 nigel 75 </pre>
1076 nigel 63 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
1077     pattern was compiled is placed in the integer it points to (see
1078     PCRE_INFO_OPTIONS above).
1079     </P>
1080     <P>
1081     If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
1082     it is used to pass back information about the first character of any matched
1083     string (see PCRE_INFO_FIRSTBYTE above).
1084     </P>
1085 nigel 91 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1086 nigel 63 <P>
1087 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1088     </P>
1089     <P>
1090     The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1091     data block that contains a compiled pattern. It is provided for the benefit of
1092     applications that operate in an object-oriented manner, where different parts
1093     of the application may be using the same compiled pattern, but you want to free
1094     the block when they are all done.
1095     </P>
1096     <P>
1097     When a pattern is compiled, the reference count field is initialized to zero.
1098     It is changed only by calling this function, whose action is to add the
1099     <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1100     function is the new value. However, the value of the count is constrained to
1101     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1102     it is forced to the appropriate limit value.
1103     </P>
1104     <P>
1105     Except when it is zero, the reference count is not correctly preserved if a
1106     pattern is compiled on one host and then transferred to a host whose byte-order
1107     is different. (This seems a highly unlikely scenario.)
1108     </P>
1109 nigel 91 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1110 nigel 77 <P>
1111 nigel 63 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1112     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1113     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1114     </P>
1115     <P>
1116     The function <b>pcre_exec()</b> is called to match a subject string against a
1117 nigel 75 compiled pattern, which is passed in the <i>code</i> argument. If the
1118 nigel 63 pattern has been studied, the result of the study should be passed in the
1119 nigel 77 <i>extra</i> argument. This function is the main matching facility of the
1120     library, and it operates in a Perl-like manner. For specialist use there is
1121     also an alternative matching function, which is described
1122     <a href="#dfamatch">below</a>
1123     in the section about the <b>pcre_dfa_exec()</b> function.
1124 nigel 63 </P>
1125     <P>
1126 nigel 75 In most applications, the pattern will have been compiled (and optionally
1127     studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1128     possible to save compiled patterns and study data, and then use them later
1129     in different processes, possibly even on different hosts. For a discussion
1130     about this, see the
1131     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1132     documentation.
1133 nigel 63 </P>
1134     <P>
1135 nigel 75 Here is an example of a simple call to <b>pcre_exec()</b>:
1136 nigel 63 <pre>
1137     int rc;
1138     int ovector[30];
1139     rc = pcre_exec(
1140     re, /* result of pcre_compile() */
1141     NULL, /* we didn't study the pattern */
1142     "some string", /* the subject string */
1143     11, /* the length of the subject string */
1144     0, /* start at offset 0 in the subject */
1145     0, /* default options */
1146 nigel 75 ovector, /* vector of integers for substring information */
1147 nigel 77 30); /* number of elements (NOT size in bytes) */
1148 nigel 75 <a name="extradata"></a></PRE>
1149 nigel 63 </P>
1150 nigel 75 <br><b>
1151     Extra data for <b>pcre_exec()</b>
1152     </b><br>
1153 nigel 63 <P>
1154     If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1155     data block. The <b>pcre_study()</b> function returns such a block (when it
1156     doesn't return NULL), but you can also create one for yourself, and pass
1157 nigel 87 additional information in it. The <b>pcre_extra</b> block contains the following
1158     fields (not necessarily in this order):
1159 nigel 63 <pre>
1160     unsigned long int <i>flags</i>;
1161     void *<i>study_data</i>;
1162     unsigned long int <i>match_limit</i>;
1163 nigel 87 unsigned long int <i>match_limit_recursion</i>;
1164 nigel 63 void *<i>callout_data</i>;
1165 nigel 75 const unsigned char *<i>tables</i>;
1166     </pre>
1167 nigel 63 The <i>flags</i> field is a bitmap that specifies which of the other fields
1168     are set. The flag bits are:
1169     <pre>
1170     PCRE_EXTRA_STUDY_DATA
1171     PCRE_EXTRA_MATCH_LIMIT
1172 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1173 nigel 63 PCRE_EXTRA_CALLOUT_DATA
1174 nigel 75 PCRE_EXTRA_TABLES
1175     </pre>
1176 nigel 63 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1177     <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1178 nigel 75 the appropriate flag bit. You should not set this yourself, but you may add to
1179     the block by setting the other fields and their corresponding flag bits.
