/[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 211 - (hide annotations) (download) (as text)
Thu Aug 9 09:52:43 2007 UTC (7 years, 1 month ago) by ph10
File MIME type: text/html
File size: 85601 byte(s)
Update UTF-8 validity check and documentation.

1 nigel 63 <html>
2     <head>
3     <title>pcreapi specification</title>
4     </head>
5     <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 nigel 75 <h1>pcreapi man page</h1>
7     <p>
8     Return to the <a href="index.html">PCRE index page</a>.
9     </p>
10 ph10 111 <p>
11 nigel 75 This page is part of the PCRE HTML documentation. It was generated automatically
12     from the original man page. If there is any nonsense in it, please consult the
13     man page, in case the conversion went wrong.
14 ph10 111 <br>
15 nigel 63 <ul>
16 nigel 75 <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17     <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18 nigel 91 <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19     <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
20     <li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
21     <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22     <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23     <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24     <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25     <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26     <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27     <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28     <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
29     <li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
30     <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
31     <li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
32     <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33     <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
34     <li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
35 nigel 93 <li><a name="TOC20" href="#SEC20">SEE ALSO</a>
36 ph10 99 <li><a name="TOC21" href="#SEC21">AUTHOR</a>
37     <li><a name="TOC22" href="#SEC22">REVISION</a>
38 nigel 63 </ul>
39 nigel 75 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
40 nigel 63 <P>
41     <b>#include &#60;pcre.h&#62;</b>
42     </P>
43     <P>
44     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
45     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
46     <b>const unsigned char *<i>tableptr</i>);</b>
47     </P>
48     <P>
49 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
50     <b>int *<i>errorcodeptr</i>,</b>
51     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
52     <b>const unsigned char *<i>tableptr</i>);</b>
53     </P>
54     <P>
55 nigel 63 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
56     <b>const char **<i>errptr</i>);</b>
57     </P>
58     <P>
59     <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
60     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
61     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
62     </P>
63     <P>
64 nigel 77 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
65     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
66     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
67     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
68     </P>
69     <P>
70 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
71     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
72     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
73     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
74     </P>
75     <P>
76     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
77     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
78     <b>int <i>buffersize</i>);</b>
79     </P>
80     <P>
81     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
82     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
83     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
84     <b>const char **<i>stringptr</i>);</b>
85     </P>
86     <P>
87     <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
88     <b>const char *<i>name</i>);</b>
89     </P>
90     <P>
91 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
92     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
93     </P>
94     <P>
95 nigel 63 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
96     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
97     <b>const char **<i>stringptr</i>);</b>
98     </P>
99     <P>
100     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
101     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
102     </P>
103     <P>
104     <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
105     </P>
106     <P>
107     <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
108     </P>
109     <P>
110     <b>const unsigned char *pcre_maketables(void);</b>
111     </P>
112     <P>
113     <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
114     <b>int <i>what</i>, void *<i>where</i>);</b>
115     </P>
116     <P>
117     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
118     <b>*<i>firstcharptr</i>);</b>
119     </P>
120     <P>
121 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
122     </P>
123     <P>
124 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
125     </P>
126     <P>
127     <b>char *pcre_version(void);</b>
128     </P>
129     <P>
130     <b>void *(*pcre_malloc)(size_t);</b>
131     </P>
132     <P>
133     <b>void (*pcre_free)(void *);</b>
134     </P>
135     <P>
136 nigel 73 <b>void *(*pcre_stack_malloc)(size_t);</b>
137     </P>
138     <P>
139     <b>void (*pcre_stack_free)(void *);</b>
140     </P>
141     <P>
142 nigel 63 <b>int (*pcre_callout)(pcre_callout_block *);</b>
143     </P>
144 nigel 75 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
145 nigel 63 <P>
146 nigel 93 PCRE has its own native API, which is described in this document. There are
147     also some wrapper functions that correspond to the POSIX regular expression
148 nigel 77 API. These are described in the
149 nigel 75 <a href="pcreposix.html"><b>pcreposix</b></a>
150 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
151     wrapper is distributed with PCRE. It is documented in the
152     <a href="pcrecpp.html"><b>pcrecpp</b></a>
153     page.
154 nigel 63 </P>
155     <P>
156 nigel 77 The native API C function prototypes are defined in the header file
157     <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
158     It can normally be accessed by adding <b>-lpcre</b> to the command for linking
159     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
160     and PCRE_MINOR to contain the major and minor release numbers for the library.
161 nigel 75 Applications can use these to include support for different releases of PCRE.
162 nigel 63 </P>
163     <P>
164 nigel 77 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
165     and <b>pcre_exec()</b> are used for compiling and matching regular expressions
166     in a Perl-compatible manner. A sample program that demonstrates the simplest
167     way of using them is provided in the file called <i>pcredemo.c</i> in the source
168     distribution. The
169 nigel 75 <a href="pcresample.html"><b>pcresample</b></a>
170     documentation describes how to run it.
171 nigel 63 </P>
172     <P>
173 nigel 77 A second matching function, <b>pcre_dfa_exec()</b>, which is not
174     Perl-compatible, is also provided. This uses a different algorithm for the
175 nigel 91 matching. The alternative algorithm finds all possible matches (at a given
176 nigel 93 point in the subject), and scans the subject just once. However, this algorithm
177     does not return captured substrings. A description of the two matching
178     algorithms and their advantages and disadvantages is given in the
179 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
180     documentation.
181     </P>
182     <P>
183 nigel 75 In addition to the main compiling and matching functions, there are convenience
184 nigel 77 functions for extracting captured substrings from a subject string that is
185     matched by <b>pcre_exec()</b>. They are:
186 nigel 63 <pre>
187     <b>pcre_copy_substring()</b>
188     <b>pcre_copy_named_substring()</b>
189     <b>pcre_get_substring()</b>
190     <b>pcre_get_named_substring()</b>
191     <b>pcre_get_substring_list()</b>
192 nigel 75 <b>pcre_get_stringnumber()</b>
193 nigel 91 <b>pcre_get_stringtable_entries()</b>
194 nigel 75 </pre>
195 nigel 63 <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
196     provided, to free the memory used for extracted strings.
197     </P>
198     <P>
199 nigel 75 The function <b>pcre_maketables()</b> is used to build a set of character tables
200 nigel 77 in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
201     or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
202     specialist use. Most commonly, no special tables are passed, in which case
203     internal tables that are generated when PCRE is built are used.
204 nigel 63 </P>
205     <P>
206     The function <b>pcre_fullinfo()</b> is used to find out information about a
207 nigel 75 compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
208 nigel 63 some of the available information, but is retained for backwards compatibility.
209     The function <b>pcre_version()</b> returns a pointer to a string containing the
210     version of PCRE and its date of release.
211     </P>
212     <P>
213 nigel 77 The function <b>pcre_refcount()</b> maintains a reference count in a data block
214     containing a compiled pattern. This is provided for the benefit of
215     object-oriented applications.
216     </P>
217     <P>
218 nigel 63 The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
219 nigel 75 the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
220 nigel 63 respectively. PCRE calls the memory management functions via these variables,
221     so a calling program can replace them if it wishes to intercept the calls. This
222     should be done before calling any PCRE functions.
223     </P>
224     <P>
225 nigel 73 The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
226     indirections to memory management functions. These special functions are used
227     only when PCRE is compiled to use the heap for remembering data, instead of
228 nigel 91 recursive function calls, when running the <b>pcre_exec()</b> function. See the
229     <a href="pcrebuild.html"><b>pcrebuild</b></a>
230     documentation for details of how to do this. It is a non-standard way of
231     building PCRE, for use in environments that have limited stacks. Because of the
232     greater use of memory management, it runs more slowly. Separate functions are
233     provided so that special-purpose external code can be used for this case. When
234     used, these functions are always called in a stack-like manner (last obtained,
235     first freed), and always for memory blocks of the same size. There is a
236     discussion about PCRE's stack usage in the
237     <a href="pcrestack.html"><b>pcrestack</b></a>
238     documentation.
239 nigel 73 </P>
240     <P>
241 nigel 63 The global variable <b>pcre_callout</b> initially contains NULL. It can be set
242     by the caller to a "callout" function, which PCRE will then call at specified
243 nigel 75 points during a matching operation. Details are given in the
244     <a href="pcrecallout.html"><b>pcrecallout</b></a>
245 nigel 63 documentation.
246     </P>
247 nigel 91 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248 nigel 63 <P>
249 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 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
266 nigel 93 pair of characters that indicate a line break". The choice of newline
267     convention affects the handling of the dot, circumflex, and dollar
268     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
269     recognized line ending sequence, the match position advancement for a
270     non-anchored pattern. The choice of newline convention does not affect the
271     interpretation of the \n or \r escape sequences.
