/[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 292 - (hide annotations) (download) (as text)
Thu Dec 27 09:27:23 2007 UTC (6 years, 7 months ago) by ph10
File MIME type: text/html
File size: 88827 byte(s)
Tidies for 7.5-RC2 release.

1 nigel 63 <html>
2     <head>
3     <title>pcreapi specification</title>
4     </head>
5     <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 nigel 75 <h1>pcreapi man page</h1>
7     <p>
8     Return to the <a href="index.html">PCRE index page</a>.
9     </p>
10 ph10 111 <p>
11 nigel 75 This page is part of the PCRE HTML documentation. It was generated automatically
12     from the original man page. If there is any nonsense in it, please consult the
13     man page, in case the conversion went wrong.
14 ph10 111 <br>
15 nigel 63 <ul>
16 nigel 75 <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17     <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18 nigel 91 <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19     <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
20     <li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
21     <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22     <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23     <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24     <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25     <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26     <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27     <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28     <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
29     <li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
30     <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
31     <li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
32     <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33     <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
34     <li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
35 nigel 93 <li><a name="TOC20" href="#SEC20">SEE ALSO</a>
36 ph10 99 <li><a name="TOC21" href="#SEC21">AUTHOR</a>
37     <li><a name="TOC22" href="#SEC22">REVISION</a>
38 nigel 63 </ul>
39 nigel 75 <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
40 nigel 63 <P>
41     <b>#include &#60;pcre.h&#62;</b>
42     </P>
43     <P>
44     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
45     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
46     <b>const unsigned char *<i>tableptr</i>);</b>
47     </P>
48     <P>
49 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
50     <b>int *<i>errorcodeptr</i>,</b>
51     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
52     <b>const unsigned char *<i>tableptr</i>);</b>
53     </P>
54     <P>
55 nigel 63 <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
56     <b>const char **<i>errptr</i>);</b>
57     </P>
58     <P>
59     <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
60     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
61     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
62     </P>
63     <P>
64 nigel 77 <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
65     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
66     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
67     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
68     </P>
69     <P>
70 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
71     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
72     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
73     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
74     </P>
75     <P>
76     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
77     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
78     <b>int <i>buffersize</i>);</b>
79     </P>
80     <P>
81     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
82     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
83     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
84     <b>const char **<i>stringptr</i>);</b>
85     </P>
86     <P>
87     <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
88     <b>const char *<i>name</i>);</b>
89     </P>
90     <P>
91 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
92     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
93     </P>
94     <P>
95 nigel 63 <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
96     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
97     <b>const char **<i>stringptr</i>);</b>
98     </P>
99     <P>
100     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
101     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
102     </P>
103     <P>
104     <b>void pcre_free_substring(const char *<i>stringptr</i>);</b>
105     </P>
106     <P>
107     <b>void pcre_free_substring_list(const char **<i>stringptr</i>);</b>
108     </P>
109     <P>
110     <b>const unsigned char *pcre_maketables(void);</b>
111     </P>
112     <P>
113     <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
114     <b>int <i>what</i>, void *<i>where</i>);</b>
115     </P>
116     <P>
117     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
118     <b>*<i>firstcharptr</i>);</b>
119     </P>
120     <P>
121 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
122     </P>
123     <P>
124 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
125     </P>
126     <P>
127     <b>char *pcre_version(void);</b>
128     </P>
129     <P>
130     <b>void *(*pcre_malloc)(size_t);</b>
131     </P>
132     <P>
133     <b>void (*pcre_free)(void *);</b>
134     </P>
135     <P>
136 nigel 73 <b>void *(*pcre_stack_malloc)(size_t);</b>
137     </P>
138     <P>
139     <b>void (*pcre_stack_free)(void *);</b>
140     </P>
141     <P>
142 nigel 63 <b>int (*pcre_callout)(pcre_callout_block *);</b>
143     </P>
144 nigel 75 <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
145 nigel 63 <P>
146 nigel 93 PCRE has its own native API, which is described in this document. There are
147     also some wrapper functions that correspond to the POSIX regular expression
148 nigel 77 API. These are described in the
149 nigel 75 <a href="pcreposix.html"><b>pcreposix</b></a>
150 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
151     wrapper is distributed with PCRE. It is documented in the
152     <a href="pcrecpp.html"><b>pcrecpp</b></a>
153     page.
154 nigel 63 </P>
155     <P>
156 nigel 77 The native API C function prototypes are defined in the header file
157     <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
158     It can normally be accessed by adding <b>-lpcre</b> to the command for linking
159     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
160     and PCRE_MINOR to contain the major and minor release numbers for the library.
161 nigel 75 Applications can use these to include support for different releases of PCRE.
162 nigel 63 </P>
163     <P>
164 nigel 77 The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
165     and <b>pcre_exec()</b> are used for compiling and matching regular expressions
166     in a Perl-compatible manner. A sample program that demonstrates the simplest
167     way of using them is provided in the file called <i>pcredemo.c</i> in the source
168     distribution. The
169 nigel 75 <a href="pcresample.html"><b>pcresample</b></a>
170     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 ph10 227 <a name="newlines"></a></P>
247 nigel 91 <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248 nigel 63 <P>
249 ph10 150 PCRE supports five different conventions for indicating line breaks in
250 nigel 93 strings: a single CR (carriage return) character, a single LF (linefeed)
251 ph10 150 character, the two-character sequence CRLF, any of the three preceding, or any
252     Unicode newline sequence. The Unicode newline sequences are the three just
253     mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
254     U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
255     (paragraph separator, U+2029).
256 nigel 93 </P>
257     <P>
258     Each of the first three conventions is used by at least one operating system as
259     its standard newline sequence. When PCRE is built, a default can be specified.
260     The default default is LF, which is the Unix standard. When PCRE is run, the
261     default can be overridden, either when a pattern is compiled, or when it is
262     matched.
263     </P>
264     <P>
265 ph10 227 At compile time, the newline convention can be specified by the <i>options</i>
266     argument of <b>pcre_compile()</b>, or it can be specified by special text at the
267     start of the pattern itself; this overrides any other settings. See the
268     <a href="pcrepattern.html"><b>pcrepattern</b></a>
269     page for details of the special character sequences.
270     </P>
271     <P>
272 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
273 nigel 93 pair of characters that indicate a line break". The choice of newline
274     convention affects the handling of the dot, circumflex, and dollar
275     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
276     recognized line ending sequence, the match position advancement for a
277 ph10 227 non-anchored pattern. There is more detail about this in the
278     <a href="#execoptions">section on <b>pcre_exec()</b> options</a>
279 ph10 231 below.
280 nigel 91 </P>
281 ph10 231 <P>
282     The choice of newline convention does not affect the interpretation of
283     the \n or \r escape sequences, nor does it affect what \R matches, which is
284     controlled in a similar way, but by separate options.
285     </P>
286 nigel 91 <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
287     <P>
288 nigel 63 The PCRE functions can be used in multi-threading applications, with the
289 nigel 73 proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
290     <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
291     callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
292 nigel 63 </P>
293     <P>
294     The compiled form of a regular expression is not altered during matching, so
295     the same compiled pattern can safely be used by several threads at once.
296     </P>
297 nigel 91 <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
298 nigel 63 <P>
299 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
300     time, possibly by a different program, and even on a host other than the one on
301     which it was compiled. Details are given in the
302     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
303 ph10 155 documentation. However, compiling a regular expression with one version of PCRE
304     for use with a different version is not guaranteed to work and may cause
305     crashes.
306 nigel 75 </P>
307 nigel 91 <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
308 nigel 75 <P>
309 nigel 63 <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
310     </P>
311     <P>
312     The function <b>pcre_config()</b> makes it possible for a PCRE client to
313     discover which optional features have been compiled into the PCRE library. The
314     <a href="pcrebuild.html"><b>pcrebuild</b></a>
315     documentation has more details about these optional features.
316     </P>
317     <P>
318     The first argument for <b>pcre_config()</b> is an integer, specifying which
319     information is required; the second argument is a pointer to a variable into
320     which the information is placed. The following information is available:
321     <pre>
322     PCRE_CONFIG_UTF8
323 nigel 75 </pre>
324 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
325     otherwise it is set to zero.
326     <pre>
327 nigel 75 PCRE_CONFIG_UNICODE_PROPERTIES
328     </pre>
329     The output is an integer that is set to one if support for Unicode character
330     properties is available; otherwise it is set to zero.
331     <pre>
332 nigel 63 PCRE_CONFIG_NEWLINE
333 nigel 75 </pre>
334 nigel 91 The output is an integer whose value specifies the default character sequence
335 nigel 93 that is recognized as meaning "newline". The four values that are supported
336 ph10 150 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
337     default should normally be the standard sequence for your operating system.
338 nigel 63 <pre>
339 ph10 231 PCRE_CONFIG_BSR
340     </pre>
341     The output is an integer whose value indicates what character sequences the \R
342     escape sequence matches by default. A value of 0 means that \R matches any
343     Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
344     or CRLF. The default can be overridden when a pattern is compiled or matched.