1180 nigel 63 </P>
1181     <P>
1182     The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1183     vast amount of resources when running patterns that are not going to match,
1184     but which have a very large number of possibilities in their search trees. The
1185 nigel 75 classic example is the use of nested unlimited repeats.
1186 nigel 63 </P>
1187     <P>
1188 nigel 75 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1189 nigel 87 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1190     number of times this function is called during a match, which has the effect of
1191     limiting the amount of backtracking that can take place. For patterns that are
1192     not anchored, the count restarts from zero for each position in the subject
1193     string.
1194 nigel 75 </P>
1195     <P>
1196 nigel 87 The default value for the limit can be set when PCRE is built; the default
1197 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1198 nigel 87 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1199     block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1200     the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1201     PCRE_ERROR_MATCHLIMIT.
1202 nigel 63 </P>
1203     <P>
1204 nigel 87 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1205     instead of limiting the total number of times that <b>match()</b> is called, it
1206     limits the depth of recursion. The recursion depth is a smaller number than the
1207     total number of calls, because not all calls to <b>match()</b> are recursive.
1208     This limit is of use only if it is set smaller than <i>match_limit</i>.
1209     </P>
1210     <P>
1211     Limiting the recursion depth limits the amount of stack that can be used, or,
1212     when PCRE has been compiled to use memory on the heap instead of the stack, the
1213     amount of heap memory that can be used.
1214     </P>
1215     <P>
1216     The default value for <i>match_limit_recursion</i> can be set when PCRE is
1217     built; the default default is the same value as the default for
1218     <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1219     with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1220     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1221     is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1222     </P>
1223     <P>
1224 nigel 63 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1225 nigel 75 which is described in the
1226     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1227     documentation.
1228 nigel 63 </P>
1229     <P>
1230 nigel 75 The <i>tables</i> field is used to pass a character tables pointer to
1231     <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1232     pattern. A non-NULL value is stored with the compiled pattern only if custom
1233     tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1234     If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1235     internal tables to be used. This facility is helpful when re-using patterns
1236     that have been saved after compiling with an external set of tables, because
1237     the external tables might be at a different address when <b>pcre_exec()</b> is
1238     called. See the
1239     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1240     documentation for a discussion of saving compiled patterns for later use.
1241 ph10 227 <a name="execoptions"></a></P>
1242 nigel 75 <br><b>
1243     Option bits for <b>pcre_exec()</b>
1244     </b><br>
1245 nigel 63 <P>
1246 nigel 75 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1247 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1248 ph10 392 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_START_OPTIMIZE,
1249     PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1250 nigel 63 <pre>
1251 nigel 75 PCRE_ANCHORED
1252     </pre>
1253     The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1254     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1255     to be anchored by virtue of its contents, it cannot be made unachored at
1256     matching time.
1257     <pre>
1258 ph10 231 PCRE_BSR_ANYCRLF
1259     PCRE_BSR_UNICODE
1260     </pre>
1261     These options (which are mutually exclusive) control what the \R escape
1262     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1263     match any Unicode newline sequence. These options override the choice that was
1264     made or defaulted when the pattern was compiled.
1265     <pre>
1266 nigel 91 PCRE_NEWLINE_CR
1267     PCRE_NEWLINE_LF
1268     PCRE_NEWLINE_CRLF
1269 ph10 150 PCRE_NEWLINE_ANYCRLF
1270 nigel 93 PCRE_NEWLINE_ANY
1271 nigel 91 </pre>
1272     These options override the newline definition that was chosen or defaulted when
1273 nigel 93 the pattern was compiled. For details, see the description of
1274     <b>pcre_compile()</b> above. During matching, the newline choice affects the
1275     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1276     the way the match position is advanced after a match failure for an unanchored
1277 ph10 227 pattern.
1278     </P>
1279     <P>
1280     When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1281     match attempt for an unanchored pattern fails when the current position is at a
1282 ph10 231 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1283 ph10 227 characters, the match position is advanced by two characters instead of one, in
1284     other words, to after the CRLF.
1285     </P>
1286     <P>
1287     The above rule is a compromise that makes the most common cases work as
1288     expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1289     set), it does not match the string "\r\nA" because, after failing at the
1290     start, it skips both the CR and the LF before retrying. However, the pattern
1291     [\r\n]A does match that string, because it contains an explicit CR or LF
1292     reference, and so advances only by one character after the first failure.