272 nigel 91 </P>
273     <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
274     <P>
275 nigel 63 The PCRE functions can be used in multi-threading applications, with the
276 nigel 73 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
277     <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
278     callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
279 nigel 63 </P>
280     <P>
281     The compiled form of a regular expression is not altered during matching, so
282     the same compiled pattern can safely be used by several threads at once.
283     </P>
284 nigel 91 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
285 nigel 63 <P>
286 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
287     time, possibly by a different program, and even on a host other than the one on
288     which it was compiled. Details are given in the
289     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
290 ph10 155 documentation. However, compiling a regular expression with one version of PCRE
291     for use with a different version is not guaranteed to work and may cause
292     crashes.
293 nigel 75 </P>
294 nigel 91 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
295 nigel 75 <P>
296 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
297     </P>
298     <P>
299     The function <b>pcre_config()</b> makes it possible for a PCRE client to
300     discover which optional features have been compiled into the PCRE library. The
301     <a href="pcrebuild.html"><b>pcrebuild</b></a>
302     documentation has more details about these optional features.
303     </P>
304     <P>
305     The first argument for <b>pcre_config()</b> is an integer, specifying which
306     information is required; the second argument is a pointer to a variable into
307     which the information is placed. The following information is available:
308     <pre>
309     PCRE_CONFIG_UTF8
310 nigel 75 </pre>
311 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
312     otherwise it is set to zero.
313     <pre>
314 nigel 75 PCRE_CONFIG_UNICODE_PROPERTIES
315     </pre>
316     The output is an integer that is set to one if support for Unicode character
317     properties is available; otherwise it is set to zero.
318     <pre>
319 nigel 63 PCRE_CONFIG_NEWLINE
320 nigel 75 </pre>
321 nigel 91 The output is an integer whose value specifies the default character sequence
322 nigel 93 that is recognized as meaning "newline". The four values that are supported
323 ph10 150 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
324     default should normally be the standard sequence for your operating system.
325 nigel 63 <pre>
326     PCRE_CONFIG_LINK_SIZE
327 nigel 75 </pre>
328 nigel 63 The output is an integer that contains the number of bytes used for internal
329     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
330     allow larger regular expressions to be compiled, at the expense of slower
331     matching. The default value of 2 is sufficient for all but the most massive
332     patterns, since it allows the compiled pattern to be up to 64K in size.
333     <pre>
334     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
335 nigel 75 </pre>
336 nigel 63 The output is an integer that contains the threshold above which the POSIX
337     interface uses <b>malloc()</b> for output vectors. Further details are given in
338 nigel 75 the
339     <a href="pcreposix.html"><b>pcreposix</b></a>
340     documentation.
341 nigel 63 <pre>
342     PCRE_CONFIG_MATCH_LIMIT
343 nigel 75 </pre>
344 nigel 63 The output is an integer that gives the default limit for the number of
345     internal matching function calls in a <b>pcre_exec()</b> execution. Further
346     details are given with <b>pcre_exec()</b> below.
347 nigel 73 <pre>
348 nigel 87 PCRE_CONFIG_MATCH_LIMIT_RECURSION
349     </pre>
350     The output is an integer that gives the default limit for the depth of
351     recursion when calling the internal matching function in a <b>pcre_exec()</b>
352     execution. Further details are given with <b>pcre_exec()</b> below.
353     <pre>
354 nigel 73 PCRE_CONFIG_STACKRECURSE
355 nigel 75 </pre>
356 nigel 77 The output is an integer that is set to one if internal recursion when running
357     <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
358     to remember their state. This is the usual way that PCRE is compiled. The
359     output is zero if PCRE was compiled to use blocks of data on the heap instead
360     of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
361     <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
362     avoiding the use of the stack.
363 nigel 73 </P>
364 nigel 91 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
365 nigel 63 <P>
366     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
367     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
368     <b>const unsigned char *<i>tableptr</i>);</b>
369 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
370     <b>int *<i>errorcodeptr</i>,</b>
371     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
372     <b>const unsigned char *<i>tableptr</i>);</b>
373 nigel 63 </P>
374     <P>
375 nigel 77 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
376     called to compile a pattern into an internal form. The only difference between
377     the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
378     <i>errorcodeptr</i>, via which a numerical error code can be returned.
379 nigel 63 </P>
380     <P>
381 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
382     <i>pattern</i> argument. A pointer to a single block of memory that is obtained
383     via <b>pcre_malloc</b> is returned. This contains the compiled code and related
384     data. The <b>pcre</b> type is defined for the returned block; this is a typedef
385     for a structure whose contents are not externally defined. It is up to the
386 nigel 91 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
387 nigel 77 </P>
388     <P>
389 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
390     depend on memory location, the complete <b>pcre</b> data block is not
391 nigel 75 fully relocatable, because it may contain a copy of the <i>tableptr</i>
392     argument, which is an address (see below).
393 nigel 63 </P>
394     <P>
395 nigel 93 The <i>options</i> argument contains various bit settings that affect the
396 nigel 75 compilation. It should be zero if no options are required. The available
397     options are described below. Some of them, in particular, those that are
398     compatible with Perl, can also be set and unset from within the pattern (see
399     the detailed description in the
400     <a href="pcrepattern.html"><b>pcrepattern</b></a>
401     documentation). For these options, the contents of the <i>options</i> argument
402     specifies their initial settings at the start of compilation and execution. The
403 nigel 91 PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
404     matching as well as at compile time.
405 nigel 63 </P>
406     <P>
407     If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
408     Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
409     NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
410 nigel 87 error message. This is a static string that is part of the library. You must
411     not try to free it. The offset from the start of the pattern to the character
412     where the error was discovered is placed in the variable pointed to by
413 nigel 63 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
414     </P>
415     <P>
416 nigel 77 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
417     <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
418     returned via this argument in the event of an error. This is in addition to the
419     textual error message. Error codes and messages are listed below.
420     </P>
421     <P>
422 nigel 63 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
423 nigel 75 character tables that are built when PCRE is compiled, using the default C
424     locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
425     call to <b>pcre_maketables()</b>. This value is stored with the compiled
426     pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
427     passed to it. For more discussion, see the section on locale support below.
428 nigel 63 </P>
429     <P>
430     This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
431     <pre>
432     pcre *re;
433     const char *error;
434     int erroffset;
435     re = pcre_compile(
436     "^A.*Z", /* the pattern */
437     0, /* default options */
438     &error, /* for error message */
439     &erroffset, /* for error offset */
440     NULL); /* use default character tables */
441 nigel 75 </pre>
442     The following names for option bits are defined in the <b>pcre.h</b> header
443     file:
444 nigel 63 <pre>
445     PCRE_ANCHORED
446 nigel 75 </pre>
447 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
448 nigel 75 constrained to match only at the first matching point in the string that is
449 nigel 63 being searched (the "subject string"). This effect can also be achieved by
450     appropriate constructs in the pattern itself, which is the only way to do it in
451     Perl.
452     <pre>
453 nigel 75 PCRE_AUTO_CALLOUT
454     </pre>
455     If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
456     all with number 255, before each pattern item. For discussion of the callout
457     facility, see the
458     <a href="pcrecallout.html"><b>pcrecallout</b></a>
459     documentation.
460     <pre>
461 nigel 63 PCRE_CASELESS
462 nigel 75 </pre>
463 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
464     letters. It is equivalent to Perl's /i option, and it can be changed within a
465 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
466     concept of case for characters whose values are less than 128, so caseless
467     matching is always possible. For characters with higher values, the concept of
468     case is supported if PCRE is compiled with Unicode property support, but not
469     otherwise. If you want to use caseless matching for characters 128 and above,
470     you must ensure that PCRE is compiled with Unicode property support as well as
471     with UTF-8 support.
472 nigel 63 <pre>
473     PCRE_DOLLAR_ENDONLY
474 nigel 75 </pre>
475 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
476     end of the subject string. Without this option, a dollar also matches
477 nigel 91 immediately before a newline at the end of the string (but not before any other
478     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
479     There is no equivalent to this option in Perl, and no way to set it within a
480     pattern.
481 nigel 63 <pre>
482     PCRE_DOTALL
483 nigel 75 </pre>
484 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
485 nigel 91 including those that indicate newline. Without it, a dot does not match when
486     the current position is at a newline. This option is equivalent to Perl's /s
487     option, and it can be changed within a pattern by a (?s) option setting. A
488 nigel 93 negative class such as [^a] always matches newline characters, independent of
489     the setting of this option.
490 nigel 63 <pre>
491 nigel 91 PCRE_DUPNAMES
492     </pre>
493     If this bit is set, names used to identify capturing subpatterns need not be
494     unique. This can be helpful for certain types of pattern when it is known that
495     only one instance of the named subpattern can ever be matched. There are more
496     details of named subpatterns below; see also the
497     <a href="pcrepattern.html"><b>pcrepattern</b></a>
498     documentation.