345     <pre>
346 nigel 63 PCRE_CONFIG_LINK_SIZE
347 nigel 75 </pre>
348 nigel 63 The output is an integer that contains the number of bytes used for internal
349     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
350     allow larger regular expressions to be compiled, at the expense of slower
351     matching. The default value of 2 is sufficient for all but the most massive
352     patterns, since it allows the compiled pattern to be up to 64K in size.
353     <pre>
354     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
355 nigel 75 </pre>
356 nigel 63 The output is an integer that contains the threshold above which the POSIX
357     interface uses <b>malloc()</b> for output vectors. Further details are given in
358 nigel 75 the
359     <a href="pcreposix.html"><b>pcreposix</b></a>
360     documentation.
361 nigel 63 <pre>
362     PCRE_CONFIG_MATCH_LIMIT
363 nigel 75 </pre>
364 nigel 63 The output is an integer that gives the default limit for the number of
365     internal matching function calls in a <b>pcre_exec()</b> execution. Further
366     details are given with <b>pcre_exec()</b> below.
367 nigel 73 <pre>
368 nigel 87 PCRE_CONFIG_MATCH_LIMIT_RECURSION
369     </pre>
370     The output is an integer that gives the default limit for the depth of
371     recursion when calling the internal matching function in a <b>pcre_exec()</b>
372     execution. Further details are given with <b>pcre_exec()</b> below.
373     <pre>
374 nigel 73 PCRE_CONFIG_STACKRECURSE
375 nigel 75 </pre>
376 nigel 77 The output is an integer that is set to one if internal recursion when running
377     <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
378     to remember their state. This is the usual way that PCRE is compiled. The
379     output is zero if PCRE was compiled to use blocks of data on the heap instead
380     of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
381     <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
382     avoiding the use of the stack.
383 nigel 73 </P>
384 nigel 91 <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
385 nigel 63 <P>
386     <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
387     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
388     <b>const unsigned char *<i>tableptr</i>);</b>
389 nigel 77 <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
390     <b>int *<i>errorcodeptr</i>,</b>
391     <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
392     <b>const unsigned char *<i>tableptr</i>);</b>
393 nigel 63 </P>
394     <P>
395 nigel 77 Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
396     called to compile a pattern into an internal form. The only difference between
397     the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
398     <i>errorcodeptr</i>, via which a numerical error code can be returned.
399 nigel 63 </P>
400     <P>
401 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
402     <i>pattern</i> argument. A pointer to a single block of memory that is obtained
403     via <b>pcre_malloc</b> is returned. This contains the compiled code and related
404     data. The <b>pcre</b> type is defined for the returned block; this is a typedef
405     for a structure whose contents are not externally defined. It is up to the
406 nigel 91 caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
407 nigel 77 </P>
408     <P>
409 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
410     depend on memory location, the complete <b>pcre</b> data block is not
411 nigel 75 fully relocatable, because it may contain a copy of the <i>tableptr</i>
412     argument, which is an address (see below).
413 nigel 63 </P>
414     <P>
415 nigel 93 The <i>options</i> argument contains various bit settings that affect the
416 nigel 75 compilation. It should be zero if no options are required. The available
417     options are described below. Some of them, in particular, those that are
418     compatible with Perl, can also be set and unset from within the pattern (see
419     the detailed description in the
420     <a href="pcrepattern.html"><b>pcrepattern</b></a>
421     documentation). For these options, the contents of the <i>options</i> argument
422     specifies their initial settings at the start of compilation and execution. The
423 nigel 91 PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
424     matching as well as at compile time.
425 nigel 63 </P>
426     <P>
427     If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
428     Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
429     NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
430 nigel 87 error message. This is a static string that is part of the library. You must
431     not try to free it. The offset from the start of the pattern to the character
432     where the error was discovered is placed in the variable pointed to by
433 nigel 63 <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
434     </P>
435     <P>
436 nigel 77 If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
437     <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
438     returned via this argument in the event of an error. This is in addition to the
439     textual error message. Error codes and messages are listed below.
440     </P>
441     <P>
442 nigel 63 If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
443 nigel 75 character tables that are built when PCRE is compiled, using the default C
444     locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
445     call to <b>pcre_maketables()</b>. This value is stored with the compiled
446     pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
447     passed to it. For more discussion, see the section on locale support below.
448 nigel 63 </P>
449     <P>
450     This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
451     <pre>
452     pcre *re;
453     const char *error;
454     int erroffset;
455     re = pcre_compile(
456     "^A.*Z", /* the pattern */
457     0, /* default options */
458     &error, /* for error message */
459     &erroffset, /* for error offset */
460     NULL); /* use default character tables */
461 nigel 75 </pre>
462     The following names for option bits are defined in the <b>pcre.h</b> header
463     file:
464 nigel 63 <pre>
465     PCRE_ANCHORED
466 nigel 75 </pre>
467 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
468 nigel 75 constrained to match only at the first matching point in the string that is
469 nigel 63 being searched (the "subject string"). This effect can also be achieved by
470     appropriate constructs in the pattern itself, which is the only way to do it in
471     Perl.
472     <pre>
473 nigel 75 PCRE_AUTO_CALLOUT
474     </pre>
475     If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
476     all with number 255, before each pattern item. For discussion of the callout
477     facility, see the
478     <a href="pcrecallout.html"><b>pcrecallout</b></a>
479     documentation.
480     <pre>
481 ph10 231 PCRE_BSR_ANYCRLF
482     PCRE_BSR_UNICODE
483     </pre>
484     These options (which are mutually exclusive) control what the \R escape
485     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
486     match any Unicode newline sequence. The default is specified when PCRE is
487     built. It can be overridden from within the pattern, or by setting an option
488     when a compiled pattern is matched.
489     <pre>
490 nigel 63 PCRE_CASELESS
491 nigel 75 </pre>
492 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
493     letters. It is equivalent to Perl's /i option, and it can be changed within a
494 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
495     concept of case for characters whose values are less than 128, so caseless
496     matching is always possible. For characters with higher values, the concept of
497     case is supported if PCRE is compiled with Unicode property support, but not
498     otherwise. If you want to use caseless matching for characters 128 and above,
499     you must ensure that PCRE is compiled with Unicode property support as well as
500     with UTF-8 support.
501 nigel 63 <pre>
502     PCRE_DOLLAR_ENDONLY
503 nigel 75 </pre>
504 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
505     end of the subject string. Without this option, a dollar also matches
506 nigel 91 immediately before a newline at the end of the string (but not before any other
507     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
508     There is no equivalent to this option in Perl, and no way to set it within a
509     pattern.
510 nigel 63 <pre>
511     PCRE_DOTALL
512 nigel 75 </pre>
513 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
514 nigel 91 including those that indicate newline. Without it, a dot does not match when
515     the current position is at a newline. This option is equivalent to Perl's /s
516     option, and it can be changed within a pattern by a (?s) option setting. A
517 nigel 93 negative class such as [^a] always matches newline characters, independent of
518     the setting of this option.
519 nigel 63 <pre>
520 nigel 91 PCRE_DUPNAMES
521     </pre>
522     If this bit is set, names used to identify capturing subpatterns need not be
523     unique. This can be helpful for certain types of pattern when it is known that
524     only one instance of the named subpattern can ever be matched. There are more
525     details of named subpatterns below; see also the
526     <a href="pcrepattern.html"><b>pcrepattern</b></a>
527     documentation.
528     <pre>
529 nigel 63 PCRE_EXTENDED
530 nigel 75 </pre>
531 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
532     ignored except when escaped or inside a character class. Whitespace does not
533     include the VT character (code 11). In addition, characters between an
534 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
535     ignored. This is equivalent to Perl's /x option, and it can be changed within a
536     pattern by a (?x) option setting.
537 nigel 63 </P>
538     <P>
539     This option makes it possible to include comments inside complicated patterns.
540     Note, however, that this applies only to data characters. Whitespace characters
541     may never appear within special character sequences in a pattern, for example
542     within the sequence (?( which introduces a conditional subpattern.
543     <pre>
544     PCRE_EXTRA
545 nigel 75 </pre>
546 nigel 63 This option was invented in order to turn on additional functionality of PCRE
547     that is incompatible with Perl, but it is currently of very little use. When
548     set, any backslash in a pattern that is followed by a letter that has no
549     special meaning causes an error, thus reserving these combinations for future
550     expansion. By default, as in Perl, a backslash followed by a letter with no
551 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
552     give a warning for this.) There are at present no other features controlled by
553     this option. It can also be set by a (?X) option setting within a pattern.
554 nigel 63 <pre>
555 nigel 77 PCRE_FIRSTLINE
556     </pre>
557     If this option is set, an unanchored pattern is required to match before or at
558 nigel 91 the first newline in the subject string, though the matched text may continue
559     over the newline.