1293     </P>
1294     <P>
1295 ph10 231 An explicit match for CR of LF is either a literal appearance of one of those
1296     characters, or one of the \r or \n escape sequences. Implicit matches such as
1297     [^X] do not count, nor does \s (which includes CR and LF in the characters
1298     that it matches).
1299     </P>
1300     <P>
1301 ph10 227 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1302     valid newline sequence and explicit \r or \n escapes appear in the pattern.
1303 nigel 91 <pre>
1304 nigel 63 PCRE_NOTBOL
1305 nigel 75 </pre>
1306     This option specifies that first character of the subject string is not the
1307     beginning of a line, so the circumflex metacharacter should not match before
1308     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1309     never to match. This option affects only the behaviour of the circumflex
1310     metacharacter. It does not affect \A.
1311 nigel 63 <pre>
1312     PCRE_NOTEOL
1313 nigel 75 </pre>
1314     This option specifies that the end of the subject string is not the end of a
1315     line, so the dollar metacharacter should not match it nor (except in multiline
1316     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1317     compile time) causes dollar never to match. This option affects only the
1318     behaviour of the dollar metacharacter. It does not affect \Z or \z.
1319 nigel 63 <pre>
1320     PCRE_NOTEMPTY
1321 nigel 75 </pre>
1322 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1323     there are alternatives in the pattern, they are tried. If all the alternatives
1324     match the empty string, the entire match fails. For example, if the pattern
1325     <pre>
1326     a?b?
1327 nigel 75 </pre>
1328 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1329     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1330     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1331     </P>
1332     <P>
1333     Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1334     of a pattern match of the empty string within its <b>split()</b> function, and
1335     when using the /g modifier. It is possible to emulate Perl's behaviour after
1336     matching a null string by first trying the match again at the same offset with
1337 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1338     starting offset (see below) and trying an ordinary match again. There is some
1339     code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1340     <pre>
1341 ph10 392 PCRE_NO_START_OPTIMIZE
1342     </pre>
1343     There are a number of optimizations that <b>pcre_exec()</b> uses at the start of
1344     a match, in order to speed up the process. For example, if it is known that a
1345     match must start with a specific character, it searches the subject for that
1346     character, and fails immediately if it cannot find it, without actually running
1347     the main matching function. When callouts are in use, these optimizations can
1348     cause them to be skipped. This option disables the "start-up" optimizations,
1349     causing performance to suffer, but ensuring that the callouts do occur.
1350     <pre>
1351 nigel 75 PCRE_NO_UTF8_CHECK
1352     </pre>
1353     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1354     string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1355     The value of <i>startoffset</i> is also checked to ensure that it points to the
1356 ph10 211 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1357     strings in the
1358     <a href="pcre.html#utf8strings">section on UTF-8 support</a>
1359     in the main
1360     <a href="pcre.html"><b>pcre</b></a>
1361     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
1362     the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,
1363     PCRE_ERROR_BADUTF8_OFFSET is returned.
1364 nigel 63 </P>
1365     <P>
1366 nigel 75 If you already know that your subject is valid, and you want to skip these
1367     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1368     calling <b>pcre_exec()</b>. You might want to do this for the second and
1369     subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1370     all the matches in a single subject string. However, you should be sure that
1371     the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1372     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1373     subject, or a value of <i>startoffset</i> that does not point to the start of a
1374     UTF-8 character, is undefined. Your program may crash.
1375     <pre>
1376     PCRE_PARTIAL
1377     </pre>
1378     This option turns on the partial matching feature. If the subject string fails
1379     to match the pattern, but at some point during the matching process the end of
1380     the subject was reached (that is, the subject partially matches the pattern and
1381     the failure to match occurred only because there were not enough subject
1382     characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1383     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1384     may appear in the pattern. These are discussed in the
1385     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1386     documentation.
1387     </P>
1388     <br><b>
1389     The string to be matched by <b>pcre_exec()</b>
1390     </b><br>
1391     <P>
1392 nigel 63 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1393 ph10 371 <i>subject</i>, a length (in bytes) in <i>length</i>, and a starting byte offset
1394     in <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of
1395     a UTF-8 character. Unlike the pattern string, the subject may contain binary
1396     zero bytes. When the starting offset is zero, the search for a match starts at
1397     the beginning of the subject, and this is by far the most common case.
1398 nigel 63 </P>
1399     <P>
1400     A non-zero starting offset is useful when searching for another match in the
1401     same subject by calling <b>pcre_exec()</b> again after a previous success.