499     <pre>
500 nigel 63 PCRE_EXTENDED
501 nigel 75 </pre>
502 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
503     ignored except when escaped or inside a character class. Whitespace does not
504     include the VT character (code 11). In addition, characters between an
505 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
506     ignored. This is equivalent to Perl's /x option, and it can be changed within a
507     pattern by a (?x) option setting.
508 nigel 63 </P>
509     <P>
510     This option makes it possible to include comments inside complicated patterns.
511     Note, however, that this applies only to data characters. Whitespace characters
512     may never appear within special character sequences in a pattern, for example
513     within the sequence (?( which introduces a conditional subpattern.
514     <pre>
515     PCRE_EXTRA
516 nigel 75 </pre>
517 nigel 63 This option was invented in order to turn on additional functionality of PCRE
518     that is incompatible with Perl, but it is currently of very little use. When
519     set, any backslash in a pattern that is followed by a letter that has no
520     special meaning causes an error, thus reserving these combinations for future
521     expansion. By default, as in Perl, a backslash followed by a letter with no
522 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
523     give a warning for this.) There are at present no other features controlled by
524     this option. It can also be set by a (?X) option setting within a pattern.
525 nigel 63 <pre>
526 nigel 77 PCRE_FIRSTLINE
527     </pre>
528     If this option is set, an unanchored pattern is required to match before or at
529 nigel 91 the first newline in the subject string, though the matched text may continue
530     over the newline.
531 nigel 77 <pre>
532 nigel 63 PCRE_MULTILINE
533 nigel 75 </pre>
534     By default, PCRE treats the subject string as consisting of a single line of
535     characters (even if it actually contains newlines). The "start of line"
536 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
537     line" metacharacter ($) matches only at the end of the string, or before a
538     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
539     Perl.
540     </P>
541     <P>
542     When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
543 nigel 91 match immediately following or immediately before internal newlines in the
544     subject string, respectively, as well as at the very start and end. This is
545     equivalent to Perl's /m option, and it can be changed within a pattern by a
546     (?m) option setting. If there are no newlines in a subject string, or no
547 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
548     <pre>
549 nigel 91 PCRE_NEWLINE_CR
550     PCRE_NEWLINE_LF
551     PCRE_NEWLINE_CRLF
552 ph10 150 PCRE_NEWLINE_ANYCRLF
553 nigel 93 PCRE_NEWLINE_ANY
554 nigel 91 </pre>
555     These options override the default newline definition that was chosen when PCRE
556     was built. Setting the first or the second specifies that a newline is
557 nigel 93 indicated by a single character (CR or LF, respectively). Setting
558     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
559 ph10 150 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
560     preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
561     that any Unicode newline sequence should be recognized. The Unicode newline
562     sequences are the three just mentioned, plus the single characters VT (vertical
563     tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
564     separator, U+2028), and PS (paragraph separator, U+2029). The last two are
565     recognized only in UTF-8 mode.
566 nigel 91 </P>
567     <P>
568 nigel 93 The newline setting in the options word uses three bits that are treated
569 ph10 150 as a number, giving eight possibilities. Currently only six are used (default
570     plus the five values above). This means that if you set more than one newline
571 nigel 93 option, the combination may or may not be sensible. For example,
572     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
573 ph10 150 other combinations may yield unused numbers and cause an error.
574 nigel 93 </P>
575     <P>
576     The only time that a line break is specially recognized when compiling a
577     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
578     class is encountered. This indicates a comment that lasts until after the next
579     line break sequence. In other circumstances, line break sequences are treated
580     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
581     as whitespace characters and are therefore ignored.
582     </P>
583     <P>
584     The newline option that is set at compile time becomes the default that is used
585     for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
586 nigel 91 <pre>
587 nigel 63 PCRE_NO_AUTO_CAPTURE
588 nigel 75 </pre>
589 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
590     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
591     were followed by ?: but named parentheses can still be used for capturing (and
592     they acquire numbers in the usual way). There is no equivalent of this option
593     in Perl.
594     <pre>
595     PCRE_UNGREEDY
596 nigel 75 </pre>
597 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
598     greedy by default, but become greedy if followed by "?". It is not compatible
599     with Perl. It can also be set by a (?U) option setting within the pattern.
600     <pre>
601     PCRE_UTF8
602 nigel 75 </pre>
603 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
604     of UTF-8 characters instead of single-byte character strings. However, it is
605 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
606 nigel 63 of this option provokes an error. Details of how this option changes the
607     behaviour of PCRE are given in the
608     <a href="pcre.html#utf8support">section on UTF-8 support</a>
609     in the main
610     <a href="pcre.html"><b>pcre</b></a>
611     page.
612 nigel 71 <pre>
613     PCRE_NO_UTF8_CHECK
614 nigel 75 </pre>
615 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
616 ph10 211 automatically checked. There is a discussion about the
617     <a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
618     in the main
619     <a href="pcre.html"><b>pcre</b></a>
620     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
621     returns an error. If you already know that your pattern is valid, and you want
622     to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
623     option. When it is set, the effect of passing an invalid UTF-8 string as a
624     pattern is undefined. It may cause your program to crash. Note that this option
625     can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
626     the UTF-8 validity checking of subject strings.
627 nigel 71 </P>
628 nigel 91 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
629 nigel 63 <P>
630 nigel 77 The following table lists the error codes than may be returned by
631     <b>pcre_compile2()</b>, along with the error messages that may be returned by
632 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
633     out of use. To avoid confusion, they have not been re-used.
634 nigel 77 <pre>
635     0 no error
636     1 \ at end of pattern
637     2 \c at end of pattern
638     3 unrecognized character follows \
639     4 numbers out of order in {} quantifier
640     5 number too big in {} quantifier
641     6 missing terminating ] for character class
642     7 invalid escape sequence in character class
643     8 range out of order in character class
644     9 nothing to repeat
645 nigel 93 10 [this code is not in use]
646 nigel 77 11 internal error: unexpected repeat
647     12 unrecognized character after (?
648     13 POSIX named classes are supported only within a class
649     14 missing )
650     15 reference to non-existent subpattern
651     16 erroffset passed as NULL
652     17 unknown option bit(s) set
653     18 missing ) after comment
654 nigel 93 19 [this code is not in use]
655 nigel 77 20 regular expression too large
656     21 failed to get memory
657     22 unmatched parentheses
658     23 internal error: code overflow
659     24 unrecognized character after (?&#60;
660     25 lookbehind assertion is not fixed length
661 nigel 91 26 malformed number or name after (?(
662 nigel 77 27 conditional group contains more than two branches
663     28 assertion expected after (?(
664 ph10 182 29 (?R or (?[+-]digits must be followed by )
665 nigel 77 30 unknown POSIX class name
666     31 POSIX collating elements are not supported
667     32 this version of PCRE is not compiled with PCRE_UTF8 support
668 nigel 93 33 [this code is not in use]
669 nigel 77 34 character value in \x{...} sequence is too large
670     35 invalid condition (?(0)
671     36 \C not allowed in lookbehind assertion
672     37 PCRE does not support \L, \l, \N, \U, or \u
673     38 number after (?C is &#62; 255
674     39 closing ) for (?C expected
675     40 recursive call could loop indefinitely
676     41 unrecognized character after (?P
677 nigel 93 42 syntax error in subpattern name (missing terminator)
678 nigel 91 43 two named subpatterns have the same name
679 nigel 77 44 invalid UTF-8 string
680     45 support for \P, \p, and \X has not been compiled
681     46 malformed \P or \p sequence
682     47 unknown property name after \P or \p
683 nigel 91 48 subpattern name is too long (maximum 32 characters)
684     49 too many named subpatterns (maximum 10,000)
685 ph10 202 50 [this code is not in use]
686 nigel 91 51 octal value is greater than \377 (not in UTF-8 mode)
687 nigel 93 52 internal error: overran compiling workspace
688     53 internal error: previously-checked referenced subpattern not found
689     54 DEFINE group contains more than one branch
690     55 repeating a DEFINE group is not allowed
691     56 inconsistent NEWLINE options"
692 ph10 182 57 \g is not followed by a braced name or an optionally braced
693     non-zero number
694     58 (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
695 nigel 77 </PRE>
696     </P>
697 nigel 91 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
698 nigel 77 <P>
699     <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
700 nigel 63 <b>const char **<i>errptr</i>);</b>
701     </P>
702     <P>
703 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
704     more time analyzing it in order to speed up the time taken for matching. The
705 nigel 63 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
706 nigel 75 argument. If studying the pattern produces additional information that will
707     help speed up matching, <b>pcre_study()</b> returns a pointer to a
708     <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
709     results of the study.
710 nigel 63 </P>
711     <P>
712 nigel 75 The returned value from <b>pcre_study()</b> can be passed directly to
713     <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
714 nigel 63 fields that can be set by the caller before the block is passed; these are
715 nigel 75 described
716     <a href="#extradata">below</a>
717     in the section on matching a pattern.