560 nigel 77 <pre>
561 nigel 63 PCRE_MULTILINE
562 nigel 75 </pre>
563     By default, PCRE treats the subject string as consisting of a single line of
564     characters (even if it actually contains newlines). The "start of line"
565 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
566     line" metacharacter ($) matches only at the end of the string, or before a
567     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
568     Perl.
569     </P>
570     <P>
571     When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
572 nigel 91 match immediately following or immediately before internal newlines in the
573     subject string, respectively, as well as at the very start and end. This is
574     equivalent to Perl's /m option, and it can be changed within a pattern by a
575     (?m) option setting. If there are no newlines in a subject string, or no
576 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
577     <pre>
578 nigel 91 PCRE_NEWLINE_CR
579     PCRE_NEWLINE_LF
580     PCRE_NEWLINE_CRLF
581 ph10 150 PCRE_NEWLINE_ANYCRLF
582 nigel 93 PCRE_NEWLINE_ANY
583 nigel 91 </pre>
584     These options override the default newline definition that was chosen when PCRE
585     was built. Setting the first or the second specifies that a newline is
586 nigel 93 indicated by a single character (CR or LF, respectively). Setting
587     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
588 ph10 150 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
589     preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
590     that any Unicode newline sequence should be recognized. The Unicode newline
591     sequences are the three just mentioned, plus the single characters VT (vertical
592     tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
593     separator, U+2028), and PS (paragraph separator, U+2029). The last two are
594     recognized only in UTF-8 mode.
595 nigel 91 </P>
596     <P>
597 nigel 93 The newline setting in the options word uses three bits that are treated
598 ph10 150 as a number, giving eight possibilities. Currently only six are used (default
599     plus the five values above). This means that if you set more than one newline
600 nigel 93 option, the combination may or may not be sensible. For example,
601     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
602 ph10 150 other combinations may yield unused numbers and cause an error.
603 nigel 93 </P>
604     <P>
605     The only time that a line break is specially recognized when compiling a
606     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
607     class is encountered. This indicates a comment that lasts until after the next
608     line break sequence. In other circumstances, line break sequences are treated
609     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
610     as whitespace characters and are therefore ignored.
611     </P>
612     <P>
613     The newline option that is set at compile time becomes the default that is used
614     for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
615 nigel 91 <pre>
616 nigel 63 PCRE_NO_AUTO_CAPTURE
617 nigel 75 </pre>
618 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
619     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
620     were followed by ?: but named parentheses can still be used for capturing (and
621     they acquire numbers in the usual way). There is no equivalent of this option
622     in Perl.
623     <pre>
624     PCRE_UNGREEDY
625 nigel 75 </pre>
626 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
627     greedy by default, but become greedy if followed by "?". It is not compatible
628     with Perl. It can also be set by a (?U) option setting within the pattern.
629     <pre>
630     PCRE_UTF8
631 nigel 75 </pre>
632 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
633     of UTF-8 characters instead of single-byte character strings. However, it is
634 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
635 nigel 63 of this option provokes an error. Details of how this option changes the
636     behaviour of PCRE are given in the
637     <a href="pcre.html#utf8support">section on UTF-8 support</a>
638     in the main
639     <a href="pcre.html"><b>pcre</b></a>
640     page.
641 nigel 71 <pre>
642     PCRE_NO_UTF8_CHECK
643 nigel 75 </pre>
644 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
645 ph10 211 automatically checked. There is a discussion about the
646     <a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
647     in the main
648     <a href="pcre.html"><b>pcre</b></a>
649     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
650     returns an error. If you already know that your pattern is valid, and you want
651     to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
652     option. When it is set, the effect of passing an invalid UTF-8 string as a
653     pattern is undefined. It may cause your program to crash. Note that this option
654     can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
655     the UTF-8 validity checking of subject strings.
656 nigel 71 </P>
657 nigel 91 <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
658 nigel 63 <P>
659 nigel 77 The following table lists the error codes than may be returned by
660     <b>pcre_compile2()</b>, along with the error messages that may be returned by
661 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
662     out of use. To avoid confusion, they have not been re-used.
663 nigel 77 <pre>
664     0 no error
665     1 \ at end of pattern
666     2 \c at end of pattern
667     3 unrecognized character follows \
668     4 numbers out of order in {} quantifier
669     5 number too big in {} quantifier
670     6 missing terminating ] for character class
671     7 invalid escape sequence in character class
672     8 range out of order in character class
673     9 nothing to repeat
674 nigel 93 10 [this code is not in use]
675 nigel 77 11 internal error: unexpected repeat
676 ph10 292 12 unrecognized character after (? or (?-
677 nigel 77 13 POSIX named classes are supported only within a class
678     14 missing )
679     15 reference to non-existent subpattern
680     16 erroffset passed as NULL
681     17 unknown option bit(s) set
682     18 missing ) after comment
683 nigel 93 19 [this code is not in use]
684 ph10 292 20 regular expression is too large
685 nigel 77 21 failed to get memory
686     22 unmatched parentheses
687     23 internal error: code overflow
688     24 unrecognized character after (?&#60;
689     25 lookbehind assertion is not fixed length
690 nigel 91 26 malformed number or name after (?(
691 nigel 77 27 conditional group contains more than two branches
692     28 assertion expected after (?(
693 ph10 182 29 (?R or (?[+-]digits must be followed by )
694 nigel 77 30 unknown POSIX class name
695     31 POSIX collating elements are not supported
696     32 this version of PCRE is not compiled with PCRE_UTF8 support
697 nigel 93 33 [this code is not in use]
698 nigel 77 34 character value in \x{...} sequence is too large
699     35 invalid condition (?(0)
700     36 \C not allowed in lookbehind assertion
701     37 PCRE does not support \L, \l, \N, \U, or \u
702     38 number after (?C is &#62; 255
703     39 closing ) for (?C expected
704     40 recursive call could loop indefinitely
705     41 unrecognized character after (?P
706 nigel 93 42 syntax error in subpattern name (missing terminator)
707 nigel 91 43 two named subpatterns have the same name
708 nigel 77 44 invalid UTF-8 string
709     45 support for \P, \p, and \X has not been compiled
710     46 malformed \P or \p sequence
711     47 unknown property name after \P or \p
712 nigel 91 48 subpattern name is too long (maximum 32 characters)
713 ph10 292 49 too many named subpatterns (maximum 10000)
714 ph10 202 50 [this code is not in use]
715 nigel 91 51 octal value is greater than \377 (not in UTF-8 mode)
716 nigel 93 52 internal error: overran compiling workspace
717     53 internal error: previously-checked referenced subpattern not found
718     54 DEFINE group contains more than one branch
719     55 repeating a DEFINE group is not allowed
720 ph10 231 56 inconsistent NEWLINE options
721 ph10 182 57 \g is not followed by a braced name or an optionally braced
722     non-zero number
723     58 (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
724 ph10 292 59 (*VERB) with an argument is not supported
725     60 (*VERB) not recognized
726     61 number is too big
727     62 subpattern name expected
728     63 digit expected after (?+
729     </pre>
730     The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
731     be used if the limits were changed when PCRE was built.
732 nigel 77 </P>
733 nigel 91 <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
734 nigel 77 <P>
735     <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
736 nigel 63 <b>const char **<i>errptr</i>);</b>
737     </P>
738     <P>
739 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
740     more time analyzing it in order to speed up the time taken for matching. The
741 nigel 63 function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
742 nigel 75 argument. If studying the pattern produces additional information that will
743     help speed up matching, <b>pcre_study()</b> returns a pointer to a
744     <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
745     results of the study.
746 nigel 63 </P>
747     <P>
748 nigel 75 The returned value from <b>pcre_study()</b> can be passed directly to
749     <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
750 nigel 63 fields that can be set by the caller before the block is passed; these are
751 nigel 75 described
752     <a href="#extradata">below</a>
753     in the section on matching a pattern.
754 nigel 63 </P>
755     <P>
756 nigel 77 If studying the pattern does not produce any additional information
757 nigel 75 <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
758     wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
759     own <b>pcre_extra</b> block.
760 nigel 63 </P>
761     <P>
762 nigel 75 The second argument of <b>pcre_study()</b> contains option bits. At present, no
763     options are defined, and this argument should always be zero.
764     </P>
765     <P>
766 nigel 63 The third argument for <b>pcre_study()</b> is a pointer for an error message. If
767     studying succeeds (even if no data is returned), the variable it points to is
768 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
769     static string that is part of the library. You must not try to free it. You
770     should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
771     sure that it has run successfully.
772 nigel 63 </P>
773     <P>
774     This is a typical call to <b>pcre_study</b>():
775     <pre>
776     pcre_extra *pe;
777     pe = pcre_study(
778     re, /* result of pcre_compile() */
779     0, /* no options exist */
780     &error); /* set to NULL or points to a message */
781 nigel 75 </pre>
782 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
783     not have a single fixed starting character. A bitmap of possible starting
784 nigel 75 bytes is created.
785     <a name="localesupport"></a></P>
786 nigel 91 <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
787 nigel 63 <P>
788 ph10 142 PCRE handles caseless matching, and determines whether characters are letters,
789 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
790 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
791 nigel 75 less than 128. Higher-valued codes never match escapes such as \w or \d, but
792     can be tested with \p if PCRE is built with Unicode character property
793 ph10 142 support. The use of locales with Unicode is discouraged. If you are handling
794     characters with codes greater than 128, you should either use UTF-8 and
795     Unicode, or use locales, but not try to mix the two.
796 nigel 63 </P>
797     <P>
798 ph10 142 PCRE contains an internal set of tables that are used when the final argument
799     of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
800     Normally, the internal tables recognize only ASCII characters. However, when
801     PCRE is built, it is possible to cause the internal tables to be rebuilt in the
802     default "C" locale of the local system, which may cause them to be different.