1402     Setting <i>startoffset</i> differs from just passing over a shortened string and
1403     setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1404     lookbehind. For example, consider the pattern
1405     <pre>
1406     \Biss\B
1407 nigel 75 </pre>
1408 nigel 63 which finds occurrences of "iss" in the middle of words. (\B matches only if
1409     the current position in the subject is not a word boundary.) When applied to
1410     the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1411     occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1412     subject, namely "issipi", it does not match, because \B is always false at the
1413     start of the subject, which is deemed to be a word boundary. However, if
1414     <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1415     set to 4, it finds the second occurrence of "iss" because it is able to look
1416     behind the starting point to discover that it is preceded by a letter.
1417     </P>
1418     <P>
1419     If a non-zero starting offset is passed when the pattern is anchored, one
1420 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1421 nigel 63 pattern does not require the match to be at the start of the subject.
1422     </P>
1423 nigel 75 <br><b>
1424     How <b>pcre_exec()</b> returns captured substrings
1425     </b><br>
1426 nigel 63 <P>
1427     In general, a pattern matches a certain portion of the subject, and in
1428     addition, further substrings from the subject may be picked out by parts of the
1429     pattern. Following the usage in Jeffrey Friedl's book, this is called
1430     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1431     a fragment of a pattern that picks out a substring. PCRE supports several other
1432     kinds of parenthesized subpattern that do not cause substrings to be captured.
1433     </P>
1434     <P>
1435 ph10 371 Captured substrings are returned to the caller via a vector of integers whose
1436     address is passed in <i>ovector</i>. The number of elements in the vector is
1437     passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>: this
1438     argument is NOT the size of <i>ovector</i> in bytes.
1439 nigel 63 </P>
1440     <P>
1441 nigel 75 The first two-thirds of the vector is used to pass back captured substrings,
1442     each substring using a pair of integers. The remaining third of the vector is
1443     used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1444 ph10 371 and is not available for passing back information. The number passed in
1445 nigel 75 <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1446     rounded down.
1447     </P>
1448     <P>
1449     When a match is successful, information about captured substrings is returned
1450     in pairs of integers, starting at the beginning of <i>ovector</i>, and
1451 ph10 371 continuing up to two-thirds of its length at the most. The first element of
1452     each pair is set to the byte offset of the first character in a substring, and
1453     the second is set to the byte offset of the first character after the end of a
1454     substring. <b>Note</b>: these values are always byte offsets, even in UTF-8
1455     mode. They are not character counts.
1456 nigel 63 </P>
1457     <P>
1458 ph10 371 The first pair of integers, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the
1459     portion of the subject string matched by the entire pattern. The next pair is
1460     used for the first capturing subpattern, and so on. The value returned by
1461     <b>pcre_exec()</b> is one more than the highest numbered pair that has been set.
1462     For example, if two substrings have been captured, the returned value is 3. If
1463     there are no capturing subpatterns, the return value from a successful match is
1464     1, indicating that just the first pair of offsets has been set.
1465     </P>
1466     <P>
1467 nigel 63 If a capturing subpattern is matched repeatedly, it is the last portion of the
1468 nigel 75 string that it matched that is returned.
1469 nigel 63 </P>
1470     <P>
1471 nigel 75 If the vector is too small to hold all the captured substring offsets, it is
1472     used as far as possible (up to two-thirds of its length), and the function
1473 ph10 371 returns a value of zero. If the substring offsets are not of interest,
1474     <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1475 nigel 63 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1476 nigel 75 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1477     has to get additional memory for use during matching. Thus it is usually
1478     advisable to supply an <i>ovector</i>.
1479 nigel 63 </P>
1480     <P>
1481 nigel 91 The <b>pcre_info()</b> function can be used to find out how many capturing
1482 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1483     <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1484     the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1485 nigel 91 </P>
1486     <P>
1487     It is possible for capturing subpattern number <i>n+1</i> to match some part of
1488     the subject when subpattern <i>n</i> has not been used at all. For example, if
1489     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1490     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1491     happens, both values in the offset pairs corresponding to unused subpatterns
1492     are set to -1.
1493     </P>
1494     <P>
1495     Offset values that correspond to unused subpatterns at the end of the
1496     expression are also set to -1. For example, if the string "abc" is matched
1497     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1498     return from the function is 2, because the highest used capturing subpattern
1499     number is 1. However, you can refer to the offsets for the second and third
1500     capturing subpatterns if you wish (assuming the vector is large enough, of
1501     course).
1502     </P>
1503     <P>
1504     Some convenience functions are provided for extracting the captured substrings
1505     as separate strings. These are described below.