718 nigel 63 </P>
719     <P>
720 nigel 77 If studying the pattern does not produce any additional information
721 nigel 75 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
722     wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
723     own <b>pcre_extra</b> block.
724 nigel 63 </P>
725     <P>
726 nigel 75 The second argument of <b>pcre_study()</b> contains option bits. At present, no
727     options are defined, and this argument should always be zero.
728     </P>
729     <P>
730 nigel 63 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
731     studying succeeds (even if no data is returned), the variable it points to is
732 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
733     static string that is part of the library. You must not try to free it. You
734     should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
735     sure that it has run successfully.
736 nigel 63 </P>
737     <P>
738     This is a typical call to <b>pcre_study</b>():
739     <pre>
740     pcre_extra *pe;
741     pe = pcre_study(
742     re, /* result of pcre_compile() */
743     0, /* no options exist */
744     &error); /* set to NULL or points to a message */
745 nigel 75 </pre>
746 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
747     not have a single fixed starting character. A bitmap of possible starting
748 nigel 75 bytes is created.
749     <a name="localesupport"></a></P>
750 nigel 91 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
751 nigel 63 <P>
752 ph10 142 PCRE handles caseless matching, and determines whether characters are letters,
753 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
754 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
755 nigel 75 less than 128. Higher-valued codes never match escapes such as \w or \d, but
756     can be tested with \p if PCRE is built with Unicode character property
757 ph10 142 support. The use of locales with Unicode is discouraged. If you are handling
758     characters with codes greater than 128, you should either use UTF-8 and
759     Unicode, or use locales, but not try to mix the two.
760 nigel 63 </P>
761     <P>
762 ph10 142 PCRE contains an internal set of tables that are used when the final argument
763     of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
764     Normally, the internal tables recognize only ASCII characters. However, when
765     PCRE is built, it is possible to cause the internal tables to be rebuilt in the
766     default "C" locale of the local system, which may cause them to be different.
767 nigel 63 </P>
768     <P>
769 ph10 142 The internal tables can always be overridden by tables supplied by the
770     application that calls PCRE. These may be created in a different locale from
771     the default. As more and more applications change to using Unicode, the need
772     for this locale support is expected to die away.
773     </P>
774     <P>
775 nigel 75 External tables are built by calling the <b>pcre_maketables()</b> function,
776     which has no arguments, in the relevant locale. The result can then be passed
777     to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
778     example, to build and use tables that are appropriate for the French locale
779     (where accented characters with values greater than 128 are treated as letters),
780     the following code could be used:
781 nigel 63 <pre>
782 nigel 75 setlocale(LC_CTYPE, "fr_FR");
783 nigel 63 tables = pcre_maketables();
784     re = pcre_compile(..., tables);
785 nigel 75 </pre>
786 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
787     are using Windows, the name for the French locale is "french".
788     </P>
789     <P>
790 nigel 75 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
791     obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
792     that the memory containing the tables remains available for as long as it is
793     needed.
794 nigel 63 </P>
795     <P>
796 nigel 75 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
797 nigel 63 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
798 nigel 75 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
799     pattern, compilation, studying and matching all happen in the same locale, but
800     different patterns can be compiled in different locales.
801 nigel 63 </P>
802     <P>
803 nigel 75 It is possible to pass a table pointer or NULL (indicating the use of the
804     internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
805     this facility could be used to match a pattern in a different locale from the
806     one in which it was compiled. Passing table pointers at run time is discussed
807     below in the section on matching a pattern.
808     </P>
809 nigel 91 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
810 nigel 75 <P>
811 nigel 63 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
812     <b>int <i>what</i>, void *<i>where</i>);</b>
813     </P>
814     <P>
815     The <b>pcre_fullinfo()</b> function returns information about a compiled
816     pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
817     nevertheless retained for backwards compability (and is documented below).
818     </P>
819     <P>
820     The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
821     pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
822     the pattern was not studied. The third argument specifies which piece of
823     information is required, and the fourth argument is a pointer to a variable
824     to receive the data. The yield of the function is zero for success, or one of
825     the following negative numbers:
826     <pre>
827     PCRE_ERROR_NULL the argument <i>code</i> was NULL
828     the argument <i>where</i> was NULL
829     PCRE_ERROR_BADMAGIC the "magic number" was not found
830     PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
831 nigel 75 </pre>
832     The "magic number" is placed at the start of each compiled pattern as an simple
833     check against passing an arbitrary memory pointer. Here is a typical call of
834     <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
835 nigel 63 <pre>
836     int rc;
837 nigel 91 size_t length;
838 nigel 63 rc = pcre_fullinfo(
839     re, /* result of pcre_compile() */
840     pe, /* result of pcre_study(), or NULL */
841     PCRE_INFO_SIZE, /* what is required */
842     &length); /* where to put the data */
843 nigel 75 </pre>
844 nigel 63 The possible values for the third argument are defined in <b>pcre.h</b>, and are
845     as follows:
846     <pre>
847     PCRE_INFO_BACKREFMAX
848 nigel 75 </pre>
849 nigel 63 Return the number of the highest back reference in the pattern. The fourth
850     argument should point to an <b>int</b> variable. Zero is returned if there are
851     no back references.
852     <pre>
853     PCRE_INFO_CAPTURECOUNT
854 nigel 75 </pre>
855 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
856 nigel 75 should point to an <b>int</b> variable.
857 nigel 63 <pre>
858 nigel 77 PCRE_INFO_DEFAULT_TABLES
859 nigel 75 </pre>
860     Return a pointer to the internal default character tables within PCRE. The
861     fourth argument should point to an <b>unsigned char *</b> variable. This
862     information call is provided for internal use by the <b>pcre_study()</b>
863     function. External callers can cause PCRE to use its internal tables by passing
864     a NULL table pointer.
865     <pre>
866 nigel 63 PCRE_INFO_FIRSTBYTE
867 nigel 75 </pre>
868 nigel 63 Return information about the first byte of any matched string, for a
869 nigel 91 non-anchored pattern. The fourth argument should point to an <b>int</b>
870     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
871     still recognized for backwards compatibility.)
872 nigel 63 </P>
873     <P>
874 nigel 75 If there is a fixed first byte, for example, from a pattern such as
875 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
876 nigel 75 <br>
877     <br>
878 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
879     starts with "^", or
880 nigel 75 <br>
881     <br>
882 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
883     (if it were set, the pattern would be anchored),
884 nigel 75 <br>
885     <br>
886 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
887     subject string or after any newline within the string. Otherwise -2 is
888     returned. For anchored patterns, -2 is returned.
889     <pre>
890     PCRE_INFO_FIRSTTABLE
891 nigel 75 </pre>
892 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
893     table indicating a fixed set of bytes for the first byte in any matching
894     string, a pointer to the table is returned. Otherwise NULL is returned. The
895     fourth argument should point to an <b>unsigned char *</b> variable.
896     <pre>
897 ph10 172 PCRE_INFO_JCHANGED
898     </pre>
899     Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
900     fourth argument should point to an <b>int</b> variable. The (?J) internal option
901 ph10 182 setting changes the local PCRE_DUPNAMES option.
902 ph10 172 <pre>
903 nigel 63 PCRE_INFO_LASTLITERAL
904 nigel 75 </pre>
905 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
906     string, other than at its start, if such a byte has been recorded. The fourth
907     argument should point to an <b>int</b> variable. If there is no such byte, -1 is
908     returned. For anchored patterns, a last literal byte is recorded only if it
909     follows something of variable length. For example, for the pattern
910     /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
911     is -1.
912 nigel 63 <pre>
913     PCRE_INFO_NAMECOUNT
914     PCRE_INFO_NAMEENTRYSIZE
915     PCRE_INFO_NAMETABLE
916 nigel 75 </pre>
917 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
918     names are just an additional way of identifying the parentheses, which still
919 nigel 91 acquire numbers. Several convenience functions such as
920     <b>pcre_get_named_substring()</b> are provided for extracting captured
921     substrings by name. It is also possible to extract the data directly, by first
922     converting the name to a number in order to access the correct pointers in the
923     output vector (described with <b>pcre_exec()</b> below). To do the conversion,
924     you need to use the name-to-number map, which is described by these three
925     values.
926 nigel 63 </P>
927     <P>
928     The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
929     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
930     entry; both of these return an <b>int</b> value. The entry size depends on the
931     length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
932     entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
933     are the number of the capturing parenthesis, most significant byte first. The
934     rest of the entry is the corresponding name, zero terminated. The names are in
935 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
936     their parentheses numbers. For example, consider the following pattern (assume
937 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
938     <pre>
939 nigel 93 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
940 nigel 75 </pre>
941 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
942     in the table is eight bytes long. The table is as follows, with non-printing
943 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
944 nigel 63 <pre>
945     00 01 d a t e 00 ??