803 nigel 63 </P>
804     <P>
805 ph10 142 The internal tables can always be overridden by tables supplied by the
806     application that calls PCRE. These may be created in a different locale from
807     the default. As more and more applications change to using Unicode, the need
808     for this locale support is expected to die away.
809     </P>
810     <P>
811 nigel 75 External tables are built by calling the <b>pcre_maketables()</b> function,
812     which has no arguments, in the relevant locale. The result can then be passed
813     to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
814     example, to build and use tables that are appropriate for the French locale
815     (where accented characters with values greater than 128 are treated as letters),
816     the following code could be used:
817 nigel 63 <pre>
818 nigel 75 setlocale(LC_CTYPE, "fr_FR");
819 nigel 63 tables = pcre_maketables();
820     re = pcre_compile(..., tables);
821 nigel 75 </pre>
822 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
823     are using Windows, the name for the French locale is "french".
824     </P>
825     <P>
826 nigel 75 When <b>pcre_maketables()</b> runs, the tables are built in memory that is
827     obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
828     that the memory containing the tables remains available for as long as it is
829     needed.
830 nigel 63 </P>
831     <P>
832 nigel 75 The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
833 nigel 63 pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
834 nigel 75 and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
835     pattern, compilation, studying and matching all happen in the same locale, but
836     different patterns can be compiled in different locales.
837 nigel 63 </P>
838     <P>
839 nigel 75 It is possible to pass a table pointer or NULL (indicating the use of the
840     internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
841     this facility could be used to match a pattern in a different locale from the
842     one in which it was compiled. Passing table pointers at run time is discussed
843     below in the section on matching a pattern.
844     </P>
845 nigel 91 <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
846 nigel 75 <P>
847 nigel 63 <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
848     <b>int <i>what</i>, void *<i>where</i>);</b>
849     </P>
850     <P>
851     The <b>pcre_fullinfo()</b> function returns information about a compiled
852     pattern. It replaces the obsolete <b>pcre_info()</b> function, which is
853     nevertheless retained for backwards compability (and is documented below).
854     </P>
855     <P>
856     The first argument for <b>pcre_fullinfo()</b> is a pointer to the compiled
857     pattern. The second argument is the result of <b>pcre_study()</b>, or NULL if
858     the pattern was not studied. The third argument specifies which piece of
859     information is required, and the fourth argument is a pointer to a variable
860     to receive the data. The yield of the function is zero for success, or one of
861     the following negative numbers:
862     <pre>
863     PCRE_ERROR_NULL the argument <i>code</i> was NULL
864     the argument <i>where</i> was NULL
865     PCRE_ERROR_BADMAGIC the "magic number" was not found
866     PCRE_ERROR_BADOPTION the value of <i>what</i> was invalid
867 nigel 75 </pre>
868     The "magic number" is placed at the start of each compiled pattern as an simple
869     check against passing an arbitrary memory pointer. Here is a typical call of
870     <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
871 nigel 63 <pre>
872     int rc;
873 nigel 91 size_t length;
874 nigel 63 rc = pcre_fullinfo(
875     re, /* result of pcre_compile() */
876     pe, /* result of pcre_study(), or NULL */
877     PCRE_INFO_SIZE, /* what is required */
878     &length); /* where to put the data */
879 nigel 75 </pre>
880 nigel 63 The possible values for the third argument are defined in <b>pcre.h</b>, and are
881     as follows:
882     <pre>
883     PCRE_INFO_BACKREFMAX
884 nigel 75 </pre>
885 nigel 63 Return the number of the highest back reference in the pattern. The fourth
886     argument should point to an <b>int</b> variable. Zero is returned if there are
887     no back references.
888     <pre>
889     PCRE_INFO_CAPTURECOUNT
890 nigel 75 </pre>
891 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
892 nigel 75 should point to an <b>int</b> variable.
893 nigel 63 <pre>
894 nigel 77 PCRE_INFO_DEFAULT_TABLES
895 nigel 75 </pre>
896     Return a pointer to the internal default character tables within PCRE. The
897     fourth argument should point to an <b>unsigned char *</b> variable. This
898     information call is provided for internal use by the <b>pcre_study()</b>
899     function. External callers can cause PCRE to use its internal tables by passing
900     a NULL table pointer.
901     <pre>
902 nigel 63 PCRE_INFO_FIRSTBYTE
903 nigel 75 </pre>
904 nigel 63 Return information about the first byte of any matched string, for a
905 nigel 91 non-anchored pattern. The fourth argument should point to an <b>int</b>
906     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
907     still recognized for backwards compatibility.)
908 nigel 63 </P>
909     <P>
910 nigel 75 If there is a fixed first byte, for example, from a pattern such as
911 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
912 nigel 75 <br>
913     <br>
914 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
915     starts with "^", or
916 nigel 75 <br>
917     <br>
918 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
919     (if it were set, the pattern would be anchored),
920 nigel 75 <br>
921     <br>
922 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
923     subject string or after any newline within the string. Otherwise -2 is
924     returned. For anchored patterns, -2 is returned.
925     <pre>
926     PCRE_INFO_FIRSTTABLE
927 nigel 75 </pre>
928 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
929     table indicating a fixed set of bytes for the first byte in any matching
930     string, a pointer to the table is returned. Otherwise NULL is returned. The
931     fourth argument should point to an <b>unsigned char *</b> variable.
932     <pre>
933 ph10 227 PCRE_INFO_HASCRORLF
934     </pre>
935     Return 1 if the pattern contains any explicit matches for CR or LF characters,
936 ph10 243 otherwise 0. The fourth argument should point to an <b>int</b> variable. An
937     explicit match is either a literal CR or LF character, or \r or \n.
938 ph10 227 <pre>
939 ph10 172 PCRE_INFO_JCHANGED
940     </pre>
941 ph10 286 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
942     0. The fourth argument should point to an <b>int</b> variable. (?J) and
943     (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
944 ph10 172 <pre>
945 nigel 63 PCRE_INFO_LASTLITERAL
946 nigel 75 </pre>
947 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
948     string, other than at its start, if such a byte has been recorded. The fourth
949     argument should point to an <b>int</b> variable. If there is no such byte, -1 is
950     returned. For anchored patterns, a last literal byte is recorded only if it
951     follows something of variable length. For example, for the pattern
952     /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
953     is -1.
954 nigel 63 <pre>
955     PCRE_INFO_NAMECOUNT
956     PCRE_INFO_NAMEENTRYSIZE
957     PCRE_INFO_NAMETABLE
958 nigel 75 </pre>
959 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
960     names are just an additional way of identifying the parentheses, which still
961 nigel 91 acquire numbers. Several convenience functions such as
962     <b>pcre_get_named_substring()</b> are provided for extracting captured
963     substrings by name. It is also possible to extract the data directly, by first
964     converting the name to a number in order to access the correct pointers in the
965     output vector (described with <b>pcre_exec()</b> below). To do the conversion,
966     you need to use the name-to-number map, which is described by these three
967     values.
968 nigel 63 </P>
969     <P>
970     The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
971     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
972     entry; both of these return an <b>int</b> value. The entry size depends on the
973     length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
974     entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
975     are the number of the capturing parenthesis, most significant byte first. The
976     rest of the entry is the corresponding name, zero terminated. The names are in
977 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
978     their parentheses numbers. For example, consider the following pattern (assume
979 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
980     <pre>
981 nigel 93 (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
982 nigel 75 </pre>
983 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
984     in the table is eight bytes long. The table is as follows, with non-printing
985 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
986 nigel 63 <pre>
987     00 01 d a t e 00 ??
988     00 05 d a y 00 ?? ??
989     00 04 m o n t h 00
990     00 02 y e a r 00 ??
991 nigel 75 </pre>
992     When writing code to extract data from named subpatterns using the
993 nigel 91 name-to-number map, remember that the length of the entries is likely to be
994 nigel 75 different for each compiled pattern.
995 nigel 63 <pre>
996 ph10 172 PCRE_INFO_OKPARTIAL
997     </pre>
998     Return 1 if the pattern can be used for partial matching, otherwise 0. The
999     fourth argument should point to an <b>int</b> variable. The
1000     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1001     documentation lists the restrictions that apply to patterns when partial
1002     matching is used.
1003     <pre>
1004 nigel 63 PCRE_INFO_OPTIONS
1005 nigel 75 </pre>
1006 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
1007     argument should point to an <b>unsigned long int</b> variable. These option bits
1008     are those specified in the call to <b>pcre_compile()</b>, modified by any
1009 ph10 197 top-level option settings at the start of the pattern itself. In other words,
1010     they are the options that will be in force when matching starts. For example,
1011     if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1012     result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1013 nigel 63 </P>
1014     <P>
1015     A pattern is automatically anchored by PCRE if all of its top-level
1016     alternatives begin with one of the following:
1017     <pre>
1018     ^ unless PCRE_MULTILINE is set
1019     \A always
1020     \G always
1021 nigel 75 .* if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
1022     </pre>
1023 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1024     <b>pcre_fullinfo()</b>.