1506 nigel 77 <a name="errorlist"></a></P>
1507 nigel 75 <br><b>
1508 nigel 91 Error return values from <b>pcre_exec()</b>
1509 nigel 75 </b><br>
1510 nigel 63 <P>
1511     If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1512     defined in the header file:
1513     <pre>
1514     PCRE_ERROR_NOMATCH (-1)
1515 nigel 75 </pre>
1516 nigel 63 The subject string did not match the pattern.
1517     <pre>
1518     PCRE_ERROR_NULL (-2)
1519 nigel 75 </pre>
1520 nigel 63 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1521     NULL and <i>ovecsize</i> was not zero.
1522     <pre>
1523     PCRE_ERROR_BADOPTION (-3)
1524 nigel 75 </pre>
1525 nigel 63 An unrecognized bit was set in the <i>options</i> argument.
1526     <pre>
1527     PCRE_ERROR_BADMAGIC (-4)
1528 nigel 75 </pre>
1529 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1530 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1531     compiled in an environment of one endianness is run in an environment with the
1532     other endianness. This is the error that PCRE gives when the magic number is
1533     not present.
1534 nigel 63 <pre>
1535 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1536 nigel 75 </pre>
1537 nigel 63 While running the pattern match, an unknown item was encountered in the
1538     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1539     of the compiled pattern.
1540     <pre>
1541     PCRE_ERROR_NOMEMORY (-6)
1542 nigel 75 </pre>
1543 nigel 63 If a pattern contains back references, but the <i>ovector</i> that is passed to
1544     <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1545     gets a block of memory at the start of matching to use for this purpose. If the
1546 nigel 75 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1547     automatically freed at the end of matching.
1548 nigel 63 <pre>
1549     PCRE_ERROR_NOSUBSTRING (-7)
1550 nigel 75 </pre>
1551 nigel 63 This error is used by the <b>pcre_copy_substring()</b>,
1552     <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1553     below). It is never returned by <b>pcre_exec()</b>.
1554     <pre>
1555     PCRE_ERROR_MATCHLIMIT (-8)
1556 nigel 75 </pre>
1557 nigel 87 The backtracking limit, as specified by the <i>match_limit</i> field in a
1558     <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1559     above.
1560     <pre>
1561 nigel 63 PCRE_ERROR_CALLOUT (-9)
1562 nigel 75 </pre>
1563 nigel 63 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1564     use by callout functions that want to yield a distinctive error code. See the
1565 nigel 75 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1566     documentation for details.
1567 nigel 71 <pre>
1568 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1569 nigel 75 </pre>
1570 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1571 nigel 73 <pre>
1572     PCRE_ERROR_BADUTF8_OFFSET (-11)
1573 nigel 75 </pre>
1574 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1575     of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1576 nigel 75 <pre>
1577 nigel 77 PCRE_ERROR_PARTIAL (-12)
1578 nigel 75 </pre>
1579     The subject string did not match, but it did match partially. See the
1580     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1581     documentation for details of partial matching.
1582     <pre>
1583 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1584 nigel 75 </pre>
1585     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1586     are not supported for partial matching. See the
1587     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1588     documentation for details of partial matching.
1589     <pre>
1590 nigel 77 PCRE_ERROR_INTERNAL (-14)
1591 nigel 75 </pre>
1592     An unexpected internal error has occurred. This error could be caused by a bug
1593     in PCRE or by overwriting of the compiled pattern.
1594     <pre>
1595 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1596 nigel 75 </pre>
1597     This error is given if the value of the <i>ovecsize</i> argument is negative.
1598 nigel 93 <pre>
1599     PCRE_ERROR_RECURSIONLIMIT (-21)
1600     </pre>
1601     The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1602     field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1603     description above.
1604     <pre>
1605     PCRE_ERROR_BADNEWLINE (-23)
1606     </pre>
1607     An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1608 nigel 73 </P>
1609 nigel 93 <P>
1610 ph10 197 Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
1611 nigel 93 </P>
1612 nigel 91 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1613 nigel 63 <P>
1614     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1615     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1616     <b>int <i>buffersize</i>);</b>
1617     </P>
1618     <P>
1619     <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1620     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1621     <b>const char **<i>stringptr</i>);</b>
1622     </P>
1623     <P>
1624     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1625     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1626     </P>
1627     <P>
1628     Captured substrings can be accessed directly by using the offsets returned by
1629     <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1630     <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1631     <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1632     as new, separate, zero-terminated strings. These functions identify substrings
1633     by number. The next section describes functions for extracting named
1634 nigel 91 substrings.