946     00 05 d a y 00 ?? ??
947     00 04 m o n t h 00
948     00 02 y e a r 00 ??
949 nigel 75 </pre>
950     When writing code to extract data from named subpatterns using the
951 nigel 91 name-to-number map, remember that the length of the entries is likely to be
952 nigel 75 different for each compiled pattern.
953 nigel 63 <pre>
954 ph10 172 PCRE_INFO_OKPARTIAL
955     </pre>
956     Return 1 if the pattern can be used for partial matching, otherwise 0. The
957     fourth argument should point to an <b>int</b> variable. The
958     <a href="pcrepartial.html"><b>pcrepartial</b></a>
959     documentation lists the restrictions that apply to patterns when partial
960     matching is used.
961     <pre>
962 nigel 63 PCRE_INFO_OPTIONS
963 nigel 75 </pre>
964 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
965     argument should point to an <b>unsigned long int</b> variable. These option bits
966     are those specified in the call to <b>pcre_compile()</b>, modified by any
967 ph10 197 top-level option settings at the start of the pattern itself. In other words,
968     they are the options that will be in force when matching starts. For example,
969     if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
970     result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
971 nigel 63 </P>
972     <P>
973     A pattern is automatically anchored by PCRE if all of its top-level
974     alternatives begin with one of the following:
975     <pre>
976     ^ unless PCRE_MULTILINE is set
977     \A always
978     \G always
979 nigel 75 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
980     </pre>
981 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
982     <b>pcre_fullinfo()</b>.
983     <pre>
984     PCRE_INFO_SIZE
985 nigel 75 </pre>
986 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
987     the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
988     place the compiled data. The fourth argument should point to a <b>size_t</b>
989     variable.
990     <pre>
991     PCRE_INFO_STUDYSIZE
992 nigel 75 </pre>
993     Return the size of the data block pointed to by the <i>study_data</i> field in
994 nigel 63 a <b>pcre_extra</b> block. That is, it is the value that was passed to
995     <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
996     created by <b>pcre_study()</b>. The fourth argument should point to a
997     <b>size_t</b> variable.
998     </P>
999 nigel 91 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
1000 nigel 63 <P>
1001     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
1002     <b>*<i>firstcharptr</i>);</b>
1003     </P>
1004     <P>
1005     The <b>pcre_info()</b> function is now obsolete because its interface is too
1006     restrictive to return all the available data about a compiled pattern. New
1007     programs should use <b>pcre_fullinfo()</b> instead. The yield of
1008     <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
1009     following negative numbers:
1010     <pre>
1011     PCRE_ERROR_NULL the argument <i>code</i> was NULL
1012     PCRE_ERROR_BADMAGIC the "magic number" was not found
1013 nigel 75 </pre>
1014 nigel 63 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
1015     pattern was compiled is placed in the integer it points to (see
1016     PCRE_INFO_OPTIONS above).
1017     </P>
1018     <P>
1019     If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
1020     it is used to pass back information about the first character of any matched
1021     string (see PCRE_INFO_FIRSTBYTE above).
1022     </P>
1023 nigel 91 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1024 nigel 63 <P>
1025 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1026     </P>
1027     <P>
1028     The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1029     data block that contains a compiled pattern. It is provided for the benefit of
1030     applications that operate in an object-oriented manner, where different parts
1031     of the application may be using the same compiled pattern, but you want to free
1032     the block when they are all done.
1033     </P>
1034     <P>
1035     When a pattern is compiled, the reference count field is initialized to zero.
1036     It is changed only by calling this function, whose action is to add the
1037     <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1038     function is the new value. However, the value of the count is constrained to
1039     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1040     it is forced to the appropriate limit value.
1041     </P>
1042     <P>
1043     Except when it is zero, the reference count is not correctly preserved if a
1044     pattern is compiled on one host and then transferred to a host whose byte-order
1045     is different. (This seems a highly unlikely scenario.)
1046     </P>
1047 nigel 91 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1048 nigel 77 <P>
1049 nigel 63 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1050     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1051     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1052     </P>
1053     <P>
1054     The function <b>pcre_exec()</b> is called to match a subject string against a
1055 nigel 75 compiled pattern, which is passed in the <i>code</i> argument. If the
1056 nigel 63 pattern has been studied, the result of the study should be passed in the
1057 nigel 77 <i>extra</i> argument. This function is the main matching facility of the
1058     library, and it operates in a Perl-like manner. For specialist use there is
1059     also an alternative matching function, which is described
1060     <a href="#dfamatch">below</a>
1061     in the section about the <b>pcre_dfa_exec()</b> function.
1062 nigel 63 </P>
1063     <P>
1064 nigel 75 In most applications, the pattern will have been compiled (and optionally
1065     studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1066     possible to save compiled patterns and study data, and then use them later
1067     in different processes, possibly even on different hosts. For a discussion
1068     about this, see the
1069     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1070     documentation.
1071 nigel 63 </P>
1072     <P>
1073 nigel 75 Here is an example of a simple call to <b>pcre_exec()</b>:
1074 nigel 63 <pre>
1075     int rc;
1076     int ovector[30];
1077     rc = pcre_exec(
1078     re, /* result of pcre_compile() */
1079     NULL, /* we didn't study the pattern */
1080     "some string", /* the subject string */
1081     11, /* the length of the subject string */
1082     0, /* start at offset 0 in the subject */
1083     0, /* default options */
1084 nigel 75 ovector, /* vector of integers for substring information */
1085 nigel 77 30); /* number of elements (NOT size in bytes) */
1086 nigel 75 <a name="extradata"></a></PRE>
1087 nigel 63 </P>
1088 nigel 75 <br><b>
1089     Extra data for <b>pcre_exec()</b>
1090     </b><br>
1091 nigel 63 <P>
1092     If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1093     data block. The <b>pcre_study()</b> function returns such a block (when it
1094     doesn't return NULL), but you can also create one for yourself, and pass
1095 nigel 87 additional information in it. The <b>pcre_extra</b> block contains the following
1096     fields (not necessarily in this order):
1097 nigel 63 <pre>
1098     unsigned long int <i>flags</i>;
1099     void *<i>study_data</i>;
1100     unsigned long int <i>match_limit</i>;
1101 nigel 87 unsigned long int <i>match_limit_recursion</i>;
1102 nigel 63 void *<i>callout_data</i>;
1103 nigel 75 const unsigned char *<i>tables</i>;
1104     </pre>
1105 nigel 63 The <i>flags</i> field is a bitmap that specifies which of the other fields
1106     are set. The flag bits are:
1107     <pre>
1108     PCRE_EXTRA_STUDY_DATA
1109     PCRE_EXTRA_MATCH_LIMIT
1110 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1111 nigel 63 PCRE_EXTRA_CALLOUT_DATA
1112 nigel 75 PCRE_EXTRA_TABLES
1113     </pre>
1114 nigel 63 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1115     <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1116 nigel 75 the appropriate flag bit. You should not set this yourself, but you may add to
1117     the block by setting the other fields and their corresponding flag bits.
1118 nigel 63 </P>
1119     <P>
1120     The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1121     vast amount of resources when running patterns that are not going to match,
1122     but which have a very large number of possibilities in their search trees. The
1123 nigel 75 classic example is the use of nested unlimited repeats.
1124 nigel 63 </P>
1125     <P>
1126 nigel 75 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1127 nigel 87 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1128     number of times this function is called during a match, which has the effect of
1129     limiting the amount of backtracking that can take place. For patterns that are
1130     not anchored, the count restarts from zero for each position in the subject
1131     string.
1132 nigel 75 </P>
1133     <P>
1134 nigel 87 The default value for the limit can be set when PCRE is built; the default
1135 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1136 nigel 87 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1137     block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1138     the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1139     PCRE_ERROR_MATCHLIMIT.
1140 nigel 63 </P>
1141     <P>
1142 nigel 87 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1143     instead of limiting the total number of times that <b>match()</b> is called, it
1144     limits the depth of recursion. The recursion depth is a smaller number than the
1145     total number of calls, because not all calls to <b>match()</b> are recursive.
1146     This limit is of use only if it is set smaller than <i>match_limit</i>.
1147     </P>
1148     <P>
1149     Limiting the recursion depth limits the amount of stack that can be used, or,
1150     when PCRE has been compiled to use memory on the heap instead of the stack, the
1151     amount of heap memory that can be used.
1152     </P>
1153     <P>
1154     The default value for <i>match_limit_recursion</i> can be set when PCRE is
1155     built; the default default is the same value as the default for
1156     <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1157     with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1158     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1159     is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1160     </P>
1161     <P>
1162 nigel 63 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1163 nigel 75 which is described in the
1164     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1165     documentation.