1025     <pre>
1026     PCRE_INFO_SIZE
1027 nigel 75 </pre>
1028 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
1029     the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
1030     place the compiled data. The fourth argument should point to a <b>size_t</b>
1031     variable.
1032     <pre>
1033     PCRE_INFO_STUDYSIZE
1034 nigel 75 </pre>
1035     Return the size of the data block pointed to by the <i>study_data</i> field in
1036 nigel 63 a <b>pcre_extra</b> block. That is, it is the value that was passed to
1037     <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
1038     created by <b>pcre_study()</b>. The fourth argument should point to a
1039     <b>size_t</b> variable.
1040     </P>
1041 nigel 91 <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
1042 nigel 63 <P>
1043     <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
1044     <b>*<i>firstcharptr</i>);</b>
1045     </P>
1046     <P>
1047     The <b>pcre_info()</b> function is now obsolete because its interface is too
1048     restrictive to return all the available data about a compiled pattern. New
1049     programs should use <b>pcre_fullinfo()</b> instead. The yield of
1050     <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
1051     following negative numbers:
1052     <pre>
1053     PCRE_ERROR_NULL the argument <i>code</i> was NULL
1054     PCRE_ERROR_BADMAGIC the "magic number" was not found
1055 nigel 75 </pre>
1056 nigel 63 If the <i>optptr</i> argument is not NULL, a copy of the options with which the
1057     pattern was compiled is placed in the integer it points to (see
1058     PCRE_INFO_OPTIONS above).
1059     </P>
1060     <P>
1061     If the pattern is not anchored and the <i>firstcharptr</i> argument is not NULL,
1062     it is used to pass back information about the first character of any matched
1063     string (see PCRE_INFO_FIRSTBYTE above).
1064     </P>
1065 nigel 91 <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1066 nigel 63 <P>
1067 nigel 77 <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1068     </P>
1069     <P>
1070     The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1071     data block that contains a compiled pattern. It is provided for the benefit of
1072     applications that operate in an object-oriented manner, where different parts
1073     of the application may be using the same compiled pattern, but you want to free
1074     the block when they are all done.
1075     </P>
1076     <P>
1077     When a pattern is compiled, the reference count field is initialized to zero.
1078     It is changed only by calling this function, whose action is to add the
1079     <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1080     function is the new value. However, the value of the count is constrained to
1081     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1082     it is forced to the appropriate limit value.
1083     </P>
1084     <P>
1085     Except when it is zero, the reference count is not correctly preserved if a
1086     pattern is compiled on one host and then transferred to a host whose byte-order
1087     is different. (This seems a highly unlikely scenario.)
1088     </P>
1089 nigel 91 <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1090 nigel 77 <P>
1091 nigel 63 <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1092     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1093     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
1094     </P>
1095     <P>
1096     The function <b>pcre_exec()</b> is called to match a subject string against a
1097 nigel 75 compiled pattern, which is passed in the <i>code</i> argument. If the
1098 nigel 63 pattern has been studied, the result of the study should be passed in the
1099 nigel 77 <i>extra</i> argument. This function is the main matching facility of the
1100     library, and it operates in a Perl-like manner. For specialist use there is
1101     also an alternative matching function, which is described
1102     <a href="#dfamatch">below</a>
1103     in the section about the <b>pcre_dfa_exec()</b> function.
1104 nigel 63 </P>
1105     <P>
1106 nigel 75 In most applications, the pattern will have been compiled (and optionally
1107     studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1108     possible to save compiled patterns and study data, and then use them later
1109     in different processes, possibly even on different hosts. For a discussion
1110     about this, see the
1111     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1112     documentation.
1113 nigel 63 </P>
1114     <P>
1115 nigel 75 Here is an example of a simple call to <b>pcre_exec()</b>:
1116 nigel 63 <pre>
1117     int rc;
1118     int ovector[30];
1119     rc = pcre_exec(
1120     re, /* result of pcre_compile() */
1121     NULL, /* we didn't study the pattern */
1122     "some string", /* the subject string */
1123     11, /* the length of the subject string */
1124     0, /* start at offset 0 in the subject */
1125     0, /* default options */
1126 nigel 75 ovector, /* vector of integers for substring information */
1127 nigel 77 30); /* number of elements (NOT size in bytes) */
1128 nigel 75 <a name="extradata"></a></PRE>
1129 nigel 63 </P>
1130 nigel 75 <br><b>
1131     Extra data for <b>pcre_exec()</b>
1132     </b><br>
1133 nigel 63 <P>
1134     If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1135     data block. The <b>pcre_study()</b> function returns such a block (when it
1136     doesn't return NULL), but you can also create one for yourself, and pass
1137 nigel 87 additional information in it. The <b>pcre_extra</b> block contains the following
1138     fields (not necessarily in this order):
1139 nigel 63 <pre>
1140     unsigned long int <i>flags</i>;
1141     void *<i>study_data</i>;
1142     unsigned long int <i>match_limit</i>;
1143 nigel 87 unsigned long int <i>match_limit_recursion</i>;
1144 nigel 63 void *<i>callout_data</i>;
1145 nigel 75 const unsigned char *<i>tables</i>;
1146     </pre>
1147 nigel 63 The <i>flags</i> field is a bitmap that specifies which of the other fields
1148     are set. The flag bits are:
1149     <pre>
1150     PCRE_EXTRA_STUDY_DATA
1151     PCRE_EXTRA_MATCH_LIMIT
1152 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1153 nigel 63 PCRE_EXTRA_CALLOUT_DATA
1154 nigel 75 PCRE_EXTRA_TABLES
1155     </pre>
1156 nigel 63 Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1157     <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1158 nigel 75 the appropriate flag bit. You should not set this yourself, but you may add to
1159     the block by setting the other fields and their corresponding flag bits.
1160 nigel 63 </P>
1161     <P>
1162     The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1163     vast amount of resources when running patterns that are not going to match,
1164     but which have a very large number of possibilities in their search trees. The
1165 nigel 75 classic example is the use of nested unlimited repeats.
1166 nigel 63 </P>
1167     <P>
1168 nigel 75 Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1169 nigel 87 (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1170     number of times this function is called during a match, which has the effect of
1171     limiting the amount of backtracking that can take place. For patterns that are
1172     not anchored, the count restarts from zero for each position in the subject
1173     string.
1174 nigel 75 </P>
1175     <P>
1176 nigel 87 The default value for the limit can be set when PCRE is built; the default
1177 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1178 nigel 87 override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1179     block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1180     the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1181     PCRE_ERROR_MATCHLIMIT.
1182 nigel 63 </P>
1183     <P>
1184 nigel 87 The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1185     instead of limiting the total number of times that <b>match()</b> is called, it
1186     limits the depth of recursion. The recursion depth is a smaller number than the
1187     total number of calls, because not all calls to <b>match()</b> are recursive.
1188     This limit is of use only if it is set smaller than <i>match_limit</i>.
1189     </P>
1190     <P>
1191     Limiting the recursion depth limits the amount of stack that can be used, or,
1192     when PCRE has been compiled to use memory on the heap instead of the stack, the
1193     amount of heap memory that can be used.
1194     </P>
1195     <P>
1196     The default value for <i>match_limit_recursion</i> can be set when PCRE is
1197     built; the default default is the same value as the default for
1198     <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1199     with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1200     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1201     is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1202     </P>
1203     <P>
1204 nigel 63 The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1205 nigel 75 which is described in the
1206     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1207     documentation.
1208 nigel 63 </P>
1209     <P>
1210 nigel 75 The <i>tables</i> field is used to pass a character tables pointer to
1211     <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1212     pattern. A non-NULL value is stored with the compiled pattern only if custom
1213     tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1214     If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1215     internal tables to be used. This facility is helpful when re-using patterns
1216     that have been saved after compiling with an external set of tables, because
1217     the external tables might be at a different address when <b>pcre_exec()</b> is
1218     called. See the
1219     <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1220     documentation for a discussion of saving compiled patterns for later use.
1221 ph10 227 <a name="execoptions"></a></P>
1222 nigel 75 <br><b>
1223     Option bits for <b>pcre_exec()</b>
1224     </b><br>
1225 nigel 63 <P>
1226 nigel 75 The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1227 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1228     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1229 nigel 63 <pre>
1230 nigel 75 PCRE_ANCHORED
1231     </pre>
1232     The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1233     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1234     to be anchored by virtue of its contents, it cannot be made unachored at
1235     matching time.
1236     <pre>
1237 ph10 231 PCRE_BSR_ANYCRLF
1238     PCRE_BSR_UNICODE
1239     </pre>
1240     These options (which are mutually exclusive) control what the \R escape
1241     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1242     match any Unicode newline sequence. These options override the choice that was
1243     made or defaulted when the pattern was compiled.