1635 nigel 63 </P>
1636     <P>
1637 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1638     further zero added on the end, but the result is not, of course, a C string.
1639     However, you can process such a string by referring to the length that is
1640     returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1641     Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1642     for handling strings containing binary zeros, because the end of the final
1643     string is not independently indicated.
1644     </P>
1645     <P>
1646 nigel 63 The first three arguments are the same for all three of these functions:
1647 nigel 75 <i>subject</i> is the subject string that has just been successfully matched,
1648 nigel 63 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1649     <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1650     captured by the match, including the substring that matched the entire regular
1651 nigel 75 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1652     than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1653     space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1654     number of elements in the vector divided by three.
1655 nigel 63 </P>
1656     <P>
1657     The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1658     extract a single substring, whose number is given as <i>stringnumber</i>. A
1659 nigel 75 value of zero extracts the substring that matched the entire pattern, whereas
1660 nigel 63 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1661     the string is placed in <i>buffer</i>, whose length is given by
1662     <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1663     obtained via <b>pcre_malloc</b>, and its address is returned via
1664     <i>stringptr</i>. The yield of the function is the length of the string, not
1665 nigel 93 including the terminating zero, or one of these error codes:
1666 nigel 63 <pre>
1667     PCRE_ERROR_NOMEMORY (-6)
1668 nigel 75 </pre>
1669 nigel 63 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1670     memory failed for <b>pcre_get_substring()</b>.
1671     <pre>
1672     PCRE_ERROR_NOSUBSTRING (-7)
1673 nigel 75 </pre>
1674 nigel 63 There is no substring whose number is <i>stringnumber</i>.
1675     </P>
1676     <P>
1677     The <b>pcre_get_substring_list()</b> function extracts all available substrings
1678     and builds a list of pointers to them. All this is done in a single block of
1679 nigel 75 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1680 nigel 63 is returned via <i>listptr</i>, which is also the start of the list of string
1681     pointers. The end of the list is marked by a NULL pointer. The yield of the
1682 nigel 93 function is zero if all went well, or the error code
1683 nigel 63 <pre>
1684     PCRE_ERROR_NOMEMORY (-6)
1685 nigel 75 </pre>
1686 nigel 63 if the attempt to get the memory block failed.
1687     </P>
1688     <P>
1689     When any of these functions encounter a substring that is unset, which can
1690     happen when capturing subpattern number <i>n+1</i> matches some part of the
1691     subject, but subpattern <i>n</i> has not been used at all, they return an empty
1692     string. This can be distinguished from a genuine zero-length substring by
1693     inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1694     substrings.
1695     </P>
1696     <P>
1697     The two convenience functions <b>pcre_free_substring()</b> and
1698     <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1699     a previous call of <b>pcre_get_substring()</b> or
1700     <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1701     the function pointed to by <b>pcre_free</b>, which of course could be called
1702     directly from a C program. However, PCRE is used in some situations where it is
1703 nigel 91 linked via a special interface to another programming language that cannot use
1704 nigel 63 <b>pcre_free</b> directly; it is for these cases that the functions are
1705     provided.
1706     </P>
1707 nigel 91 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1708 nigel 63 <P>
1709 nigel 75 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1710     <b>const char *<i>name</i>);</b>
1711     </P>
1712     <P>
1713 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1714     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1715     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1716     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1717     </P>
1718     <P>
1719     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1720     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1721     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1722     <b>const char **<i>stringptr</i>);</b>
1723     </P>
1724     <P>
1725 nigel 75 To extract a substring by name, you first have to find associated number.
1726     For example, for this pattern
1727 nigel 63 <pre>
1728 nigel 93 (a+)b(?&#60;xxx&#62;\d+)...
1729 nigel 75 </pre>
1730 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1731     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1732     calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1733     pattern, and the second is the name. The yield of the function is the
1734 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1735     that name.
1736 nigel 63 </P>
1737     <P>
1738 nigel 75 Given the number, you can extract the substring directly, or use one of the
1739     functions described in the previous section. For convenience, there are also
1740     two functions that do the whole job.
1741 nigel 63 </P>
1742     <P>
1743 nigel 91 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1744     <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1745 nigel 75 functions that extract by number. As these are described in the previous
1746     section, they are not re-described here. There are just two differences:
1747 nigel 63 </P>
1748     <P>
1749     First, instead of a substring number, a substring name is given. Second, there
1750     is an extra argument, given at the start, which is a pointer to the compiled
1751     pattern. This is needed in order to gain access to the name-to-number
1752     translation table.