1166 nigel 63 </P>
1167     <P>
1168 nigel 75 The <i>tables</i> field is used to pass a character tables pointer to
1169     <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1170     pattern. A non-NULL value is stored with the compiled pattern only if custom
1171     tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1172     If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1173     internal tables to be used. This facility is helpful when re-using patterns
1174     that have been saved after compiling with an external set of tables, because
1175     the external tables might be at a different address when <b>pcre_exec()</b> is
1176     called. See the
1177     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1178     documentation for a discussion of saving compiled patterns for later use.
1179 nigel 63 </P>
1180 nigel 75 <br><b>
1181     Option bits for <b>pcre_exec()</b>
1182     </b><br>
1183 nigel 63 <P>
1184 nigel 75 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1185 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1186     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1187 nigel 63 <pre>
1188 nigel 75 PCRE_ANCHORED
1189     </pre>
1190     The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1191     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1192     to be anchored by virtue of its contents, it cannot be made unachored at
1193     matching time.
1194     <pre>
1195 nigel 91 PCRE_NEWLINE_CR
1196     PCRE_NEWLINE_LF
1197     PCRE_NEWLINE_CRLF
1198 ph10 150 PCRE_NEWLINE_ANYCRLF
1199 nigel 93 PCRE_NEWLINE_ANY
1200 nigel 91 </pre>
1201     These options override the newline definition that was chosen or defaulted when
1202 nigel 93 the pattern was compiled. For details, see the description of
1203     <b>pcre_compile()</b> above. During matching, the newline choice affects the
1204     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1205     the way the match position is advanced after a match failure for an unanchored
1206 ph10 150 pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
1207     set, and a match attempt fails when the current position is at a CRLF sequence,
1208     the match position is advanced by two characters instead of one, in other
1209     words, to after the CRLF.
1210 nigel 91 <pre>
1211 nigel 63 PCRE_NOTBOL
1212 nigel 75 </pre>
1213     This option specifies that first character of the subject string is not the
1214     beginning of a line, so the circumflex metacharacter should not match before
1215     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1216     never to match. This option affects only the behaviour of the circumflex
1217     metacharacter. It does not affect \A.
1218 nigel 63 <pre>
1219     PCRE_NOTEOL
1220 nigel 75 </pre>
1221     This option specifies that the end of the subject string is not the end of a
1222     line, so the dollar metacharacter should not match it nor (except in multiline
1223     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1224     compile time) causes dollar never to match. This option affects only the
1225     behaviour of the dollar metacharacter. It does not affect \Z or \z.
1226 nigel 63 <pre>
1227     PCRE_NOTEMPTY
1228 nigel 75 </pre>
1229 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1230     there are alternatives in the pattern, they are tried. If all the alternatives
1231     match the empty string, the entire match fails. For example, if the pattern
1232     <pre>
1233     a?b?
1234 nigel 75 </pre>
1235 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1236     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1237     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1238     </P>
1239     <P>
1240     Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1241     of a pattern match of the empty string within its <b>split()</b> function, and
1242     when using the /g modifier. It is possible to emulate Perl's behaviour after
1243     matching a null string by first trying the match again at the same offset with
1244 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1245     starting offset (see below) and trying an ordinary match again. There is some
1246     code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1247     <pre>
1248     PCRE_NO_UTF8_CHECK
1249     </pre>
1250     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1251     string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1252     The value of <i>startoffset</i> is also checked to ensure that it points to the
1253 ph10 211 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1254     strings in the
1255     <a href="pcre.html#utf8strings">section on UTF-8 support</a>
1256     in the main
1257     <a href="pcre.html"><b>pcre</b></a>
1258     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
1259     the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,
1260     PCRE_ERROR_BADUTF8_OFFSET is returned.
1261 nigel 63 </P>
1262     <P>
1263 nigel 75 If you already know that your subject is valid, and you want to skip these
1264     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1265     calling <b>pcre_exec()</b>. You might want to do this for the second and
1266     subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1267     all the matches in a single subject string. However, you should be sure that
1268     the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1269     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1270     subject, or a value of <i>startoffset</i> that does not point to the start of a
1271     UTF-8 character, is undefined. Your program may crash.
1272     <pre>
1273     PCRE_PARTIAL
1274     </pre>
1275     This option turns on the partial matching feature. If the subject string fails
1276     to match the pattern, but at some point during the matching process the end of
1277     the subject was reached (that is, the subject partially matches the pattern and
1278     the failure to match occurred only because there were not enough subject
1279     characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1280     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1281     may appear in the pattern. These are discussed in the
1282     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1283     documentation.
1284     </P>
1285     <br><b>
1286     The string to be matched by <b>pcre_exec()</b>
1287     </b><br>
1288     <P>
1289 nigel 63 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1290 nigel 73 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1291 nigel 75 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1292     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1293     bytes. When the starting offset is zero, the search for a match starts at the
1294     beginning of the subject, and this is by far the most common case.
1295 nigel 63 </P>
1296     <P>
1297     A non-zero starting offset is useful when searching for another match in the
1298     same subject by calling <b>pcre_exec()</b> again after a previous success.
1299     Setting <i>startoffset</i> differs from just passing over a shortened string and
1300     setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1301     lookbehind. For example, consider the pattern
1302     <pre>
1303     \Biss\B
1304 nigel 75 </pre>
1305 nigel 63 which finds occurrences of "iss" in the middle of words. (\B matches only if
1306     the current position in the subject is not a word boundary.) When applied to
1307     the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1308     occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1309     subject, namely "issipi", it does not match, because \B is always false at the
1310     start of the subject, which is deemed to be a word boundary. However, if
1311     <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1312     set to 4, it finds the second occurrence of "iss" because it is able to look
1313     behind the starting point to discover that it is preceded by a letter.
1314     </P>
1315     <P>
1316     If a non-zero starting offset is passed when the pattern is anchored, one
1317 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1318 nigel 63 pattern does not require the match to be at the start of the subject.
1319     </P>
1320 nigel 75 <br><b>
1321     How <b>pcre_exec()</b> returns captured substrings
1322     </b><br>
1323 nigel 63 <P>
1324     In general, a pattern matches a certain portion of the subject, and in
1325     addition, further substrings from the subject may be picked out by parts of the
1326     pattern. Following the usage in Jeffrey Friedl's book, this is called
1327     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1328     a fragment of a pattern that picks out a substring. PCRE supports several other
1329     kinds of parenthesized subpattern that do not cause substrings to be captured.
1330     </P>
1331     <P>
1332     Captured substrings are returned to the caller via a vector of integer offsets
1333     whose address is passed in <i>ovector</i>. The number of elements in the vector
1334 nigel 75 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1335     this argument is NOT the size of <i>ovector</i> in bytes.
1336 nigel 63 </P>
1337     <P>
1338 nigel 75 The first two-thirds of the vector is used to pass back captured substrings,
1339     each substring using a pair of integers. The remaining third of the vector is
1340     used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1341     and is not available for passing back information. The length passed in
1342     <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1343     rounded down.
1344     </P>
1345     <P>
1346     When a match is successful, information about captured substrings is returned
1347     in pairs of integers, starting at the beginning of <i>ovector</i>, and
1348 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1349     pair is set to the offset of the first character in a substring, and the second
1350     is set to the offset of the first character after the end of a substring. The
1351     first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1352     subject string matched by the entire pattern. The next pair is used for the
1353     first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1354 nigel 91 is one more than the highest numbered pair that has been set. For example, if
1355     two substrings have been captured, the returned value is 3. If there are no
1356     capturing subpatterns, the return value from a successful match is 1,
1357     indicating that just the first pair of offsets has been set.
1358 nigel 63 </P>
1359     <P>
1360     If a capturing subpattern is matched repeatedly, it is the last portion of the
1361 nigel 75 string that it matched that is returned.
1362 nigel 63 </P>
1363     <P>
1364 nigel 75 If the vector is too small to hold all the captured substring offsets, it is
1365     used as far as possible (up to two-thirds of its length), and the function
1366     returns a value of zero. In particular, if the substring offsets are not of
1367     interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1368 nigel 63 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1369 nigel 75 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1370     has to get additional memory for use during matching. Thus it is usually
1371     advisable to supply an <i>ovector</i>.
1372 nigel 63 </P>
1373     <P>
1374 nigel 91 The <b>pcre_info()</b> function can be used to find out how many capturing
1375 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1376     <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1377     the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1378 nigel 91 </P>
1379     <P>
1380     It is possible for capturing subpattern number <i>n+1</i> to match some part of
1381     the subject when subpattern <i>n</i> has not been used at all. For example, if
1382     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1383     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1384     happens, both values in the offset pairs corresponding to unused subpatterns
1385     are set to -1.