1244     <pre>
1245 nigel 91 PCRE_NEWLINE_CR
1246     PCRE_NEWLINE_LF
1247     PCRE_NEWLINE_CRLF
1248 ph10 150 PCRE_NEWLINE_ANYCRLF
1249 nigel 93 PCRE_NEWLINE_ANY
1250 nigel 91 </pre>
1251     These options override the newline definition that was chosen or defaulted when
1252 nigel 93 the pattern was compiled. For details, see the description of
1253     <b>pcre_compile()</b> above. During matching, the newline choice affects the
1254     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1255     the way the match position is advanced after a match failure for an unanchored
1256 ph10 227 pattern.
1257     </P>
1258     <P>
1259     When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1260     match attempt for an unanchored pattern fails when the current position is at a
1261 ph10 231 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1262 ph10 227 characters, the match position is advanced by two characters instead of one, in
1263     other words, to after the CRLF.
1264     </P>
1265     <P>
1266     The above rule is a compromise that makes the most common cases work as
1267     expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1268     set), it does not match the string "\r\nA" because, after failing at the
1269     start, it skips both the CR and the LF before retrying. However, the pattern
1270     [\r\n]A does match that string, because it contains an explicit CR or LF
1271     reference, and so advances only by one character after the first failure.
1272     </P>
1273     <P>
1274 ph10 231 An explicit match for CR of LF is either a literal appearance of one of those
1275     characters, or one of the \r or \n escape sequences. Implicit matches such as
1276     [^X] do not count, nor does \s (which includes CR and LF in the characters
1277     that it matches).
1278     </P>
1279     <P>
1280 ph10 227 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1281     valid newline sequence and explicit \r or \n escapes appear in the pattern.
1282 nigel 91 <pre>
1283 nigel 63 PCRE_NOTBOL
1284 nigel 75 </pre>
1285     This option specifies that first character of the subject string is not the
1286     beginning of a line, so the circumflex metacharacter should not match before
1287     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1288     never to match. This option affects only the behaviour of the circumflex
1289     metacharacter. It does not affect \A.
1290 nigel 63 <pre>
1291     PCRE_NOTEOL
1292 nigel 75 </pre>
1293     This option specifies that the end of the subject string is not the end of a
1294     line, so the dollar metacharacter should not match it nor (except in multiline
1295     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1296     compile time) causes dollar never to match. This option affects only the
1297     behaviour of the dollar metacharacter. It does not affect \Z or \z.
1298 nigel 63 <pre>
1299     PCRE_NOTEMPTY
1300 nigel 75 </pre>
1301 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1302     there are alternatives in the pattern, they are tried. If all the alternatives
1303     match the empty string, the entire match fails. For example, if the pattern
1304     <pre>
1305     a?b?
1306 nigel 75 </pre>
1307 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1308     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1309     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1310     </P>
1311     <P>
1312     Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1313     of a pattern match of the empty string within its <b>split()</b> function, and
1314     when using the /g modifier. It is possible to emulate Perl's behaviour after
1315     matching a null string by first trying the match again at the same offset with
1316 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1317     starting offset (see below) and trying an ordinary match again. There is some
1318     code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1319     <pre>
1320     PCRE_NO_UTF8_CHECK
1321     </pre>
1322     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1323     string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1324     The value of <i>startoffset</i> is also checked to ensure that it points to the
1325 ph10 211 start of a UTF-8 character. There is a discussion about the validity of UTF-8
1326     strings in the
1327     <a href="pcre.html#utf8strings">section on UTF-8 support</a>
1328     in the main
1329     <a href="pcre.html"><b>pcre</b></a>
1330     page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
1331     the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,
1332     PCRE_ERROR_BADUTF8_OFFSET is returned.
1333 nigel 63 </P>
1334     <P>
1335 nigel 75 If you already know that your subject is valid, and you want to skip these
1336     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1337     calling <b>pcre_exec()</b>. You might want to do this for the second and
1338     subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1339     all the matches in a single subject string. However, you should be sure that
1340     the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1341     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1342     subject, or a value of <i>startoffset</i> that does not point to the start of a
1343     UTF-8 character, is undefined. Your program may crash.
1344     <pre>
1345     PCRE_PARTIAL
1346     </pre>
1347     This option turns on the partial matching feature. If the subject string fails
1348     to match the pattern, but at some point during the matching process the end of
1349     the subject was reached (that is, the subject partially matches the pattern and
1350     the failure to match occurred only because there were not enough subject
1351     characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1352     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1353     may appear in the pattern. These are discussed in the
1354     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1355     documentation.
1356     </P>
1357     <br><b>
1358     The string to be matched by <b>pcre_exec()</b>
1359     </b><br>
1360     <P>
1361 nigel 63 The subject string is passed to <b>pcre_exec()</b> as a pointer in
1362 nigel 73 <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1363 nigel 75 <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1364     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1365     bytes. When the starting offset is zero, the search for a match starts at the
1366     beginning of the subject, and this is by far the most common case.
1367 nigel 63 </P>
1368     <P>
1369     A non-zero starting offset is useful when searching for another match in the
1370     same subject by calling <b>pcre_exec()</b> again after a previous success.
1371     Setting <i>startoffset</i> differs from just passing over a shortened string and
1372     setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1373     lookbehind. For example, consider the pattern
1374     <pre>
1375     \Biss\B
1376 nigel 75 </pre>
1377 nigel 63 which finds occurrences of "iss" in the middle of words. (\B matches only if
1378     the current position in the subject is not a word boundary.) When applied to
1379     the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
1380     occurrence. If <b>pcre_exec()</b> is called again with just the remainder of the
1381     subject, namely "issipi", it does not match, because \B is always false at the
1382     start of the subject, which is deemed to be a word boundary. However, if
1383     <b>pcre_exec()</b> is passed the entire string again, but with <i>startoffset</i>
1384     set to 4, it finds the second occurrence of "iss" because it is able to look
1385     behind the starting point to discover that it is preceded by a letter.
1386     </P>
1387     <P>
1388     If a non-zero starting offset is passed when the pattern is anchored, one
1389 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1390 nigel 63 pattern does not require the match to be at the start of the subject.
1391     </P>
1392 nigel 75 <br><b>
1393     How <b>pcre_exec()</b> returns captured substrings
1394     </b><br>
1395 nigel 63 <P>
1396     In general, a pattern matches a certain portion of the subject, and in
1397     addition, further substrings from the subject may be picked out by parts of the
1398     pattern. Following the usage in Jeffrey Friedl's book, this is called
1399     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1400     a fragment of a pattern that picks out a substring. PCRE supports several other
1401     kinds of parenthesized subpattern that do not cause substrings to be captured.
1402     </P>
1403     <P>
1404     Captured substrings are returned to the caller via a vector of integer offsets
1405     whose address is passed in <i>ovector</i>. The number of elements in the vector
1406 nigel 75 is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1407     this argument is NOT the size of <i>ovector</i> in bytes.
1408 nigel 63 </P>
1409     <P>
1410 nigel 75 The first two-thirds of the vector is used to pass back captured substrings,
1411     each substring using a pair of integers. The remaining third of the vector is
1412     used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1413     and is not available for passing back information. The length passed in
1414     <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1415     rounded down.
1416     </P>
1417     <P>
1418     When a match is successful, information about captured substrings is returned
1419     in pairs of integers, starting at the beginning of <i>ovector</i>, and
1420 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1421     pair is set to the offset of the first character in a substring, and the second
1422     is set to the offset of the first character after the end of a substring. The
1423     first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1424     subject string matched by the entire pattern. The next pair is used for the
1425     first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1426 nigel 91 is one more than the highest numbered pair that has been set. For example, if
1427     two substrings have been captured, the returned value is 3. If there are no
1428     capturing subpatterns, the return value from a successful match is 1,
1429     indicating that just the first pair of offsets has been set.
1430 nigel 63 </P>
1431     <P>
1432     If a capturing subpattern is matched repeatedly, it is the last portion of the
1433 nigel 75 string that it matched that is returned.
1434 nigel 63 </P>
1435     <P>
1436 nigel 75 If the vector is too small to hold all the captured substring offsets, it is
1437     used as far as possible (up to two-thirds of its length), and the function
1438     returns a value of zero. In particular, if the substring offsets are not of
1439     interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1440 nigel 63 <i>ovecsize</i> as zero. However, if the pattern contains back references and
1441 nigel 75 the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1442     has to get additional memory for use during matching. Thus it is usually
1443     advisable to supply an <i>ovector</i>.
1444 nigel 63 </P>
1445     <P>
1446 nigel 91 The <b>pcre_info()</b> function can be used to find out how many capturing
1447 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1448     <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1449     the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1450 nigel 91 </P>
1451     <P>
1452     It is possible for capturing subpattern number <i>n+1</i> to match some part of
1453     the subject when subpattern <i>n</i> has not been used at all. For example, if
1454     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1455     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1456     happens, both values in the offset pairs corresponding to unused subpatterns
1457     are set to -1.
1458     </P>
1459     <P>
1460     Offset values that correspond to unused subpatterns at the end of the
1461     expression are also set to -1. For example, if the string "abc" is matched
1462     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1463     return from the function is 2, because the highest used capturing subpattern
1464     number is 1. However, you can refer to the offsets for the second and third
1465     capturing subpatterns if you wish (assuming the vector is large enough, of
1466     course).