1753     </P>
1754     <P>
1755     These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1756 ph10 128 then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
1757     appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1758     the behaviour may not be what you want (see the next section).
1759 nigel 63 </P>
1760 ph10 392 <P>
1761     <b>Warning:</b> If the pattern uses the "(?|" feature to set up multiple
1762     subpatterns with the same number, you cannot use names to distinguish them,
1763     because names are not included in the compiled code. The matching process uses
1764     only numbers.
1765     </P>
1766 nigel 91 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1767 nigel 63 <P>
1768 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1769     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1770     </P>
1771     <P>
1772     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1773     are not required to be unique. Normally, patterns with duplicate names are such
1774     that in any one match, only one of the named subpatterns participates. An
1775     example is shown in the
1776     <a href="pcrepattern.html"><b>pcrepattern</b></a>
1777 ph10 208 documentation.
1778     </P>
1779     <P>
1780     When duplicates are present, <b>pcre_copy_named_substring()</b> and
1781     <b>pcre_get_named_substring()</b> return the first substring corresponding to
1782     the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1783     returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
1784     returns one of the numbers that are associated with the name, but it is not
1785     defined which it is.
1786     </P>
1787     <P>
1788 nigel 91 If you want to get full details of all captured substrings for a given name,
1789     you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1790     argument is the compiled pattern, and the second is the name. The third and
1791     fourth are pointers to variables which are updated by the function. After it
1792     has run, they point to the first and last entries in the name-to-number table
1793     for the given name. The function itself returns the length of each entry, or
1794 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1795     described above in the section entitled <i>Information about a pattern</i>.
1796     Given all the relevant entries for the name, you can extract each of their
1797     numbers, and hence the captured data, if any.
1798 nigel 91 </P>
1799     <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1800     <P>
1801 nigel 77 The traditional matching function uses a similar algorithm to Perl, which stops
1802     when it finds the first match, starting at a given point in the subject. If you
1803     want to find all possible matches, or the longest possible match, consider
1804     using the alternative matching function (see below) instead. If you cannot use
1805     the alternative function, but still need to find all possible matches, you
1806     can kludge it up by making use of the callout facility, which is described in
1807     the
1808     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1809     documentation.
1810     </P>
1811     <P>
1812     What you have to do is to insert a callout right at the end of the pattern.
1813     When your callout function is called, extract and save the current matched
1814     substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1815     other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1816     will yield PCRE_ERROR_NOMATCH.
1817     <a name="dfamatch"></a></P>
1818 nigel 91 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1819 nigel 77 <P>
1820     <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1821     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1822     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1823     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1824     </P>
1825     <P>
1826     The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1827 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1828     just once, and does not backtrack. This has different characteristics to the
1829     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1830     patterns are not supported. Nevertheless, there are times when this kind of
1831     matching can be useful. For a discussion of the two matching algorithms, see
1832     the
1833 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
1834     documentation.
1835     </P>
1836     <P>
1837     The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1838     <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1839     different way, and this is described below. The other common arguments are used
1840     in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1841     here.
1842     </P>
1843     <P>
1844     The two additional arguments provide workspace for the function. The workspace
1845     vector should contain at least 20 elements. It is used for keeping track of
1846     multiple paths through the pattern tree. More workspace will be needed for
1847 nigel 91 patterns and subjects where there are a lot of potential matches.
1848 nigel 77 </P>
1849     <P>
1850 nigel 87 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1851 nigel 77 <pre>
1852     int rc;
1853     int ovector[10];
1854     int wspace[20];
1855 nigel 87 rc = pcre_dfa_exec(
1856 nigel 77 re, /* result of pcre_compile() */
1857     NULL, /* we didn't study the pattern */
1858     "some string", /* the subject string */
1859     11, /* the length of the subject string */
1860     0, /* start at offset 0 in the subject */
1861     0, /* default options */
1862     ovector, /* vector of integers for substring information */
1863     10, /* number of elements (NOT size in bytes) */
1864     wspace, /* working space vector */
1865     20); /* number of elements (NOT size in bytes) */
1866     </PRE>
1867     </P>
1868     <br><b>
1869     Option bits for <b>pcre_dfa_exec()</b>
1870     </b><br>
1871     <P>
1872     The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1873 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1874     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1875 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1876     the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1877     <pre>
1878     PCRE_PARTIAL
1879     </pre>
1880     This has the same general effect as it does for <b>pcre_exec()</b>, but the
1881     details are slightly different. When PCRE_PARTIAL is set for
1882     <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1883     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1884     complete matches, but there is still at least one matching possibility. The
1885     portion of the string that provided the partial match is set as the first
1886     matching string.