1386     </P>
1387     <P>
1388     Offset values that correspond to unused subpatterns at the end of the
1389     expression are also set to -1. For example, if the string "abc" is matched
1390     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1391     return from the function is 2, because the highest used capturing subpattern
1392     number is 1. However, you can refer to the offsets for the second and third
1393     capturing subpatterns if you wish (assuming the vector is large enough, of
1394     course).
1395     </P>
1396     <P>
1397     Some convenience functions are provided for extracting the captured substrings
1398     as separate strings. These are described below.
1399 nigel 77 <a name="errorlist"></a></P>
1400 nigel 75 <br><b>
1401 nigel 91 Error return values from <b>pcre_exec()</b>
1402 nigel 75 </b><br>
1403 nigel 63 <P>
1404     If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1405     defined in the header file:
1406     <pre>
1407     PCRE_ERROR_NOMATCH (-1)
1408 nigel 75 </pre>
1409 nigel 63 The subject string did not match the pattern.
1410     <pre>
1411     PCRE_ERROR_NULL (-2)
1412 nigel 75 </pre>
1413 nigel 63 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1414     NULL and <i>ovecsize</i> was not zero.
1415     <pre>
1416     PCRE_ERROR_BADOPTION (-3)
1417 nigel 75 </pre>
1418 nigel 63 An unrecognized bit was set in the <i>options</i> argument.
1419     <pre>
1420     PCRE_ERROR_BADMAGIC (-4)
1421 nigel 75 </pre>
1422 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1423 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1424     compiled in an environment of one endianness is run in an environment with the
1425     other endianness. This is the error that PCRE gives when the magic number is
1426     not present.
1427 nigel 63 <pre>
1428 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1429 nigel 75 </pre>
1430 nigel 63 While running the pattern match, an unknown item was encountered in the
1431     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1432     of the compiled pattern.
1433     <pre>
1434     PCRE_ERROR_NOMEMORY (-6)
1435 nigel 75 </pre>
1436 nigel 63 If a pattern contains back references, but the <i>ovector</i> that is passed to
1437     <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1438     gets a block of memory at the start of matching to use for this purpose. If the
1439 nigel 75 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1440     automatically freed at the end of matching.
1441 nigel 63 <pre>
1442     PCRE_ERROR_NOSUBSTRING (-7)
1443 nigel 75 </pre>
1444 nigel 63 This error is used by the <b>pcre_copy_substring()</b>,
1445     <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1446     below). It is never returned by <b>pcre_exec()</b>.
1447     <pre>
1448     PCRE_ERROR_MATCHLIMIT (-8)
1449 nigel 75 </pre>
1450 nigel 87 The backtracking limit, as specified by the <i>match_limit</i> field in a
1451     <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1452     above.
1453     <pre>
1454 nigel 63 PCRE_ERROR_CALLOUT (-9)
1455 nigel 75 </pre>
1456 nigel 63 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1457     use by callout functions that want to yield a distinctive error code. See the
1458 nigel 75 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1459     documentation for details.
1460 nigel 71 <pre>
1461 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1462 nigel 75 </pre>
1463 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1464 nigel 73 <pre>
1465     PCRE_ERROR_BADUTF8_OFFSET (-11)
1466 nigel 75 </pre>
1467 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1468     of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1469 nigel 75 <pre>
1470 nigel 77 PCRE_ERROR_PARTIAL (-12)
1471 nigel 75 </pre>
1472     The subject string did not match, but it did match partially. See the
1473     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1474     documentation for details of partial matching.
1475     <pre>
1476 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1477 nigel 75 </pre>
1478     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1479     are not supported for partial matching. See the
1480     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1481     documentation for details of partial matching.
1482     <pre>
1483 nigel 77 PCRE_ERROR_INTERNAL (-14)
1484 nigel 75 </pre>
1485     An unexpected internal error has occurred. This error could be caused by a bug
1486     in PCRE or by overwriting of the compiled pattern.
1487     <pre>
1488 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1489 nigel 75 </pre>
1490     This error is given if the value of the <i>ovecsize</i> argument is negative.
1491 nigel 93 <pre>
1492     PCRE_ERROR_RECURSIONLIMIT (-21)
1493     </pre>
1494     The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1495     field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1496     description above.
1497     <pre>
1498     PCRE_ERROR_BADNEWLINE (-23)
1499     </pre>
1500     An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1501 nigel 73 </P>
1502 nigel 93 <P>
1503 ph10 197 Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
1504 nigel 93 </P>
1505 nigel 91 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1506 nigel 63 <P>
1507     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1508     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1509     <b>int <i>buffersize</i>);</b>
1510     </P>
1511     <P>
1512     <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1513     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1514     <b>const char **<i>stringptr</i>);</b>
1515     </P>
1516     <P>
1517     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1518     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1519     </P>
1520     <P>
1521     Captured substrings can be accessed directly by using the offsets returned by
1522     <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1523     <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1524     <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1525     as new, separate, zero-terminated strings. These functions identify substrings
1526     by number. The next section describes functions for extracting named
1527 nigel 91 substrings.
1528 nigel 63 </P>
1529     <P>
1530 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1531     further zero added on the end, but the result is not, of course, a C string.
1532     However, you can process such a string by referring to the length that is
1533     returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1534     Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1535     for handling strings containing binary zeros, because the end of the final
1536     string is not independently indicated.
1537     </P>
1538     <P>
1539 nigel 63 The first three arguments are the same for all three of these functions:
1540 nigel 75 <i>subject</i> is the subject string that has just been successfully matched,
1541 nigel 63 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1542     <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1543     captured by the match, including the substring that matched the entire regular
1544 nigel 75 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1545     than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1546     space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1547     number of elements in the vector divided by three.
1548 nigel 63 </P>
1549     <P>
1550     The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1551     extract a single substring, whose number is given as <i>stringnumber</i>. A
1552 nigel 75 value of zero extracts the substring that matched the entire pattern, whereas
1553 nigel 63 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1554     the string is placed in <i>buffer</i>, whose length is given by
1555     <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1556     obtained via <b>pcre_malloc</b>, and its address is returned via
1557     <i>stringptr</i>. The yield of the function is the length of the string, not
1558 nigel 93 including the terminating zero, or one of these error codes:
1559 nigel 63 <pre>
1560     PCRE_ERROR_NOMEMORY (-6)
1561 nigel 75 </pre>
1562 nigel 63 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1563     memory failed for <b>pcre_get_substring()</b>.
1564     <pre>
1565     PCRE_ERROR_NOSUBSTRING (-7)
1566 nigel 75 </pre>
1567 nigel 63 There is no substring whose number is <i>stringnumber</i>.
1568     </P>
1569     <P>
1570     The <b>pcre_get_substring_list()</b> function extracts all available substrings
1571     and builds a list of pointers to them. All this is done in a single block of
1572 nigel 75 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1573 nigel 63 is returned via <i>listptr</i>, which is also the start of the list of string
1574     pointers. The end of the list is marked by a NULL pointer. The yield of the
1575 nigel 93 function is zero if all went well, or the error code
1576 nigel 63 <pre>
1577     PCRE_ERROR_NOMEMORY (-6)
1578 nigel 75 </pre>
1579 nigel 63 if the attempt to get the memory block failed.
1580     </P>
1581     <P>
1582     When any of these functions encounter a substring that is unset, which can
1583     happen when capturing subpattern number <i>n+1</i> matches some part of the
1584     subject, but subpattern <i>n</i> has not been used at all, they return an empty
1585     string. This can be distinguished from a genuine zero-length substring by
1586     inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1587     substrings.
1588     </P>
1589     <P>
1590     The two convenience functions <b>pcre_free_substring()</b> and
1591     <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1592     a previous call of <b>pcre_get_substring()</b> or
1593     <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1594     the function pointed to by <b>pcre_free</b>, which of course could be called
1595     directly from a C program. However, PCRE is used in some situations where it is
1596 nigel 91 linked via a special interface to another programming language that cannot use
1597 nigel 63 <b>pcre_free</b> directly; it is for these cases that the functions are
1598     provided.
1599     </P>
1600 nigel 91 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1601 nigel 63 <P>
1602 nigel 75 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1603     <b>const char *<i>name</i>);</b>
1604     </P>
1605     <P>
1606 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1607     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1608     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1609     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1610     </P>
1611     <P>
1612     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1613     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1614     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1615     <b>const char **<i>stringptr</i>);</b>
1616     </P>
1617     <P>
1618 nigel 75 To extract a substring by name, you first have to find associated number.
1619     For example, for this pattern
1620 nigel 63 <pre>
1621 nigel 93 (a+)b(?&#60;xxx&#62;\d+)...
1622 nigel 75 </pre>
1623 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1624     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1625     calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1626     pattern, and the second is the name. The yield of the function is the
1627 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1628     that name.
1629 nigel 63 </P>
1630     <P>
1631 nigel 75 Given the number, you can extract the substring directly, or use one of the
1632     functions described in the previous section. For convenience, there are also
1633     two functions that do the whole job.