1467     </P>
1468     <P>
1469     Some convenience functions are provided for extracting the captured substrings
1470     as separate strings. These are described below.
1471 nigel 77 <a name="errorlist"></a></P>
1472 nigel 75 <br><b>
1473 nigel 91 Error return values from <b>pcre_exec()</b>
1474 nigel 75 </b><br>
1475 nigel 63 <P>
1476     If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1477     defined in the header file:
1478     <pre>
1479     PCRE_ERROR_NOMATCH (-1)
1480 nigel 75 </pre>
1481 nigel 63 The subject string did not match the pattern.
1482     <pre>
1483     PCRE_ERROR_NULL (-2)
1484 nigel 75 </pre>
1485 nigel 63 Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1486     NULL and <i>ovecsize</i> was not zero.
1487     <pre>
1488     PCRE_ERROR_BADOPTION (-3)
1489 nigel 75 </pre>
1490 nigel 63 An unrecognized bit was set in the <i>options</i> argument.
1491     <pre>
1492     PCRE_ERROR_BADMAGIC (-4)
1493 nigel 75 </pre>
1494 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1495 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1496     compiled in an environment of one endianness is run in an environment with the
1497     other endianness. This is the error that PCRE gives when the magic number is
1498     not present.
1499 nigel 63 <pre>
1500 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1501 nigel 75 </pre>
1502 nigel 63 While running the pattern match, an unknown item was encountered in the
1503     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1504     of the compiled pattern.
1505     <pre>
1506     PCRE_ERROR_NOMEMORY (-6)
1507 nigel 75 </pre>
1508 nigel 63 If a pattern contains back references, but the <i>ovector</i> that is passed to
1509     <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1510     gets a block of memory at the start of matching to use for this purpose. If the
1511 nigel 75 call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1512     automatically freed at the end of matching.
1513 nigel 63 <pre>
1514     PCRE_ERROR_NOSUBSTRING (-7)
1515 nigel 75 </pre>
1516 nigel 63 This error is used by the <b>pcre_copy_substring()</b>,
1517     <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1518     below). It is never returned by <b>pcre_exec()</b>.
1519     <pre>
1520     PCRE_ERROR_MATCHLIMIT (-8)
1521 nigel 75 </pre>
1522 nigel 87 The backtracking limit, as specified by the <i>match_limit</i> field in a
1523     <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1524     above.
1525     <pre>
1526 nigel 63 PCRE_ERROR_CALLOUT (-9)
1527 nigel 75 </pre>
1528 nigel 63 This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1529     use by callout functions that want to yield a distinctive error code. See the
1530 nigel 75 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1531     documentation for details.
1532 nigel 71 <pre>
1533 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1534 nigel 75 </pre>
1535 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1536 nigel 73 <pre>
1537     PCRE_ERROR_BADUTF8_OFFSET (-11)
1538 nigel 75 </pre>
1539 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1540     of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1541 nigel 75 <pre>
1542 nigel 77 PCRE_ERROR_PARTIAL (-12)
1543 nigel 75 </pre>
1544     The subject string did not match, but it did match partially. See the
1545     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1546     documentation for details of partial matching.
1547     <pre>
1548 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1549 nigel 75 </pre>
1550     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1551     are not supported for partial matching. See the
1552     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1553     documentation for details of partial matching.
1554     <pre>
1555 nigel 77 PCRE_ERROR_INTERNAL (-14)
1556 nigel 75 </pre>
1557     An unexpected internal error has occurred. This error could be caused by a bug
1558     in PCRE or by overwriting of the compiled pattern.
1559     <pre>
1560 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1561 nigel 75 </pre>
1562     This error is given if the value of the <i>ovecsize</i> argument is negative.
1563 nigel 93 <pre>
1564     PCRE_ERROR_RECURSIONLIMIT (-21)
1565     </pre>
1566     The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1567     field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1568     description above.
1569     <pre>
1570     PCRE_ERROR_BADNEWLINE (-23)
1571     </pre>
1572     An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1573 nigel 73 </P>
1574 nigel 93 <P>
1575 ph10 197 Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
1576 nigel 93 </P>
1577 nigel 91 <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1578 nigel 63 <P>
1579     <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1580     <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
1581     <b>int <i>buffersize</i>);</b>
1582     </P>
1583     <P>
1584     <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1585     <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
1586     <b>const char **<i>stringptr</i>);</b>
1587     </P>
1588     <P>
1589     <b>int pcre_get_substring_list(const char *<i>subject</i>,</b>
1590     <b>int *<i>ovector</i>, int <i>stringcount</i>, const char ***<i>listptr</i>);</b>
1591     </P>
1592     <P>
1593     Captured substrings can be accessed directly by using the offsets returned by
1594     <b>pcre_exec()</b> in <i>ovector</i>. For convenience, the functions
1595     <b>pcre_copy_substring()</b>, <b>pcre_get_substring()</b>, and
1596     <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1597     as new, separate, zero-terminated strings. These functions identify substrings
1598     by number. The next section describes functions for extracting named
1599 nigel 91 substrings.
1600 nigel 63 </P>
1601     <P>
1602 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1603     further zero added on the end, but the result is not, of course, a C string.
1604     However, you can process such a string by referring to the length that is
1605     returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1606     Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1607     for handling strings containing binary zeros, because the end of the final
1608     string is not independently indicated.
1609     </P>
1610     <P>
1611 nigel 63 The first three arguments are the same for all three of these functions:
1612 nigel 75 <i>subject</i> is the subject string that has just been successfully matched,
1613 nigel 63 <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1614     <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1615     captured by the match, including the substring that matched the entire regular
1616 nigel 75 expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1617     than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1618     space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1619     number of elements in the vector divided by three.
1620 nigel 63 </P>
1621     <P>
1622     The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1623     extract a single substring, whose number is given as <i>stringnumber</i>. A
1624 nigel 75 value of zero extracts the substring that matched the entire pattern, whereas
1625 nigel 63 higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1626     the string is placed in <i>buffer</i>, whose length is given by
1627     <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1628     obtained via <b>pcre_malloc</b>, and its address is returned via
1629     <i>stringptr</i>. The yield of the function is the length of the string, not
1630 nigel 93 including the terminating zero, or one of these error codes:
1631 nigel 63 <pre>
1632     PCRE_ERROR_NOMEMORY (-6)
1633 nigel 75 </pre>
1634 nigel 63 The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1635     memory failed for <b>pcre_get_substring()</b>.
1636     <pre>
1637     PCRE_ERROR_NOSUBSTRING (-7)
1638 nigel 75 </pre>
1639 nigel 63 There is no substring whose number is <i>stringnumber</i>.
1640     </P>
1641     <P>
1642     The <b>pcre_get_substring_list()</b> function extracts all available substrings
1643     and builds a list of pointers to them. All this is done in a single block of
1644 nigel 75 memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1645 nigel 63 is returned via <i>listptr</i>, which is also the start of the list of string
1646     pointers. The end of the list is marked by a NULL pointer. The yield of the
1647 nigel 93 function is zero if all went well, or the error code
1648 nigel 63 <pre>
1649     PCRE_ERROR_NOMEMORY (-6)
1650 nigel 75 </pre>
1651 nigel 63 if the attempt to get the memory block failed.
1652     </P>
1653     <P>
1654     When any of these functions encounter a substring that is unset, which can
1655     happen when capturing subpattern number <i>n+1</i> matches some part of the
1656     subject, but subpattern <i>n</i> has not been used at all, they return an empty
1657     string. This can be distinguished from a genuine zero-length substring by
1658     inspecting the appropriate offset in <i>ovector</i>, which is negative for unset
1659     substrings.
1660     </P>
1661     <P>
1662     The two convenience functions <b>pcre_free_substring()</b> and
1663     <b>pcre_free_substring_list()</b> can be used to free the memory returned by
1664     a previous call of <b>pcre_get_substring()</b> or
1665     <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1666     the function pointed to by <b>pcre_free</b>, which of course could be called
1667     directly from a C program. However, PCRE is used in some situations where it is
1668 nigel 91 linked via a special interface to another programming language that cannot use
1669 nigel 63 <b>pcre_free</b> directly; it is for these cases that the functions are
1670     provided.
1671     </P>
1672 nigel 91 <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1673 nigel 63 <P>
1674 nigel 75 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1675     <b>const char *<i>name</i>);</b>
1676     </P>
1677     <P>
1678 nigel 63 <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1679     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1680     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1681     <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1682     </P>
1683     <P>
1684     <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1685     <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1686     <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1687     <b>const char **<i>stringptr</i>);</b>
1688     </P>
1689     <P>
1690 nigel 75 To extract a substring by name, you first have to find associated number.
1691     For example, for this pattern
1692 nigel 63 <pre>
1693 nigel 93 (a+)b(?&#60;xxx&#62;\d+)...
1694 nigel 75 </pre>
1695 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1696     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1697     calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1698     pattern, and the second is the name. The yield of the function is the
1699 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1700     that name.
1701 nigel 63 </P>
1702     <P>
1703 nigel 75 Given the number, you can extract the substring directly, or use one of the
1704     functions described in the previous section. For convenience, there are also
1705     two functions that do the whole job.