1887     <pre>
1888     PCRE_DFA_SHORTEST
1889     </pre>
1890     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1891 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1892     works, this is necessarily the shortest possible match at the first possible
1893     matching point in the subject string.
1894 nigel 77 <pre>
1895     PCRE_DFA_RESTART
1896     </pre>
1897     When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1898     a partial match, it is possible to call it again, with additional subject
1899     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1900     option requests this action; when it is set, the <i>workspace</i> and
1901     <i>wscount</i> options must reference the same vector as before because data
1902     about the match so far is left in them after a partial match. There is more
1903     discussion of this facility in the
1904     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1905     documentation.
1906     </P>
1907     <br><b>
1908     Successful returns from <b>pcre_dfa_exec()</b>
1909     </b><br>
1910     <P>
1911     When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1912     substring in the subject. Note, however, that all the matches from one run of
1913     the function start at the same point in the subject. The shorter matches are
1914     all initial substrings of the longer matches. For example, if the pattern
1915     <pre>
1916     &#60;.*&#62;
1917     </pre>
1918     is matched against the string
1919     <pre>
1920     This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1921     </pre>
1922     the three matched strings are
1923     <pre>
1924     &#60;something&#62;
1925     &#60;something&#62; &#60;something else&#62;
1926     &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1927     </pre>
1928     On success, the yield of the function is a number greater than zero, which is
1929     the number of matched substrings. The substrings themselves are returned in
1930     <i>ovector</i>. Each string uses two elements; the first is the offset to the
1931 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1932     the same start offset. (Space could have been saved by giving this only once,
1933     but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1934     returns data, even though the meaning of the strings is different.)
1935 nigel 77 </P>
1936     <P>
1937     The strings are returned in reverse order of length; that is, the longest
1938     matching string is given first. If there were too many matches to fit into
1939     <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1940     the longest matches.
1941     </P>
1942     <br><b>
1943     Error returns from <b>pcre_dfa_exec()</b>
1944     </b><br>
1945     <P>
1946     The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1947     Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1948     described
1949     <a href="#errorlist">above.</a>
1950     There are in addition the following errors that are specific to
1951     <b>pcre_dfa_exec()</b>:
1952     <pre>
1953     PCRE_ERROR_DFA_UITEM (-16)
1954     </pre>
1955     This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1956     that it does not support, for instance, the use of \C or a back reference.
1957     <pre>
1958     PCRE_ERROR_DFA_UCOND (-17)
1959     </pre>
1960 nigel 93 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1961     uses a back reference for the condition, or a test for recursion in a specific
1962     group. These are not supported.
1963 nigel 77 <pre>
1964     PCRE_ERROR_DFA_UMLIMIT (-18)
1965     </pre>
1966     This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1967     block that contains a setting of the <i>match_limit</i> field. This is not
1968     supported (it is meaningless).
1969     <pre>
1970     PCRE_ERROR_DFA_WSSIZE (-19)
1971     </pre>
1972     This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1973     <i>workspace</i> vector.
1974     <pre>
1975     PCRE_ERROR_DFA_RECURSE (-20)
1976     </pre>
1977     When a recursive subpattern is processed, the matching function calls itself
1978     recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1979     error is given if the output vector is not large enough. This should be
1980     extremely rare, as a vector of size 1000 is used.
1981     </P>
1982 nigel 93 <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1983 nigel 77 <P>
1984 nigel 93 <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1985     <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1986     <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1987     </P>
1988 ph10 99 <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1989 nigel 93 <P>
1990 ph10 99 Philip Hazel
1991 nigel 63 <br>
1992 ph10 99 University Computing Service
1993     <br>
1994     Cambridge CB2 3QH, England.
1995     <br>
1996     </P>
1997     <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1998     <P>
1999 ph10 416 Last updated: 11 April 2009
2000 ph10 99 <br>
2001 ph10 392 Copyright &copy; 1997-2009 University of Cambridge.
2002 ph10 99 <br>
2003 nigel 75 <p>
2004     Return to the <a href="index.html">PCRE index page</a>.
2005     </p>

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12