1634 nigel 63 </P>
1635     <P>
1636 nigel 91 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1637     <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1638 nigel 75 functions that extract by number. As these are described in the previous
1639     section, they are not re-described here. There are just two differences:
1640 nigel 63 </P>
1641     <P>
1642     First, instead of a substring number, a substring name is given. Second, there
1643     is an extra argument, given at the start, which is a pointer to the compiled
1644     pattern. This is needed in order to gain access to the name-to-number
1645     translation table.
1646     </P>
1647     <P>
1648     These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1649 ph10 128 then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
1650     appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1651     the behaviour may not be what you want (see the next section).
1652 nigel 63 </P>
1653 nigel 91 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1654 nigel 63 <P>
1655 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1656     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1657     </P>
1658     <P>
1659     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1660     are not required to be unique. Normally, patterns with duplicate names are such
1661     that in any one match, only one of the named subpatterns participates. An
1662     example is shown in the
1663     <a href="pcrepattern.html"><b>pcrepattern</b></a>
1664 ph10 208 documentation.
1665     </P>
1666     <P>
1667     When duplicates are present, <b>pcre_copy_named_substring()</b> and
1668     <b>pcre_get_named_substring()</b> return the first substring corresponding to
1669     the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1670     returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
1671     returns one of the numbers that are associated with the name, but it is not
1672     defined which it is.
1673     </P>
1674     <P>
1675 nigel 91 If you want to get full details of all captured substrings for a given name,
1676     you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1677     argument is the compiled pattern, and the second is the name. The third and
1678     fourth are pointers to variables which are updated by the function. After it
1679     has run, they point to the first and last entries in the name-to-number table
1680     for the given name. The function itself returns the length of each entry, or
1681 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1682     described above in the section entitled <i>Information about a pattern</i>.
1683     Given all the relevant entries for the name, you can extract each of their
1684     numbers, and hence the captured data, if any.
1685 nigel 91 </P>
1686     <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1687     <P>
1688 nigel 77 The traditional matching function uses a similar algorithm to Perl, which stops
1689     when it finds the first match, starting at a given point in the subject. If you
1690     want to find all possible matches, or the longest possible match, consider
1691     using the alternative matching function (see below) instead. If you cannot use
1692     the alternative function, but still need to find all possible matches, you
1693     can kludge it up by making use of the callout facility, which is described in
1694     the
1695     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1696     documentation.
1697     </P>
1698     <P>
1699     What you have to do is to insert a callout right at the end of the pattern.
1700     When your callout function is called, extract and save the current matched
1701     substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1702     other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1703     will yield PCRE_ERROR_NOMATCH.
1704     <a name="dfamatch"></a></P>
1705 nigel 91 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1706 nigel 77 <P>
1707     <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1708     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1709     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1710     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1711     </P>
1712     <P>
1713     The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1714 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1715     just once, and does not backtrack. This has different characteristics to the
1716     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1717     patterns are not supported. Nevertheless, there are times when this kind of
1718     matching can be useful. For a discussion of the two matching algorithms, see
1719     the
1720 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
1721     documentation.
1722     </P>
1723     <P>
1724     The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1725     <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1726     different way, and this is described below. The other common arguments are used
1727     in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1728     here.
1729     </P>
1730     <P>
1731     The two additional arguments provide workspace for the function. The workspace
1732     vector should contain at least 20 elements. It is used for keeping track of
1733     multiple paths through the pattern tree. More workspace will be needed for
1734 nigel 91 patterns and subjects where there are a lot of potential matches.
1735 nigel 77 </P>
1736     <P>
1737 nigel 87 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1738 nigel 77 <pre>
1739     int rc;
1740     int ovector[10];
1741     int wspace[20];
1742 nigel 87 rc = pcre_dfa_exec(
1743 nigel 77 re, /* result of pcre_compile() */
1744     NULL, /* we didn't study the pattern */
1745     "some string", /* the subject string */
1746     11, /* the length of the subject string */
1747     0, /* start at offset 0 in the subject */
1748     0, /* default options */
1749     ovector, /* vector of integers for substring information */
1750     10, /* number of elements (NOT size in bytes) */
1751     wspace, /* working space vector */
1752     20); /* number of elements (NOT size in bytes) */
1753     </PRE>
1754     </P>
1755     <br><b>
1756     Option bits for <b>pcre_dfa_exec()</b>
1757     </b><br>
1758     <P>
1759     The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1760 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1761     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1762 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1763     the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1764     <pre>
1765     PCRE_PARTIAL
1766     </pre>
1767     This has the same general effect as it does for <b>pcre_exec()</b>, but the
1768     details are slightly different. When PCRE_PARTIAL is set for
1769     <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1770     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1771     complete matches, but there is still at least one matching possibility. The
1772     portion of the string that provided the partial match is set as the first
1773     matching string.
1774     <pre>
1775     PCRE_DFA_SHORTEST
1776     </pre>
1777     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1778 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1779     works, this is necessarily the shortest possible match at the first possible
1780     matching point in the subject string.
1781 nigel 77 <pre>
1782     PCRE_DFA_RESTART
1783     </pre>
1784     When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1785     a partial match, it is possible to call it again, with additional subject
1786     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1787     option requests this action; when it is set, the <i>workspace</i> and
1788     <i>wscount</i> options must reference the same vector as before because data
1789     about the match so far is left in them after a partial match. There is more
1790     discussion of this facility in the
1791     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1792     documentation.
1793     </P>
1794     <br><b>
1795     Successful returns from <b>pcre_dfa_exec()</b>
1796     </b><br>
1797     <P>
1798     When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1799     substring in the subject. Note, however, that all the matches from one run of
1800     the function start at the same point in the subject. The shorter matches are
1801     all initial substrings of the longer matches. For example, if the pattern
1802     <pre>
1803     &#60;.*&#62;
1804     </pre>
1805     is matched against the string
1806     <pre>
1807     This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1808     </pre>
1809     the three matched strings are
1810     <pre>
1811     &#60;something&#62;
1812     &#60;something&#62; &#60;something else&#62;
1813     &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1814     </pre>
1815     On success, the yield of the function is a number greater than zero, which is
1816     the number of matched substrings. The substrings themselves are returned in
1817     <i>ovector</i>. Each string uses two elements; the first is the offset to the
1818 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1819     the same start offset. (Space could have been saved by giving this only once,
1820     but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1821     returns data, even though the meaning of the strings is different.)
1822 nigel 77 </P>
1823     <P>
1824     The strings are returned in reverse order of length; that is, the longest
1825     matching string is given first. If there were too many matches to fit into
1826     <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1827     the longest matches.
1828     </P>
1829     <br><b>
1830     Error returns from <b>pcre_dfa_exec()</b>
1831     </b><br>
1832     <P>
1833     The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1834     Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1835     described
1836     <a href="#errorlist">above.</a>
1837     There are in addition the following errors that are specific to
1838     <b>pcre_dfa_exec()</b>:
1839     <pre>
1840     PCRE_ERROR_DFA_UITEM (-16)
1841     </pre>
1842     This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1843     that it does not support, for instance, the use of \C or a back reference.
1844     <pre>
1845     PCRE_ERROR_DFA_UCOND (-17)
1846     </pre>
1847 nigel 93 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1848     uses a back reference for the condition, or a test for recursion in a specific
1849     group. These are not supported.
1850 nigel 77 <pre>
1851     PCRE_ERROR_DFA_UMLIMIT (-18)
1852     </pre>
1853     This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1854     block that contains a setting of the <i>match_limit</i> field. This is not
1855     supported (it is meaningless).
1856     <pre>
1857     PCRE_ERROR_DFA_WSSIZE (-19)
1858     </pre>
1859     This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1860     <i>workspace</i> vector.
1861     <pre>
1862     PCRE_ERROR_DFA_RECURSE (-20)
1863     </pre>
1864     When a recursive subpattern is processed, the matching function calls itself
1865     recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1866     error is given if the output vector is not large enough. This should be
1867     extremely rare, as a vector of size 1000 is used.
1868     </P>
1869 nigel 93 <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1870 nigel 77 <P>
1871 nigel 93 <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1872     <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1873     <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1874     </P>
1875 ph10 99 <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1876 nigel 93 <P>
1877 ph10 99 Philip Hazel
1878 nigel 63 <br>
1879 ph10 99 University Computing Service
1880     <br>
1881     Cambridge CB2 3QH, England.
1882     <br>
1883     </P>
1884     <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1885     <P>
1886 ph10 211 Last updated: 09 August 2007
1887 ph10 99 <br>
1888     Copyright &copy; 1997-2007 University of Cambridge.
1889     <br>
1890 nigel 75 <p>
1891     Return to the <a href="index.html">PCRE index page</a>.
1892     </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