1706 nigel 63 </P>
1707     <P>
1708 nigel 91 Most of the arguments of <b>pcre_copy_named_substring()</b> and
1709     <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1710 nigel 75 functions that extract by number. As these are described in the previous
1711     section, they are not re-described here. There are just two differences:
1712 nigel 63 </P>
1713     <P>
1714     First, instead of a substring number, a substring name is given. Second, there
1715     is an extra argument, given at the start, which is a pointer to the compiled
1716     pattern. This is needed in order to gain access to the name-to-number
1717     translation table.
1718     </P>
1719     <P>
1720     These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1721 ph10 128 then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
1722     appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1723     the behaviour may not be what you want (see the next section).
1724 nigel 63 </P>
1725 nigel 91 <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1726 nigel 63 <P>
1727 nigel 91 <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1728     <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1729     </P>
1730     <P>
1731     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1732     are not required to be unique. Normally, patterns with duplicate names are such
1733     that in any one match, only one of the named subpatterns participates. An
1734     example is shown in the
1735     <a href="pcrepattern.html"><b>pcrepattern</b></a>
1736 ph10 208 documentation.
1737     </P>
1738     <P>
1739     When duplicates are present, <b>pcre_copy_named_substring()</b> and
1740     <b>pcre_get_named_substring()</b> return the first substring corresponding to
1741     the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1742     returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
1743     returns one of the numbers that are associated with the name, but it is not
1744     defined which it is.
1745     </P>
1746     <P>
1747 nigel 91 If you want to get full details of all captured substrings for a given name,
1748     you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1749     argument is the compiled pattern, and the second is the name. The third and
1750     fourth are pointers to variables which are updated by the function. After it
1751     has run, they point to the first and last entries in the name-to-number table
1752     for the given name. The function itself returns the length of each entry, or
1753 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1754     described above in the section entitled <i>Information about a pattern</i>.
1755     Given all the relevant entries for the name, you can extract each of their
1756     numbers, and hence the captured data, if any.
1757 nigel 91 </P>
1758     <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1759     <P>
1760 nigel 77 The traditional matching function uses a similar algorithm to Perl, which stops
1761     when it finds the first match, starting at a given point in the subject. If you
1762     want to find all possible matches, or the longest possible match, consider
1763     using the alternative matching function (see below) instead. If you cannot use
1764     the alternative function, but still need to find all possible matches, you
1765     can kludge it up by making use of the callout facility, which is described in
1766     the
1767     <a href="pcrecallout.html"><b>pcrecallout</b></a>
1768     documentation.
1769     </P>
1770     <P>
1771     What you have to do is to insert a callout right at the end of the pattern.
1772     When your callout function is called, extract and save the current matched
1773     substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1774     other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1775     will yield PCRE_ERROR_NOMATCH.
1776     <a name="dfamatch"></a></P>
1777 nigel 91 <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1778 nigel 77 <P>
1779     <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1780     <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1781     <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1782     <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1783     </P>
1784     <P>
1785     The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1786 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1787     just once, and does not backtrack. This has different characteristics to the
1788     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1789     patterns are not supported. Nevertheless, there are times when this kind of
1790     matching can be useful. For a discussion of the two matching algorithms, see
1791     the
1792 nigel 77 <a href="pcrematching.html"><b>pcrematching</b></a>
1793     documentation.
1794     </P>
1795     <P>
1796     The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1797     <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1798     different way, and this is described below. The other common arguments are used
1799     in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1800     here.
1801     </P>
1802     <P>
1803     The two additional arguments provide workspace for the function. The workspace
1804     vector should contain at least 20 elements. It is used for keeping track of
1805     multiple paths through the pattern tree. More workspace will be needed for
1806 nigel 91 patterns and subjects where there are a lot of potential matches.
1807 nigel 77 </P>
1808     <P>
1809 nigel 87 Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1810 nigel 77 <pre>
1811     int rc;
1812     int ovector[10];
1813     int wspace[20];
1814 nigel 87 rc = pcre_dfa_exec(
1815 nigel 77 re, /* result of pcre_compile() */
1816     NULL, /* we didn't study the pattern */
1817     "some string", /* the subject string */
1818     11, /* the length of the subject string */
1819     0, /* start at offset 0 in the subject */
1820     0, /* default options */
1821     ovector, /* vector of integers for substring information */
1822     10, /* number of elements (NOT size in bytes) */
1823     wspace, /* working space vector */
1824     20); /* number of elements (NOT size in bytes) */
1825     </PRE>
1826     </P>
1827     <br><b>
1828     Option bits for <b>pcre_dfa_exec()</b>
1829     </b><br>
1830     <P>
1831     The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1832 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1833     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1834 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1835     the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1836     <pre>
1837     PCRE_PARTIAL
1838     </pre>
1839     This has the same general effect as it does for <b>pcre_exec()</b>, but the
1840     details are slightly different. When PCRE_PARTIAL is set for
1841     <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1842     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1843     complete matches, but there is still at least one matching possibility. The
1844     portion of the string that provided the partial match is set as the first
1845     matching string.
1846     <pre>
1847     PCRE_DFA_SHORTEST
1848     </pre>
1849     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1850 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1851     works, this is necessarily the shortest possible match at the first possible
1852     matching point in the subject string.
1853 nigel 77 <pre>
1854     PCRE_DFA_RESTART
1855     </pre>
1856     When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1857     a partial match, it is possible to call it again, with additional subject
1858     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1859     option requests this action; when it is set, the <i>workspace</i> and
1860     <i>wscount</i> options must reference the same vector as before because data
1861     about the match so far is left in them after a partial match. There is more
1862     discussion of this facility in the
1863     <a href="pcrepartial.html"><b>pcrepartial</b></a>
1864     documentation.
1865     </P>
1866     <br><b>
1867     Successful returns from <b>pcre_dfa_exec()</b>
1868     </b><br>
1869     <P>
1870     When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1871     substring in the subject. Note, however, that all the matches from one run of
1872     the function start at the same point in the subject. The shorter matches are
1873     all initial substrings of the longer matches. For example, if the pattern
1874     <pre>
1875     &#60;.*&#62;
1876     </pre>
1877     is matched against the string
1878     <pre>
1879     This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1880     </pre>
1881     the three matched strings are
1882     <pre>
1883     &#60;something&#62;
1884     &#60;something&#62; &#60;something else&#62;
1885     &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1886     </pre>
1887     On success, the yield of the function is a number greater than zero, which is
1888     the number of matched substrings. The substrings themselves are returned in
1889     <i>ovector</i>. Each string uses two elements; the first is the offset to the
1890 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1891     the same start offset. (Space could have been saved by giving this only once,
1892     but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1893     returns data, even though the meaning of the strings is different.)
1894 nigel 77 </P>
1895     <P>
1896     The strings are returned in reverse order of length; that is, the longest
1897     matching string is given first. If there were too many matches to fit into
1898     <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1899     the longest matches.
1900     </P>
1901     <br><b>
1902     Error returns from <b>pcre_dfa_exec()</b>
1903     </b><br>
1904     <P>
1905     The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1906     Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1907     described
1908     <a href="#errorlist">above.</a>
1909     There are in addition the following errors that are specific to
1910     <b>pcre_dfa_exec()</b>:
1911     <pre>
1912     PCRE_ERROR_DFA_UITEM (-16)
1913     </pre>
1914     This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1915     that it does not support, for instance, the use of \C or a back reference.
1916     <pre>
1917     PCRE_ERROR_DFA_UCOND (-17)
1918     </pre>
1919 nigel 93 This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1920     uses a back reference for the condition, or a test for recursion in a specific
1921     group. These are not supported.
1922 nigel 77 <pre>
1923     PCRE_ERROR_DFA_UMLIMIT (-18)
1924     </pre>
1925     This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1926     block that contains a setting of the <i>match_limit</i> field. This is not
1927     supported (it is meaningless).
1928     <pre>
1929     PCRE_ERROR_DFA_WSSIZE (-19)
1930     </pre>
1931     This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1932     <i>workspace</i> vector.
1933     <pre>
1934     PCRE_ERROR_DFA_RECURSE (-20)
1935     </pre>
1936     When a recursive subpattern is processed, the matching function calls itself
1937     recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1938     error is given if the output vector is not large enough. This should be
1939     extremely rare, as a vector of size 1000 is used.
1940     </P>
1941 nigel 93 <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1942 nigel 77 <P>
1943 nigel 93 <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1944     <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1945     <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1946     </P>
1947 ph10 99 <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1948 nigel 93 <P>
1949 ph10 99 Philip Hazel
1950 nigel 63 <br>
1951 ph10 99 University Computing Service
1952     <br>
1953     Cambridge CB2 3QH, England.
1954     <br>
1955     </P>
1956     <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1957     <P>
1958 ph10 292 Last updated: 26 December 2007
1959 ph10 99 <br>
1960     Copyright &copy; 1997-2007 University of Cambridge.
1961     <br>
1962 nigel 75 <p>
1963     Return to the <a href="index.html">PCRE index page</a>.
1964     </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